Advertisement

Transforming growth factor-β1 and inducible nitric oxide synthase signaling were involved in effects of prostaglandin E2 on progression of lower limb varicose veins

Open AccessPublished:January 19, 2021DOI:https://doi.org/10.1016/j.jvsv.2020.12.083

      Abstract

      Objective

      The vital pathogenesis of varicose veins includes remodeling of the extracellular matrix and decreased vascular tone. Prostaglandin E2 (PGE2), a small molecule substance and inflammatory medium that belongs to the arachidonic acid derivatives, has the capacity to influence the expression of metalloproteinase and the vascular tone of the venous wall. The purpose of the present study was to investigate the role of PGE2 in the development of varicose veins in lower limbs.

      Methods

      The collected venous specimens were analyzed using hematoxylin and eosin, Masson's trichrome, and immunohistochemical staining. Transforming growth factor (TGF)-β1, PGE2, CD31, and α-smooth muscle actin antibody were used to detect the expression and distribution of these proteins. The effect of PGE2 on the proliferation, migration, and tube formation capacity of human umbilical vein endothelial cells (HUVECs) was detected in vitro. The effect of TGF-β1 on the expression of PGE2 and matrix metalloproteinases (MMPs) was assessed using Western blotting. Quantitative reverse transcription polymerase chain reaction was used to evaluate the effect of PGE2 on the expression of nitric oxide synthase (NOS) and other genes.

      Results

      The expression of PGE2 and TGF-β1 in varicose veins was upregulated in the media tunica and intima tunica, and a strong positive correlation was found between PGE2 and TGF-β1 expression in both varicose veins (95% confidence interval, 0.5207-0.9582; R = 0.848; P = .0005) and normal veins (95% confidence interval, 0.2530-0.8532; R = 0.643; P = .003). PGE2 promoted the migration and tube formation ability of HUVECs. Moreover, PGE2 also upregulated the expression of MMP-1 and TGF-β1 in HUVECs and increased the mRNA level of inducible NOS.

      Conclusions

      PGE2 can affect the remodeling of the extracellular matrix and reduce the elasticity of the vascular walls by promoting the synthesis of TGF-β1 and MMP-1. PGE2 can also reduce the tension of the great saphenous vein by promoting the expression of inducible NOS, thus aggravating the blood stasis.

      Clinical Relevance

      Lower limb varicose veins are characterized by remodeling of the extracellular matrix and a lower vascular tone, which contribute to the reflux of venous blood and the higher pressure in the vascular wall. In the present study, prostaglandin E2 was shown to contribute to the progression of lower limb varicose veins. Prostaglandin E2 not only affected the remodeling of the extracellular matrix but also reduced the wall tension of the great saphenous veins by promoting the expression of inducible nitric oxide synthase, thus aggravating blood stasis. These two mechanisms might explain why tortuous and dilated varicose veins appear gradually.

