Abstract
Aims: Chronic thromboembolic pulmonary hypertension (CTEPH) is a condition with a poor prognosis in which the pulmonary arteries are occluded by organized thrombi. Pulmonary thromboendarterectomy (PEA) is an effective treatment for CTEPH; however, the literature on its histopathological examination is lacking. This study aimed to investigate the histopathological findings and protein and gene expression in PEA specimens, establish an optimal histopathological evaluation method, and clarify the mechanisms of thrombus organization and disease progression in CTEPH.
Methods: In total, 50 patients with CTEPH who underwent PEA were analyzed. The patients were categorized according to their clinical data into two groups: good and poor postoperative courses. The relationship between their histopathological findings and the clinical course was examined. Immunohistochemical studies confirmed the expression of oxidants, antioxidants, and smooth muscle cell (SMC) differentiation markers and their changes during the progression of thrombus organization. The mRNA expression analysis of 102 samples from 27 cases included oxidants, antioxidants, and vasoconstrictor endothelin-1.
Results: In the PEA specimens, colander-like lesions (aggregations of recanalized blood vessels with well-differentiated SMCs) were significantly more common in the good postoperative course group than in the poor postoperative course group; analysis of proteins and genes proposed that oxidative and antioxidant mechanisms were involved. In the colander-like lesions, there was an increase in endothelin-1 mRNA and protein expression of endothelin receptor A.
Conclusions: Colander-like lesions in PEA specimens must be identified. Additionally, SMC differentiation in recanalized vessels and the expression of vasoconstrictors and their receptors may contribute to the progression of CTEPH.
Abbreviations: chronic thromboembolic pulmonary hypertension (CTEPH), pulmonary hypertension (PH), pulmonary thromboendarterectomy (PEA), balloon pulmonary angioplasty (BPA), pulmonary vascular resistance (PVR), formalin-fixed and paraffin-embedded (FFPE), hematoxylin and eosin stain (HE), Elastica van Gieson stain (EVG), smooth muscle cell (SMC), anti-vitamin D3 upregulated protein 1/thioredoxin-interacting protein (VDUP/TXNIP), thioredoxin (TRX), endothelin receptor A (ET-a), immunofluorescence (IF).
Introduction
Chronic thromboembolic pulmonary hypertension (CTEPH) is one of the main diseases that lead to pulmonary hypertension (PH) and right-sided heart failure. It is a pathological condition that has a poor prognosis and causes chronic pulmonary artery occlusion and stenosis as a result of the formation of an organized thrombus in the pulmonary artery. Current treatment options for CTEPH comprise conservative and surgical or catheter treatments. Conservative treatments include oxygen therapy, anticoagulant therapy for preventing blood clot recurrence, and pulmonary vasodilators for PH.
By contrast, pulmonary thromboendarterectomy (PEA) and balloon pulmonary angioplasty (BPA) are considered effective surgical and catheter treatments1-4). Several studies deal with PEA effects and surgical methodology; nevertheless, only a few regard the pathological examination of PEA specimens. Additionally, there is currently no established method for evaluating PEA lesions5, 6). One previous study on the histological findings and clinical course of PEA has reported that colander-like lesions, which are a characteristic histological finding of CTEPH, are associated with decreased postoperative pulmonary vascular resistance (PVR)6). Reportedly, endothelin receptor A (ET-a) is expressed in the contractile phenotype of smooth muscle cells (SMCs) that surround recanalized channels in a more organized chronic thrombus, which is a region that is consistent with the colander-like lesion7). Endothelin-1, which binds to ET-a, is secreted by the endothelial cells and promotes vasoconstriction and SMC proliferation; moreover, reportedly, blood levels of endothelin-1 as well as idiopathic PH are increased in CTEPH8).
These phenomena propose that differentiated SMCs and endothelial cells in recanalized vessels are involved in the progression and worsening of CTEPH.
Nevertheless, previous studies have shown that oxidative and antioxidant mechanisms are involved in atherosclerosis development, atherogenesis, and thrombus formation9). Among them, the antioxidant marker thioredoxin (TRX) is involved in pulmonary artery SMC proliferation via hypoxia-inducible factor (HIF) activation under hypoxic conditions10) and in angiogenesis and vascular SMC proliferation via the nuclear factor kappa B (NFκB) pathway11).
We have previously reported that TRX and its inhibitor, the oxidative marker anti-vitamin D3 upregulated protein 1/TRX-interacting protein (VDUP/TXNIP), are involved in the mechanism of atherosclerosis development under shear stress12).
Thus, we investigated the relationship between the oxidative and antioxidative mechanisms (TRX and TXNIP) and the progression of thrombus organogenesis in CTEPH.
Aim
In this study, we attempted to classify histopathological PEA specimens from patients with CTEPH with objective indices and examined their relevance based on clinical findings and postoperative data.