      Keywords

      Article Highlights
      • Type of Research: A single-center, retrospective cohort study
      • Key Findings: Prostaglandin E2 (PGE2) and transforming growth factor (TGF)-β1 were upregulated in varicose veins, with a strong positive correlation found between PGE2 and TGF-β1 in both varicose veins (n = 19) and normal veins (n = 12). Moreover, PGE2 also upregulated the expression of inducible nitric oxide synthase, matrix metalloproteinase-1, and TGF-β1 in human umbilical vein endothelial cells.
      • Take Home Message: PGE2 can affect the remodeling of the extracellular matrix and reduce the elasticity of vascular walls by promoting the synthesis of TGF-β1 and matrix metalloproteinase-1. PGE2 can also reduce the tension of the great saphenous vein by promoting the expression of inducible nitric oxide synthase, thus aggravating the blood stasis.
      Varicose veins, especially lower extremity varicose veins, are a common and frequent presentation in the field of vascular surgery. Due to the dysfunction of the vascular valves or obstruction in the saphenous veins, the blood in the femoral vein flows in reverse to the great saphenous vein. This reflux becomes blocked when the person is standing or straining, because the intravenous pressure in the calf section becomes greatly elevated by ∼20 mm Hg.
      • Martin A.
      • Odling-Smee W.
      Pressure changes in varicose veins.
      Such pressure increases subsequently cause many typical symptoms, including the dilation and tortuosity of the superficial veins and dysfunction of the microvasculature of the skin.
      • Chang M.Y.
      • Chiang P.T.
      • Chung Y.C.
      • Ho S.Y.
      • Lin S.D.
      • Lin S.R.
      • et al.
      Apoptosis and angiogenesis in varicose veins using gene expression profiling.
      The incidence rate of varicose veins can be as great as 20% to 40%, and more than one half of those aged >40 years will experience varicose veins in the lower limbs to various degrees.
      • Chang M.Y.
      • Chiang P.T.
      • Chung Y.C.
      • Ho S.Y.
      • Lin S.D.
      • Lin S.R.
      • et al.
      Apoptosis and angiogenesis in varicose veins using gene expression profiling.
      The accumulation of venous blood in the branches or trunks of the great saphenous veins finally results in fatigue and swelling of the legs. In addition, the aberrant supply of oxygen and nutrition contributes to the development of skin symptoms, such as atrophy, desquamation, itching, and ulcers.
      • Naoum J.J.
      • Hunter G.C.
      Pathogenesis of varicose veins and implications for clinical management.
      In the development of varicose veins, remodeling of the extracellular matrix (ECM) and the low vascular tone are responsible for the venous blood reflux and the higher pressure in the vascular wall.
      • Raffetto J.D.
      Pathophysiology of chronic venous disease and venous ulcers.
      Specifically, ECM remodeling can alter the composition of the veins, making the venous wall tortuous and dilated. In addition, the lower vascular tone can dilate the vessels and make the valves fail to close completely.
      Prostaglandin E2 (PGE2), a small lipid molecule, is a 20-carbon unsaturated fatty acid and plays an important role in cellular growth.
      • Gomez I.
      • Foudi N.
      • Longrois D.
      • Norel X.
      The role of prostaglandin E2 in human vascular inflammation.
      PGE2 has a dilation effect on human saphenous veins and rabbit jugular veins and has the capacity to alter the expression of matrix metalloproteinases (MMPs).
      • Lawrence R.A.
      • Jones R.L.
      Investigation of the prostaglandin E (EP-) receptor subtype mediating relaxation of the rabbit jugular vein.
      ,
      • Foudi N.
      • Kotelevets L.
      • Gomez I.
      • Louedec L.
      • Longrois D.
      • Chastre E.
      Differential reactivity of human mammary artery and saphenous vein to prostaglandin E(2): implication for cardiovascular grafts.
      These characteristics provide the essential premise for the possible role of PGE2 in the progression of varicose veins.
      Transforming growth factor (TGF)-β1, which can change the ratio of MMPs and the tissue inhibitor of MMPs and regulates the secretion of ECM, is considered a crucial factor in the pathogenesis of varicose veins.
      • Serralheiro P.
      • Novais A.
      • Cairrão E.
      • Maia C.
      • Costa Almeida C.M.
      • Verde I.
      Variability of MMP/TIMP and TGF-beta1 receptors throughout the clinical progression of chronic venous disease.
      It seems that a constant relationship exists between TGF-β1 and PGE2. In the progression of renal disease or prostate cancer, TGF-β1 can directly stimulate the expression of PGE2, exerting enormous effects on the disease.
      • Vo B.T.
      • Morton Jr., D.
      • Komaragiri S.
      • Millena A.C.
      • Leath C.
      • Khan S.A.
      TGF-β effects on prostate cancer cell migration and invasion are mediated by PGE2 through activation of PI3K/AKT/mTOR pathway.
      ,
      • Jing L.
      • Yi-De Z.
      • Xiao-Lan C.
      • Xue-Ling Z.
      • Xu C.
      • Jian-Hua W.
      • et al.
      The protective effect of the EP2 receptor on TGF-β1 induced podocyte injury via the PI3K/Akt signaling pathway.
      However, the correlation of these two factors has not been studied in the occurrence or progression of varicose veins.
      In the present study, we explored the correlation between PGE2 and TGF-β1 and the possible effects of PGE2 on the progression of varicose veins in the hope of informing drug treatment for varicose veins in the future. We also explored whether PGE2 has the capacity to affect the expression of MMPs and alter the venous tone to some extent.