We examined PEA specimens for histopathology, primarily recanalization vessel changes and SMC differentiation using SMC differentiation markers, as well as for the involvement of oxidative and antioxidant mechanisms.
In addition, we compared and examined the histopathological findings and protein and gene expression and discussed the mechanism of lesion exacerbation in CTEPH.
Methods
Patient Selection and Ethical Approval
A total of 50 patients with CTEPH who underwent PEA at Tokyo Medical University Hospital between February 2012 and March 2017 were analyzed. All patients who underwent PEA were classified using preoperative image analysis (such as contrast-enhanced computed tomography or pulmonary angiography) as either central-type or central-plus-peripheral-type pulmonary artery thrombosis and San Diego classification13) type I/II with or without type III/IV. For retrospective studies using the previously collected specimens, we provided the patients with the option of opting out by posting the relevant documents for refusal of study inclusion. For prospective studies using newly collected specimens, all patients gave written informed consent to the attending physician. The Medical Ethics Committee of Tokyo Medical University approved the study design, and the study complied with the Declaration of Helsinki and its subsequent amendments.
Clinical Data Analysis
In all cases, hemodynamic data were confirmed via right heart catheterization before and after PEA. The patients were then classified into the following two groups based on the postoperative data: the good clinical course group without residual PH and the poor clinical course group with residual PH, which included cases of death. Postoperative residual PH was defined, based on previous literature, as PVR of ≥ 500 dynes/s/cm−5 4). Moreover, we examined the differences in the background and clinical indicators between the two groups.
Histopathological Evaluation
Formalin-fixed and paraffin-embedded (FFPE) samples of the PEA specimens were prepared as in Fig.1, and histopathological findings were examined using the hematoxylin and eosin (HE) and Elastica van Gieson (EVG) stains. We analyzed four histological findings, namely, colander-like lesions (aggregation of the recanalized vascular lesions), early stage of organization, fresh thrombus, and atherosclerosis (Fig.2: a–f).
There were several samples where the four types of histological findings were mixed, and their boundaries were unclear. Thus, we considered it difficult to evaluate the area of lesions and used a method of counting the number of lesions, which is generally easy to introduce.
The method of lesion counting was set as follows: i) count recognizable lesions such as clusters of multiple recanalized vessels or lesions larger than 1 mm; ii) when multiple lesions coexisted at sites separated by at least 5 mm in one section, all were counted as the number of lesions (for example, colander-like lesion 2, the early phase of organization 1, and fresh thrombus 1); iii) regarding the early stages of organization, lesions that spread discontinuously in one section were treated as a single lesion, and when recanalized vessels were included, they were limited to lesions without surrounding SMCs; and iv) for atherosclerosis, CD68 macrophage aggregation was also referred to besides histological images such as lipid deposition.
Normalization was performed by dividing the total number of lesions in the specimen by the total slice number, and a numerical value was calculated for each specimen (Fig.1). Observable lesions in all FFPE samples were counted and normalized by dividing by the total number of slices. The reason for this normalization is that PEA specimens show continuous lesions from the central side to the bifurcation. Previous reports have calculated the number of colander-like lesions4); however, calculating the exact number of contiguous lesions is practically challenging. Histopathological images are frequently observed by performing many serial sections from one location, and when diagnosing a PEA specimen, the number of sections varies depending on the facility and operator who cuts it out. Thus, to correct this, we adjusted for the total lesion number/section number.
In this study, the average number of slices prepared in each case was 13.3 and 9.5 for the right and left, respectively.
We studied whether the calculated values of each finding varied between the postoperatively good and poor groups.
Immunohistochemistry
An immunohistochemical study was carried out on FFPE tissue. The primary antibodies, clones, and dilutions used in this study were as follows: anti-α smooth muscle actin (αSMA) (1A4, 1/100, Agilent, Santa Clara, CA, USA), anti-h-caldesmon (h-CD, 1/50, Agilent), anti-CD68 (PG-M1, 1/100, Agilent), anti-smoothelin (R4A, 1/2, kindly provided by Dr. van Eys GJ., University of Maastricht, The Netherlands), anti-S100-A4 (A5114, 1/4000, Agilent), anti-VDUP/TXNIP (JY2, 1/200, MBL Medical & Biological Laboratories, Tokyo, Japan), anti-TRX (ab26320, 1/2000, Abcam, Cambridge, UK), anti-ET-a (ab76259, 1/200, Abcam), and anti-endothelin-1 (12191-1-AP, 1/400, Proteintech, Rosemont, IL, USA).
For all antibodies, negative and positive controls were routinely performed using the heat- or protease-induced epitope retrieval process. Table 1 describes the epitope retrieval process (temperature, time, and type of solution) and the immunohistochemistry system (kit or autostainer).
We compared the immunostaining findings with the histological findings and examined the protein-expressing cells and their distribution.