      Methods

       Specimen collection

      Varicose vein tissue samples were collected from 19 patients admitted for operative treatment of varicose veins. Control saphenous vein tissues were collected from the residual vein tissues of 12 patients who had undergone coronary artery bypass graft surgery (age, 50 ± 4 years). The specimens were all collected from the First Affiliated Hospital of Xi'an Jiaotong University (Shaanxi, China). The human ethics committees of the First Affiliated Hospital of Xi'an Jiaotong University approved the present study, which was executed in accordance with the principles of the Declaration of Helsinki and the recommendations of Good Clinical Practice. The samples were collected after the study subjects had provided written informed consent.
      Varicose veins were diagnosed from the physical examination and Doppler ultrasound findings. The patients enrolled in our study had had CEAP (clinical, etiology, anatomy, and pathophysiology classification) class 2 to 6 and varicose veins with clinical signs such as skin pigmentation, ulceration, and edema.
      • Raffetto J.D.
      Pathophysiology of chronic venous disease and venous ulcers.
      The CEAP class was C2 in 40%, C3 in 25%, C4 in 20%, C5 in 10%, and C6 in 5%. Patients with venous obstructions from neoplasms or deep vein thrombosis were excluded from the present study. The varicose great saphenous veins collected from the 19 patients were all dilated and >6 mm.

       Histopathologic analysis of tissue sections

      To analyze the venous structures using light microscopy, 20 varicose vein and 12 control saphenous vein samples were fixed in 4% formaldehyde and processed for paraffin embedding. The specimens had been removed from the operating room and kept refrigerated until use. Sections 5-μm thick were cut from the tissue blocks, and the sections were processed using a hematoxylin and eosin (H&E) staining kit and Masson's trichrome staining kit (both from Nanjing SenBeiJia Biological Technology Co, Ltd, Nanjing, China) in accordance with a previous study.
      • Aravind B.
      • Saunders B.
      • Navin T.
      • Sandison A.
      • Monaco C.
      • Paleolog E.M.
      • et al.
      Inhibitory effect of TIMP influences the morphology of varicose veins.

       Immunostaining of PGE2, TGF-β1, CD31, and α-smooth muscle actin

      We also stained the specimens for CD31 and α-smooth muscle actin (SMA) to confirm the presence of endothelium and evaluate the expression intensity of the vascular smooth muscle cells (VSMCs), respectively. After the paraffin-embedded tissue sections from the varicose and normal veins had been deparaffinized and rehydrated, they were incubated in a pressure kettle (Haier, Qingdao, China) with ethylenediaminetetraacetic acid antigen repair solution for 6 minutes. Endogenous peroxidase was then blocked using a 3% hydrogen peroxide solution. The sections were successively incubated with the primary antibody overnight in a 4°C refrigerator. The primary antibody included goat anti-rabbit CD31 antibody (1:30), goat anti-rabbit α-SMA antibody (1:80), goat anti-rabbit PGE2 antibody (1:100), and goat anti-rabbit TGF-β1 antibody (1:100; all from Abcam, Cambridge, UK). Subsequently, the sections were incubated with horseradish peroxidase-conjugated rabbit anti-human antibody (1:5000; Wuhan Boster Biological Technology, Ltd, Wuhan, China). Negative control sections were treated with phosphate-buffered saline (PBS) solution to replace the primary antibody. The sections were counterstained with hematoxylin and observed under a microscope.
      • Aravind B.
      • Saunders B.
      • Navin T.
      • Sandison A.
      • Monaco C.
      • Paleolog E.M.
      • et al.
      Inhibitory effect of TIMP influences the morphology of varicose veins.
      The specific scoring rules used were as follows: <5% pigmented cells, 0 points; 5% to 25%, 1 point; 26% to 50%, 2 points; 51% to 75%, 3 points; and >75% pigmented cells, 4 points. The staining intensity was scored as follows: negative, 0 point; weakly positive, 1 point; positive, 2 points; and strongly positive, 3 points. The final immunohistochemical (IHC) score was the staining percentage score multiplied by the staining intensity score. Four fields were randomly selected for each slice, and the average score was recorded. The outcome was evaluated by two of us independently. If the outcome of the two evaluators differed, a third person also evaluated the sections.