Table 1. Data of immunohistochemistry
| Primary antibody |
Epitope retrieval process (temperature, time, and type of solution), secondary antibodies, detection kit, or autostainer |
| Anti-αSMA |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) (Nichirei, Tokyo, Japan) |
|
Hisogine simple stain MAX-PO (MULTI) (Nichirei) |
| Anti-h-caldesmon |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) |
|
Hisogine simple stain MAX-PO (MULTI) |
| Anti-CD68 |
Treated with Pepsin, 20 minutes at room temperature |
|
Hisogine simple stain MAX-PO (MULTI) |
| Anti-smoothelin |
Autostainer: Ventana Discovery XT (Roche diagnostics, Basel, Switzerland) |
|
Heat-induced epitope retrieval in EDTA buffer 120 minutes |
| Anti-S100A4 |
Autostainer: Ventana Discovery XT |
|
Heat-induced epitope retrieval in EDTA buffer 30 minutes |
| Anti-VDUP/TXNIP |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) |
|
Hisogine simple stain MAX-PO (MULTI) |
| Anti-TRX |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) |
|
Hisogine simple stain MAX-PO (MULTI) |
| Anti-ET-a |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) |
|
Hisogine simple stain MAX-PO (MULTI) |
| Anti-endothelin 1 |
Autoclave at 105 ℃ for 20 minutes in Histofine antigen retrieval solution (pH 9) |
|
Hisogine simple stain MAX-PO (MULTI) |
α smooth muscle actin: αSMA; anti-vitamin D3 upregulated protein 1/thioredoxin-interacting protein: VDUP/TXNIP; thioredoxin: TRX; endothelin receptor A: ET-a
Double immunofluorescent (IF) staining was performed to confirm the localization of the antioxidant marker TRX in the early phase of the organization and in the colander-like lesions. A combination of anti-CD31/TRX, CD68/TRX, αSMA/TRX, and smoothelin/TRX were double stained to observe the localization of TRX in the endothelial cells, macrophages, and undifferentiated or well-differentiated SMCs.
FFPE samples were used for double IF staining. After paraffin removal and rehydration, sections were submitted for antigen retrieval with EnVision FLEX Target Retrieval Solution High pH (Agilent) for 30 min at 98℃. The sections were incubated with primary antibodies, which are rabbit polyclonal and mouse monoclonal antibodies sequentially, in antibody diluent (Agilent) for 1 h at 20–25℃. The primary antibodies, clones, and dilutions used for double IF staining were as follows: anti-TRX rabbit polyclonal (ab26320, 1/1000, Abcam), anti-smoothelin (R4A, 1/2, provided as described above), anti-αSMA (1A4, 1/400, Agilent), anti-CD31 (1A10, 1/100, Leica Biosystems, Nussloch, Germany), and CD68 (KP-1, 1/1200, Agilent). Subsequently, the sections were incubated with Alexa Flour 488 and 568 conjugated secondary antibodies (Alexa Fluor TM 568 goat antimouse IgG, Alexa Fluor TM 488 goat antirabbit IgG, Thermo Fisher Scientific, Carlsbad, CA, USA) for 1 h at 20–25℃. ProLong™ Diamond Antifade Mountant with 4′,6-diamidino-2-phenylindole (DAPI) (Thermo Fisher Scientific) was used for nuclear stain and encapsulation. Images were captured using an All-in-one Fluorescence Microscope BZ-X710 (Keyence, Osaka, Japan).
Gene Expression Analysis
Samples for RNA analysis were collected from multiple locations of each of the 27 specimens, and 102 samples could be examined. We harvested and cryopreserved tissue pieces from submitted fresh PEA specimens and used them as samples for RNA analysis. When collecting samples for RNA, FFPE was prepared at the same site (Fig.1), histopathological findings were confirmed, and each sample was classified into three groups: atherosclerosis/intima, fresh thrombi/early phase of organization, and colander-like lesion. Colander-like lesions were classified when recanalized vessels with SMCs were dominant, even with a small amount of fresh thrombi.
Cryopreserved samples were homogenized in TRIzol™ Reagent (Thermo Fisher Scientific), and total RNA was extracted according to the manufacturer’s protocol. Using RiverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO, Osaka, Japan), the extracted total RNA was subjected to DNase treatment and cDNA synthesis.
The mRNA expression of the oxidant/antioxidant markers TXNIP, TRX, and the vasoconstrictor endothelin-1 was analyzed using real-time reverse transcription polymerase chain reaction. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used for endogenous control and was evaluated using the ΔCT method.
Gene expression assays with TaqMan probes (thioredoxins: Hs0155214_g1, TXNIP: Hs01006900_g1, GAPDH: Hs02758991_g1, and endothelin-1: Hs00174961_m) was conducted using the QuantStudio3 Detection System (Applied Biosystems, Waltham, MA, USA) according to the manufacturer’s protocols, and each experiment was performed in duplicates.
Statistical Analysis
The results are expressed as mean±standard deviation, and the Mann–Whitney U test, Student’s t-test, and Fisher’s exact test were performed using SPSS statistical analysis software (version 28.0 for Windows; IBM, Armonk, NY, USA). Statistical significance was considered at P＜0.05.
Results
Patient Characteristics
Table 2 presents the clinical background and laboratory data before and after PEA. Among the 50 patients, eight had a poor postoperative course. Seven cases of residual PH (PVR ≥ 500 dynes/s/cm−5) after PEA and one postoperative death (postoperative PVR was not measured) were found. All patients in the poor clinical course group were women and were significantly older than those in the good prognosis group (P=0.047). Furthermore, the preoperative PVR was significantly higher, and preoperative cardiac output (CO) was significantly lower in the poor group than that in the good prognosis group (P=0.039, each).
Table 2. Clinical and laboratory data (pre- or post-PEA)
|
Total |
Good clinical course
(PVR＜500)
|
Poor clinical course
(PVR ≥ 500 or death)
|
P-value
|
|
|
| Case number |
50 |
42 |
8 (1: death after PEA) |
|
|
|
| Age (years)＊
|
60.4±13.6 |
58.3±13.3 |
69.1±12.0 |
0.047 |
|
|
| Male: Female (number)＊＊
|
17:33 |
17:25 |
0:8 |
0.039 |
|
|
| Coagulation disorder＊＊
|
22.0% |
19.5% |
37.5% |
0.351 |
|
|
| Past history of DVT＊＊
|
70.0% |
68.3% |
75.0% |
1.000 |
|
|
| With peripheral lesions (imaging)＊＊
|
22.0% |
16.7% |
50.0% |
0.059 |
|
|
| Pre mean PAP (mmHg)＊
|
41.7±9.8 |
41.5±10.3 |
42.9±7.4 |
0.507 |
|
|
| Pre CO (L/min)＊
|
4.02±1.35 |
4.20±1.32 |
3.07±1.19 |
0.039 |
|
|
| Pre PVR (dyne・sec・cm－5)＊
|
781.3±412.4 |
725.1±383.8 |
1076.1±457.4 |
0.039 |
|
|
| Pre BNP (pg/ml)＊
|
184.7±201.0 |
178.2±213.2 |
216.0±134.4 |
0.238 |
|
|
| Pre 6MWD (m)＊
|
326.4±109.2 |
330.0±110.0 |
300.0±110.7 |
0.473 |
|
|
| Post mean PAP (mmHg)＊＊＊
|
21.2±8.3 |
19.3±7.2 |
32.6±5.0 |
6.980E-16 |
4.8546E-15 |
0.008 |
| Post CO (L/min)＊＊＊
|
4.13±1.34 |
4.32±1.36 |
3.00±0.29 |
0.726 |
0.624 |
0.712 |
| Post PVR (dyne・sec・cm－5)＊＊＊
|
280.9±181.8 |
219.4±102.2 |
649.9±87.9 |
1.1767E-11 |
6.8298E-11 |
0.066 |
| Post BNP (pg/ml)＊＊＊
|
131.6±156.9 |
91.0±73.3 |
334.2±279.8 |
0.240 |
0.104 |
0.609 |
| Post 6MWD (m)＊＊＊
|
340.8±108.2 |
355.3±106.6 |
245.0±65.4 |
0.106 |
0.040 |
0.046 |
| BPA additional treatment＊＊
|
30.0% |
21.4% |
75.0% |
0.002 |
|
|
Data are presented as mean±SD
＊Mann–Whitney U test, comparison between groups
＊＊Fisher’s exact test, comparison between groups
＊＊＊Student’s t-test, comparison before and after PEA (P-values from left to right indicate pre- and postoperative t-tests for total, good, or poor clinical course, respectively.)
DVT: deep venous thrombosis, PAP: pulmonary arterial pressure, CO: cardiac output, PVR: pulmonary vascular resistance, BNP: brain natriuretic peptide, 6MWD: six-minute walk distance, PEA: pulmonary thromboendarterectomy, SD: standard deviation
The total number of patients with preoperative CO ＜3.0 L/min was 14, of which five were in the poor clinical course group. Regarding the clinical course group, nine of 42 and five of eight patients in the good and poor clinical course groups, respectively, had a preoperative CO of ＜3.0 L/min.
The total number of patients with preoperative PVR of ≥ 1000 dynes/s/cm−5 was 14, of which four were in the poor clinical course group. For clinical course group, the number of patients with a preoperative PVR of ≥ 1000 dynes/s/cm−5 was 10 of 42 and four of eight patients in the good and poor clinical course groups, respectively.
No significant difference was observed between the two groups for other background factors (presence or absence of coagulation abnormality and deep vein thrombosis), brain natriuretic peptide, or other indicators.
The proportion of patients with peripheral lesions on imaging studies tended to be higher in the poor clinical course group than in the good clinical course group, although the difference was not statistically significant.