       Cell culture

      Human umbilical vein endothelial cells (HUVECs) were cultured at 37°C and 5% carbon dioxide in high-glucose HyClone Dulbecco's modified Eagle medium (DMEM; Cytiva, Global Life Sciences Solutions USA LLC, Marlborough, Mass) containing 10% fetal bovine serum (FBS) and a penicillin–streptomycin mixture (1:100; containing 10,000 U/mL penicillin and 10 mg/mL streptomycin; Sigma-Aldrich, St Louis, Mo). When the cells had reached 80% confluence, they were passaged by digestion with 0.25% trypsin containing ethylenediaminetetraacetic acid (Beyotime Biotechnology, Nantong, China).

       Cell counting kit-8 assay

      To evaluate the proliferation of HUVECs after adding PGE2, we performed the cell counting kit-8 assay. When the HUVECs had reached 80% confluence, they were starved for 24 hours and then counted with a hemocytometer. After the cell concentration was adjusted to 5 × 104 cells/mL, a 100-μL cell suspension containing PGE2 (Sigma-Aldrich) was added to 96-well plates. The study was divided into four groups with different concentrations of PGE2 (0, 0.1, 1, and 10 μM). An equal volume of PBS was added to the remaining surrounding wells. After labeling the plates, the cell culture plate was placed in the incubator for 24 and 48 hours. A 10-μL cell counting kit-8 solution (Beyotime Biotechnology) was added to the well and incubated for 2 to 4 hours.
      • Sun C.
      • Feng S.B.
      • Cao Z.W.
      • Jun Jie B.
      • Hou H.
      Up-regulated expression of matrix metalloproteinases in endothelial cells mediates platelet microvesicle-induced angiogenesis.
      Next, the absorbance (optical density) at 450 nm was measured using a microplate reader (BioTek Instruments, Winooski, Vt).

       Wound healing assay

      To evaluate the migration of HUVECs after adding PGE2, we used a wound healing assay. After the HUVECs had grown to 80% confluence, a vertical line was created by scraping the cells with a sterile pipette tip to create a scratch wound.
      • Sun C.
      • Feng S.B.
      • Cao Z.W.
      • Jun Jie B.
      • Hou H.
      Up-regulated expression of matrix metalloproteinases in endothelial cells mediates platelet microvesicle-induced angiogenesis.
      Next, the cells were washed twice with PBS solution and incubated in FBS-free DMEM at different concentrations (0, 0.1, 1, or 10 μM) of PGE2. After incubation for 12 and 24 hours, the cells that had migrated to the wound area were photographed using an inverted light microscope (Leica, Wetzler, Germany).

       Transwell assay

      The migration capacity of the HUVECs was determined using a Boyden Transwell chamber (Corning, Corning, NY) with 8.0-μm pore-size polycarbonate filter inserts in 24-well plates.
      • Sun C.
      • Feng S.B.
      • Cao Z.W.
      • Jun Jie B.
      • Hou H.
      Up-regulated expression of matrix metalloproteinases in endothelial cells mediates platelet microvesicle-induced angiogenesis.
      After the HUVECs had grown to 80% confluence, the cells were starved and treated with different concentrations of PGE2 (0, 0.1, 1, or 10 μM) for 24 hours. Next, the cells were digested and collected from six-well plates. The resuspended HUVECs (1 × 105/mL; 100 μL) were seeded in the upper compartment. The lower compartment contained 600 μL of DMEM supplemented with 10% FBS to stimulate cell migration. After incubating for 24 hours at 37°C, the medium in the upper chamber was aspirated, and the cells on the upper surface of the membrane were removed by gentle scraping with a cotton swab. After fixation with methanol at room temperature for 15 minutes and staining with crystal violet stain solution for 20 minutes, the migratory cells on the lower surface of the membrane of five random fields were counted.