Based on the clinical course, 15 out of 50 patients were determined to have peripheral lesions, and BPA was performed for additional treatment. The proportion of patients who underwent additional BPA was significantly higher in the poor clinical course group than in the good clinical course group (P=0.002). Although these data are limited to patients undergoing follow-up and may be biased by the additional treatment of peripheral lesions due to poor postoperative course, if PH does not improve with PEA alone, the complication of peripheral lesions and the associated poor course may be the cause.
Relationship Between Histopathological Findings and Postoperative Course
Fig.3 shows a comparison of the histological findings of PEA tissue specimens in the good and poor clinical course groups. In the PEA specimens of patients with CTEPH, colander-like lesions comprising multiple recanalizing blood vessels were observed.
The number of colander-like lesions observed was significantly higher in the PEA tissue specimens of the good clinical course group than in those of the poor clinical course group (Fig.3a). There were no significant differences between the two groups in the early organization, fresh thrombi, and atherosclerosis (Fig.3b–d).
In total, six of eight patients in the poor clinical course group had no colander-like lesions, and one had only one colander-like lesion.
Histopathologically, seven of nine cases with no colander-like lesions were in the poor clinical course group. The remaining two patients without colander-like lesions were classified in the good postoperative course group. However, both had a slightly higher PVR (378 and 444 dynes/s/cm−5, respectively), and one patient had PH recurrence with a PVR of ≥ 700 dynes/s/cm−5 1 year after PEA.
Immunohistochemical Findings
We investigated SMC properties, inflammatory cells, and oxidant/antioxidant markers in organized thrombi. Immunohistochemical study showed that in the early stage of organization (Fig.4a), only undifferentiated SMCs were present, which are defined as α-SMA (+)/h-caldesmon (− or focal +)/S-100A4(− or focal +)/smoothelin (−) (Fig.4b–e). By contrast, recanalizing blood vessels in the colander-like lesions (Fig.5a) were surrounded by well-differentiated SMCs, which are defined as αSMA (+)/h-caldesmon (+)/S100A4 (+)/smoothelin (+) (Fig.5b–e). ET-a, which is the receptor for the vasoconstrictor endothelin, was also confirmed in the well-differentiated SMCs of the colander-like lesions rather than in the early phase of organization (Figs.4f, 5f).
In the early stage of organization, endothelin-1, which is secreted by the vascular endothelium and has vasoconstrictor and cell proliferative effects, was positive in neovascular endothelial cells, undifferentiated SMCs, and macrophages (Fig.4j). In colander-like lesions, endothelin-1 was more prominently positive in recanalized vascular endothelial cells but was also positive in perivascular differentiated SMCs (Fig.5j).
The oxidant/antioxidant markers TXNIP and TRX were expressed in infiltrating macrophages, endothelial cells, and SMCs in the early phase of organization; nevertheless, their expression tended to be reduced in the differentiated SMCs of the colander-like lesions (Fig.4g, h and Fig.5g, h).
Thus, we considered that it is a characteristic of the CTEPH pathological finding that the differentiation of SMCs progresses with the organization of the thrombus. Oxidant and antioxidant mechanisms have been suggested to be involved in this process.
Since normal human pulmonary vessels could not be used as controls in this study, we refer to previous reports on its expression in normal pulmonary arteries. Eneothelin-1 immunoreactivity was observed in the vascular endothelial cells and SMCs throughout the pulmonary arterioles and muscular pulmonary arteries, and its expression was increased in pulmonary artery hypertension (PAH). ET-a has been reported to be expressed mainly in the SMCs of arteries with small size, and the distribution of ET-a in pulmonary arteries was dominant in the proximal in the control group, mainly in the distal segment in PAH14). Nevertheless, TRX and TXNIP have not been reported to be expressed in normal pulmonary arteries, but TRX has been reported in interstitial lung disease. In vessels other than pulmonary arteries, we have previously reported TXNIP and TRX expression in the aortic endothelium and confirmed that TXNIP expression was suppressed via p21Sdi/Cip/Waf1 under laminar shear stress12).
Double Immunofluorescence Staining
Fig.6 shows the changes in the localization of the antioxidant marker TRX in early organizing (Fig.6a–d) and colander-like lesions (Fig.6e–h).
In the early phase of organization, TRX localized in CD31-positive neovascular endothelial cells, CD68-positive macrophages, and αSMA-positive undifferentiated SMCs (Fig.6b–d). Smoothelin-positive well-differentiated SMCs were not observed (Fig.6a).
In colander-like lesions, TRX localization was observed in CD31-positive endothelial cells of recanalized vessels and perivascular αSMA-positive (smoothelin-negative) SMCs, but attenuated or absent TRX was observed in smoothelin-positive SMCs (Fig.6e–g). Macrophages were rarely observed in colander-like lesions (Fig.6h).