       Tube formation assay

      To analyze the function of PGE2 on the angiogenesis capacity of HUVECs, we used the tube formation assay. Matrigel (Corning) was thawed on ice at 4°C overnight. The solution was added to a 96-well plate (50 μL/well) and incubated at 37°C for 1 hour to allow for matrix gel solidification. Next, the HUVECs were trypsinized and seeded (1 × 104 cells/well) with different concentrations of PGE2 (0, 0.1, 1, or 10 μM) in serum-free DMEM. The plate was then incubated at 37°C for 6 hours. Tube formation was observed under an inverted light microscope (10×), and five representative fields from each well were photographed.
      • Sun C.
      • Feng S.B.
      • Cao Z.W.
      • Jun Jie B.
      • Hou H.
      Up-regulated expression of matrix metalloproteinases in endothelial cells mediates platelet microvesicle-induced angiogenesis.

       Western blot analysis

      The HUVEC monolayers were lysed, and 40 μg of denatured protein samples were loaded onto gels and separated by gel electrophoresis, followed by transfer to a polyvinylidene fluoride membrane (Roche, Basel, Switzerland). The membranes were blocked and incubated overnight at 4°C in the presence of a primary antibody (dilution in accordance with the manufacturer's instructions). The primary antibody included MMP-1 antibody (1:1000), MMP-2 antibody (1:1000), MMP-9 antibody (1:1000), and TGF-β1 antibody (1:1000; Abcam for all) and glyceraldehyde 3-phosphate dehydrogenase antibody (1:5000; Wuhan Boster Biological Technology, Ltd). Next, the membranes were incubated with the secondary horseradish peroxidase-conjugated antibody (1:5000; Wuhan Boster Biological Technology, Ltd) and developed with enhanced chemiluminescence substrate solution (MilliporeSigma, Burlington, Mass) according to the manufacturer's instructions. Next, the protein band signal intensity was detected using chemiluminescence and quantified using ImageJ software (National Institutes of Health, Bethesda, Md). The protein levels of PGE2 and TGF-β1 in the tissues were not detected, because a very small amount of the normal great saphenous vein had been collected.

       RNA isolation and quantitative reverse transcription polymerase chain reaction analysis

      Total RNA was extracted using the RNA Extraction Kit (Takara Bio Inc, Kusatsu, Japan), and cDNA was synthesized using the Primescript RT Master Mix (Takara Bio Inc). The primers used for real-time polymerase chain reaction analyses are listed in the Table. The reaction conditions were as follows: 95°C for 5 seconds and 60°C for 30 seconds, with 40 cycle counts. Each sample was analyzed at least in triplicate.
      TablePrimers used in quantitative real-time polymerase chain reaction analyses
      GeneSequence (5′ to 3′)
      CSE
      Forward: AAAGACGCCTCCTCACAAGG
      Reverse: AAGGCAATTCCTAGTGGGATTTC
      iNOS
      Forward: AATGTGGAGAAAGCCCCCTG
      Reverse: TGCATCCAGCTTGACCAGAG
      eNOS
      Forward: AGCTGCCCTGATGGAGATGT
      Reverse: CCCGAACACACAGAACCTGA
      HO-1
      Forward: CTCCCAGGGCCATGAACTTT
      Reverse: GGGAAGATGCCATAGGCTCC
      β-Actin
      Forward: CTCCATCCTGGCCTCGCTGT
      Reverse: GCTGTCACCTTCACCGTTCC
      CSE, Cystathionine gamma lyase; eNOS, endothelial nitric oxide synthase; HO-1, heme oxygenase-1; iNOS, inducible nitric oxide synthase.