Gene Expression Analysis
Based on the histological findings, 102 samples obtained from 27 patients were classified into three groups: the colander-like lesion, the fresh thrombus/early organization, and the intima with atherosclerosis groups. The mRNA expression of TXNIP (oxidation marker), TRX (antioxidant marker), and endothelin-1 (vasoconstrictor) were confirmed and compared between the three groups.
The expression of endothelin-1 was significantly higher in the colander-like lesion group than in the other two groups (Fig.7a). No significant difference was found in the TRX expression in all three groups (Fig.7c); however, TXNIP was significantly increased in the colander-like lesion group compared with the fresh thrombus/early organization phase group and tended to increase compared with atherosclerosis/intima group (Fig.7b).
Discussion
PEA is an effective treatment for CTEPH; however, postoperative PH can sometimes occur. Thus, we investigated which pathological findings in PEA tissue specimens could be predictors of residual postoperative PH.
Of the clinical data factors, age, sex (female), preoperative PVR, and preoperative CO could be predictors of postoperative residual PH. However, CTEPH is predominant in women, and patients in poor preoperative conditions are likely to have a poor postoperative course; hence, clearly predicting the postoperative course using only clinical data is challenging. Patients with poor preoperative conditions (low level of preoperative CO and high level of preoperative PVR) accounted for a high proportion of patients in the poor clinical course group, but the postoperative prediction from preoperative CO and PVR was difficult.
By contrast, our histopathological study has shown that the PEA specimens from patients with a good postoperative course had significantly more colander-like lesions. These lesions, which are characteristic findings in CTEPH, are the responsible lesions that cause the condition, and their removal is important for preventing postoperative PH. We conclude that to predict the postoperative course, the presence or absence of colander-like lesions in PEA specimens should be histopathologically confirmed.
The presence or absence of peripheral lesions also affected the clinical course, but preoperative imaging studies did not always coincide with additional BPA treatment. Hence, histopathological confirmation is important because the postoperative course depends on whether colander-like lesions can be reached and sampled.
Although a previous study has reported that resection of colander-like lesions contributes to reduced postoperative PVR6), our study is the first to compare this finding with other findings, such as fresh thrombi and atherosclerosis.
Next, we investigated the mechanism of thrombus organization and maintenance of SMC differentiation, the involvement of oxidant and antioxidant mechanisms, and the factors that exacerbate the condition.
Colander-like lesions in CTEPH are caused by the recanalization of occlusive thrombi. Various mechanisms have been investigated for thrombus recanalization, including the involvement of circulating progenitor cells forming channels, endothelialization, myofibroblastic invasion, and deposition of matrix metalloproteinases that have been reported15). An occlusive thrombus is a hypoxic state where blood flow and oxygen supply are disrupted, and the environment is different from that of a mural thrombus, including the action of HIF. Elevated HIF1α and HIF2α in animal models of pulmonary thrombosis have been reported16). Such environmental factors may influence SMC proliferation and differentiation in the colander-like lesions.
In this study, the differentiation of SMCs was observed in the organizing thrombus and the recanalized vessels as the thrombus organization progressed. Well-differentiated SMCs around the recanalized blood vessels in the colander-like lesions were h-caldesmon (+), S100A4 (+), smoothelin (+), and ET-a (+), which demonstrated that they have a contractile capacity. Contrarily, TXNIP and TRX, which are oxidant/antioxidant markers, were expressed in macrophages, endothelial cells, and SMCs in the early stage of organization; nevertheless, their expression tended to decrease in well-differentiated SMCs in the colander-like lesions. In endothelial cells, TXNIP and TRX were still expressed in the colander-like lesions. We confirmed the location of TRX with double IF staining as well as immunohistochemical staining.
TRX is involved in pulmonary artery SMC proliferation via HIF activation under hypoxic conditions10) and in angiogenesis and vascular SMC proliferation via the NFκB pathway11). However, TXNIP suppresses TRX9). From the above, we speculated that the TRX, which is secreted from endothelial cells, macrophages, and undifferentiated SMCs in the early stage of organization, is involved in the proliferation of SMCs and increases in TXNIP. The proliferated SMCs were transformed into well-differentiated SMCs positive for smoothelin and ET-a, and with contractile ability by the influence of blood pressure and flow. Biomechanical stimuli, including blood pressure and flow (shear stress), cause changes in the properties and proliferative capacity of SMCs and endothelial cells17), and it is assumed that changes in blood flow and pressure due to recanalization are involved in this mechanism.
Based on the above, we considered that the expression of TRX and TXNIP decreased in the well-differentiated SMCs (Fig.8).
The expression level of TRX and its activity and localization are important. Reportedly, TRX subcellular localization is altered by various factors and that nuclear translocation is observed in the presence of oxidative stress and downregulation of TXNIP18, 19). In this study, we did not search for the localization of TRX and TXNIP in detail; however, it is possible that in the early phase of organization, oxidative stress in thrombotic and hypoxic conditions may cause TRX translocation into the nucleus. In this study’s double IF staining, TRX is positive mainly in the cytoplasm and also appears positive in some nuclei, which suggests that further studies are necessary.