       Statistical analysis

      The results are presented as the mean ± standard deviation, and the data were analyzed using Prism, version 5 (GraphPad, San Diego, Calif). Differences in the IHC scores for the varicose veins and normal veins were analyzed using the Student t test, with a probability value of P < .05 considered statistically significant. Differences among three or more experimental groups in the cell study were analyzed using one-way analysis of variance, followed by a Tukey-Kramer multiple comparisons test, with P values <.05 considered statistically significant.

      Results

       Clinical results

      The study population consisted of 19 patients with varicose veins (13 men and 6 women) and 12 control patients (9 men and 3 women) with coronary artery disease who had undergone coronary artery bypass graft surgery. The median age of the 19 patients with varicose veins was 38 years (range, 28-66 years) and that of the 12 control subjects was 51 years (range, 42-80 years). None of the control subjects had clinical evidence of venous insufficiency or varicose veins.

       Histologic features of varicose and normal veins

      The sections processed using H&E staining revealed that the tunica media of the varicose veins was thicker than that of the normal veins, with an apparently thicker subintimal area in the varicose veins (Fig 1, A). Moreover, the distribution of the smooth muscle cells (SMCs) was not as even as that in the normal veins. The orientation of the SMCs was mainly circular in the normal veins and they were well-organized. In contrast, the orientation of the cells in the varicose veins was relatively unorganized and irregular. Masson's trichrome staining showed large amounts of collagen deposited in the ECM of the varicose veins (Fig 1, B).
      Figure thumbnail gr1
      Fig 1Histologic features of varicose veins and normal veins. A, Hematoxylin and eosin (H&E) staining of the sections. B, Masson trichrome staining of the sections. Scale bar equals 100 μm.

       Expression of PGE2, TGF-β1, CD31, and α-SMA in varicose and normal veins

      In the tunica intima and tunica media, the expression of PGE2 and TGF-β1 in the varicose veins was much greater than that in the normal veins (Fig 2, A). After scoring the sections, the IHC score for these two factors in the varicose veins was significantly greater statistically (Fig 2, B). Next, the correlation of PGE2 and TGF-β1 was calculated, and both factors correlated positively in the varicose veins and normal veins both (Fig 2, B). IHC staining was performed using CD31 and α-SMA, the molecular marker of endothelial cells and SMCs, respectively. In addition to the stained intima, CD31 antibody stained the vasa vasorum in the tunica media, with the density higher in the varicose veins (Fig 2, C). In addition, the thicker subintimal area was stained by α-SMA antibody, manifesting its feature of SMCs (Fig 2, C). Further results of immunofluorescent staining showed that PGE2 was significantly expressed in both CD31+ endothelial cells and VSMCs (Fig 2, D).
      Figure thumbnail gr2
      Fig 2Expression of prostaglandin E2 (PGE2), transforming growth factor (TGF)-β1, CD31, and α-smooth muscle actin (SMA) in varicose veins and normal veins. A, Expression of PGE2 and TGF-β1 in the specimens. Scale bar equals 100 μm. B, Results of statistical analysis of PGE2 and TGF-β1. C, Expression of CD31 and α-SMA in the specimens. Scale bar equals 250 μm. ∗∗∗Significantly different (P < .05) compared with untreated controls.