Endothelin-1 is a secreted protein, which is a type of autacoid produced by endothelial cells and has vasoconstrictor and cell proliferative effects20). We believe that the positive immunostaining of endothelin-1 indicates not only in situ production but also the binding and action of the secreted protein. Endothelin-1 was immunohistochemically positive in vascular endothelial cells and SMCs in the early stage of organization and colander-like lesions, which suggests that it is also involved in smooth muscle proliferation.
By contrast, endothelin-1 mRNA expression was increased in the colander-like lesions compared with fresh and early-organized thrombi. This is assumed to be produced by endothelial cells in the recanalized vessels. Therefore, we believe that endothelin-1 secreted from the recanalized vascular endothelial cells binds to the ET-a localized in the well-differentiated SMCs around the recanalized blood vessel, causing smooth muscle contraction and contributing to PH exacerbation.
It has been previously reported that the elevation of endothelin-1 observed in idiopathic pulmonary arterial hypertension is also found in CTEPH and that ET-a is expressed in the organized thrombi of PEA specimens7). Although this study concluded that ET-a expression in the contractile phenotype of SMCs surrounding recanalized channels in more organized chronic thrombus was present, other SMC markers were not shown. Moreover, no studies have reported the involvement of oxidant and antioxidant mechanisms in organized thrombi or the expression of ET-a associated with SMC differentiation markers of recanalized vessels.
The expression of TXNIP mRNA was increased in colander-like lesions compared with that in the early stage of organization and fresh thrombi. Combined with the immunohistochemical findings, we speculated that TXNIP is continuously produced in the endothelial cells of recanalized blood vessels and that it exerts an antioxidant effect on the surrounding SMCs.
Although no significant difference in the TRX mRNA expression was observed among all groups, the activation of TRX is more important than its expression in the oxidant/antioxidant mechanism9). Therefore, we believe that these results do not exclude the involvement of oxidant/antioxidant mechanisms in CTEPH.
Conclusions
In CTEPH, recanalized vessels with well-differentiated SMCs are formed at the thrombus formation site through a process of organization and recanalization with the involvement of oxidant/antioxidant mechanisms. These well-differentiated SMCs have endothelin receptors and are assumed to contract in response to endothelin secreted by endothelial cells, which result in PH exacerbation.
Therefore, it should be confirmed whether this collection of recanalized vascular lesions with well-differentiated SMCs (colander-like lesions) has been resected when observing PEA specimens to predict the possibility of PH development.
Acknowledgements
We are grateful to Dr. Guillaume van Eys, University of Maastricht, The Netherlands, for graciously providing the smoothelin antibody.
Conflict of Interest
None.
Notice of Grant Support
This research was supported by a grant (Scholarship Fund for Young Researchers) from The Science Research Promotion Fund (Tokyo, Japan) and grant-in-aid to Tokyo Medical University, Supporting Positive Activities for Female Researchers from the Japanese Ministry of Education, Culture, Sports, Science and Technology.
References
- 1) Kim NH, Delcroix M, Jais X, Madani MM, Matsubara H, Mayer E, Ogo T, Tapson VF, Ghofrani HA, and Jenkins DP: Chronic thromboembolic pulmonary hypertension. Eur Respir J, 2019; 53: 1801915
- 2) Fukuda K, Date H, Doi S, Fukumoto Y, Fukushima N, Hatano M, Ito H, Kuwana M, Matsubara H, Momomura SI, Nishimura M, Ogino H, Satoh T, Shimokawa H, Yamauchi-Takihara K, Tatsumi K, Ishibashi-Ueda H, Yamada N, Yoshida S, Abe K, Ogawa A, Ogo T, Kasai T, Kataoka M, Kawakami T, Kogaki S, Nakamura M, Nakayama T, Nishizaki M, Sugimura K, Tanabe N, Tsujino I, Yao A, Akasaka T, Ando M, Kimura T, Kuriyama T, Nakanishi N, Nakanishi T, and Tsutsui H; Japanese Circulation Society and the Japanese Pulmonary Circulation and Pulmonary Hypertension Society Joint Working Group: Guidelines for the treatment of pulmonary hypertension (JCS 2017/JPCPHS 2017). Circ J, 2019; 83: 842-945