       Effects of PGE2 on proliferation, migration, and tube formation capacity of HUVECs

      The addition of different concentrations of PGE2 (0.1, 1, or 10 μM) into the cultured HUVECs and incubation for 24 and 48 hours revealed that PGE2 had no proliferation effects on the HUVECs (Fig 3, A). This result suggested that the effects of PGE2 did not result from the regulation of the cell number of HUVECs. Regarding the migration capacity, we found that PGE2 could facilitate the migration of HUVECs in the wound healing assay and Transwell assay, and 1 μM PGE2 seemed to have the strongest effect (Fig 3, B and C). Furthermore, we detected capillary-like structures using the tube formation assay and found that PGE2 had the capacity to promote the tube formation capacity of HUVECs (Fig 3, D).
      Figure thumbnail gr3
      Fig 3Effects of prostaglandin E2 (PGE2) on the proliferation, migration, and tube formation capacity of human umbilical vein endothelial cells (HUVECs). A, PGE2 had no proliferation effect on the HUVECs. B, PGE2 facilitated the migration of HUVECs in the wound healing assay. Scale bar equals 100 μm. C, PGE2 facilitated the migration of HUVECs in the Transwell assay. Scale bar equals 50 μm. D, PGE2 promoted the tube formation capacity of HUVECs. Scale bar equals 100 μm. ∗Significantly different (P < .05) compared with untreated controls.

       PGE2 upregulates MMP-1 and TGF-β1 expression

      MMPs and TGF-β1 are closely associated with the remodeling of ECM. MMPs can degrade the ECM, and TGF-β1 can boost the secretion of ECM. To explore the influence of PGE2 on these factors, we added 1 μM PGE2 to HUVECs and incubated for 24 hours. Finally, we found that the expression of MMP-1 and TGF-β1 was upregulated but the expression of MMP-2 and MMP-9 was not altered (Fig 4).
      Figure thumbnail gr4
      Fig 4Prostaglandin E2 (PGE2) upregulated the expression of matrix metalloproteinase (MMP)-1 and transforming growth factor (TGF)-β1. ∗Significantly different (P < .05) compared with untreated controls.

       PGE2 elevates mRNA level of inducible nitric oxide (NO) synthase (iNOS) in HUVECs

      Therefore, endothelial NO synthase (eNOS), iNOS, cystathionine gamma lyase (CSE), and heme oxygenase (HO)-1 were selectively measured. We found no changes in the mRNA level of CSE, HO-1, or eNOS (Fig 5). The mRNA level of iNOS was upregulated, suggesting that PGE2 could facilitate the generation of NO in vitro.
      Figure thumbnail gr5
      Fig 5Prostaglandin E2 (PGE2) elevated the mRNA level of inducible nitric oxide synthase (iNOS) in human umbilical vein endothelial cells (HUVECs). ∗Significantly different (P < .05) compared with untreated controls.