- 3) Moser KM, Daily PO, Peterson K, Dembitsky W, Vapnek JM, Shure D, Utley J, and Archibald C: Thromboendarterectomy for chronic, major-vessel thromboembolic pulmonary hypertension. Immediate and long-term results in 42 patients. Ann Intern Med, 1987; 107: 560-565
- 4) Feinstein JA, Goldhaber SZ, Lock JE, Ferndandes SM, and Landzberg MJ: Balloon pulmonary angioplasty for treatment of chronic thromboembolic pulmonary hypertension. Circulation, 2001; 103: 10-13
- 5) Bernard J and Yi ES: Pulmonary thromboendarterectomy: a clinicopathologic study of 200 consecutive pulmonary thromboendarterectomy cases in one institution. Hum Pathol, 2007; 38: 871-877
- 6) Hosokawa K, Ishibashi-Ueda H, Kishi T, Nakanishi N, Kyotani S, and Ogino H: Histopathological multiple recanalized lesion is critical element of outcome after pulmonary thromboendarterectomy. Int Heart J, 2011; 52: 377-381
- 7) Southwood M, MacKenzie Ross RV, Kuc RE, Hagan G, Sheares KK, Jenkins DP, Goddard M, Davenport AP, and Pepke-Zaba J: Endothelin ETA receptors predominate in chronic thromboembolic pulmonary hypertension. Life Sci, 2016; 159: 104-110
- 8) Reesink HJ, Meijer RC, Lutter R, Boomsma F, Jansen HM, Kloek JJ, and Bresser P: Hemodynamic and clinical correlates of endothelin-1 in chronic thromboembolic pulmonary. Circ J, 2006; 70: 1058-1063
- 9) Madrigal-Matute J, Martinez-Pinna R, Fernandez-Garcia CE, Ramos-Mozo P, Burillo E, Egido J, Blanco-Colio LM, and Martin-Ventura JL: Cell stress proteins in atherothrombosis. Oxid Med Cell Longev, 2012; 2012: 232464
- 10) Chen B, Nelin VE, Locy ML, Jin Y, and Tipple TE: Thioredoxin-1 mediates hypoxia-induced pulmonary artery smooth muscle cell proliferation. Am J Physiol Lung Cell Mol Physiol, 2013; 305: L389-L395
- 11) Schulze PC, De Keulenaer GW, Yoshioka J, Kassik KA, and Lee RT: Vitamin D3-upregulated protein-1 (VDUP-1) regulates redox-dependent vascular smooth muscle cell proliferation through interaction with thioredoxin. Circ Res, 2002; 91: 689-695
- 12) Obikane H, Abiko Y, Ueno H, Kusumi Y, Esumi M, and Mitsumata M: Effect of endothelial cell proliferation on atherogenesis: a role of p21(Sdi/Cip/Waf1) in monocyte adhesion to endothelial cells. Atherosclerosis, 2010; 212: 116-122
- 13) Thistlethwaite PA, Mo M, Madani MM, Deutsch R, Blanchard D, Kapelanski DP, and Jamieson SW: Operative classification of thromboembolic disease determines outcome after pulmonary endarterectomy. J Thorac Cardiovasc Surg, 2002; 124: 1203-1211
- 14) Huang H, Zhang P, Wang Z, Tang F, and Jiang Z: Activation of endothelin-1 receptor signaling pathways is associated with neointima formation, neoangiogenesis and irreversible pulmonary artery hypertension in patients with congenital heart disease. Circ J, 2011; 75: 1463-1471
- 15) Evans CE, Palazon A, Sim J, Tyrakis PA, Prodger A, Lu X, Chan S, Bendahl PO, Belting M, Von Euler L, Rundqvist H, Johnson RS, and Branco C: Modelling pulmonary microthrombosis coupled to metastasis: distinct effects of thrombogenesis on tumorigenesis. Biol Open, 2017; 6: 688-697
- 16) Lu X, Prodger A, Sim J, and Evans C: Pulmonary thrombosis promotes tumorigenesis via myeloid hypoxia-inducible factors. Biomolecules, 2022; 12: 1354
- 17) Qiu J, Zheng Y, Hu J, Liao D, Gregersen H, Deng X, Fan Y, and Wang G: Biomechanical regulation of vascular smooth muscle cell functions: from in vitro to in vivo understanding. J R Soc Interface, 2013; 11: 20130852
- 18) Wang R, Guo Y, Li L, Luo M, Peng L, Lv D, Cheng Z, Xue Q, Wang L, and Huang J: Role of thioredoxin-interacting protein in mediating endothelial dysfunction in hypertension. Genes Dis, 2020; 9: 753-765
- 19) Ogata FT, Batista WL, Sartori A, Gesteira TF, Masutani H, Arai RJ, Yodoi J, Stern A, and Monteiro HP: Nitrosative/oxidative stress conditions regulate thioredoxin-interacting protein (TXNIP) expression and thioredoxin-1 (TRX-1) nuclear localization. PLoS One, 2013; 8: e84588
- 20) Maruyama H, Dewachter C, Belhaj A, Rondelet B, Sakai S, Remmelink M, Vachiery JL, Naeije R, Dewachter L: Endothelin-Bone morphogenetic protein type 2 receptor interaction induces pulmonary artery smooth muscle cell hyperplasia in pulmonary arterial hypertension. J Heart Lung Transplant, 2015; 34: 468-478