      Discussion

      Varicose veins of the lower limbs are one of the most common diseases treated by vascular surgery. The dilated and tortuous great saphenous vein, not only affects the beauty of the legs, but also seriously affects patients' quality of life.
      • Takase S.
      • Pascarella L.
      • Lerond L.
      • Bergan J.J.
      • Nbein G.W.
      Venous hypertension, inflammation and valve remodeling.
      Treatment of varicose veins will often be surgery or interventional therapies, which can achieve immediate results. Therefore, research on the treatment of varicose veins of the great saphenous vein of the lower limbs has mainly focused on clinical treatment, with relatively fewer basic research studies. In the present study, we first collected the remaining normal veins from coronary artery bypass grafting surgery and varicose veins after great saphenous vein stripping surgery and analyzed them using H&E, Masson's trichrome, and IHC staining. The most obvious pathologic changes found in the varicose veins were the thicker media of the vessels and the deposition of collagen in ECM. Studies have shown that the expression of collagen type I, collagen type IV, laminin, and tenascin is upregulated in varicose veins and that the expression of collagen type III is downregulated.
      • Kirsch D.
      • Dienes H.P.
      • Küchle R.
      • Duschner H.
      • Wahl W.
      • Böttger T.
      • et al.
      Changes in the extracellular matrix of the vein wall—the cause of primary varicosis?.
      This change in composition can increase the toughness of the blood vessels and weaken the elasticity of the tube wall and, thus, affect the constriction function of the blood vessels.
      • Buján J.
      • Gimeno M.J.
      • Jiménez J.A.
      • Kielty C.M.
      • Mecham R.P.
      • Juan M.B.
      Expression of elastic components in healthy and varicose veins.
      Medication is often used as an adjunct therapy; however, its use is not appropriate for all patients at different stages of varicosity. Drug treatment can effectively relieve patients' symptoms, with different therapeutic uses at different stages of the disease. The drugs used most often include flavonoids and coumarin.
      • Raffetto J.D.
      • Khalil R.A.
      Ca(2+)-dependent contraction by the saponoside escin in rat vena cava: implications in venotonic treatment of varicose veins.
      These drugs can reduce the feelings of heaviness, swelling, pain, edema, and other symptoms. They can improve the venous blood wall tension, reduce the permeability of the microvascular, and promote the return of lymphatic fluid and have certain anti-inflammatory effects.
      • Hagen M.D.
      • Johnson E.D.
      • Adelman A.
      What treatments are effective for varicose veins?.
      In the present study, our results suggest that upregulation of PGE2 in varicose veins can contribute to its occurrence. Hence, the inhibition of PGE2 or its receptor might block the pathogenesis of varicose veins by affecting ECM remodeling and vascular tone. In clinical studies, an inhibitor of E-type prostanoid receptor 4 (a receptor of PGE2) has been used to treat cancer. However, as an inflammatory cytokine, PGE2 has rarely been studied in the field of vascular surgery. Therefore, further study of the pathologic mechanisms of varicose veins could help inform better drug treatment of this disease.
      In addition, we found that PGE2 in varicose veins was highly expressed in the media and intima, which could promote the proliferation of VSMCs. Biagi et al
      • Biagi G.
      • Lapilli A.
      • Zendron R.
      • Piccinni L.
      • Coccheri S.
      Prostanoid production in varicose veins: evidence for decreased prostacyclin with increased thromboxane A2 and prostaglandin E2 formation.
      reported that varicose veins can generate more PGE2 than can normal veins after adding arachidonic acid ex vivo. This is consistent with our results. However, Gomez et al
      • Gomez I.
      • Benyahia C.
      • Louedec L.
      • Leséche G.
      • Jacob M.-P.
      • Longrois D.
      • et al.
      Decreased PGE2 content reduces MMP-1 activity and consequently increases collagen density in human varicose vein.
      showed that the expression of PGE2 and its receptor E-type prostanoid receptor 4 was downregulated. This difference might have resulted from the following reasons.
      • Gomez I.
      • Benyahia C.
      • Louedec L.
      • Leséche G.
      • Jacob M.-P.
      • Longrois D.
      • et al.
      Decreased PGE2 content reduces MMP-1 activity and consequently increases collagen density in human varicose vein.
      First, the age of the patients in the control group in our study was much younger than that in the study by Gomez et al.
      • Gomez I.
      • Benyahia C.
      • Louedec L.
      • Leséche G.
      • Jacob M.-P.
      • Longrois D.
      • et al.
      Decreased PGE2 content reduces MMP-1 activity and consequently increases collagen density in human varicose vein.
      Second, the patient groups stratified by CEAP clinical class were significantly different. Finally, the collected varicose veins in our study were all dilated, although they were not all dilated in the study by Gomez et al.
      • Gomez I.
      • Benyahia C.
      • Louedec L.
      • Leséche G.
      • Jacob M.-P.
      • Longrois D.
      • et al.
      Decreased PGE2 content reduces MMP-1 activity and consequently increases collagen density in human varicose vein.
      Eschrich et al
      • Eschrich J.
      • Meyer R.
      • Kuk H.
      • Andreas H.W.
      • Thomas N.
      • Sebastian D.
      • et al.
      Varicose remodeling of veins is suppressed by 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors.
      observed that the expression of proliferating cell nuclear antigen in varicose veins was upregulated and the proliferation capacity of VSMCs was enhanced, which might be closely related to the thickening of the middle membrane. Although PGE2 had no effect on the proliferation of HUVECs, it did promote the migration and angiogenic abilities of HUVECs. In the present study, more CD31+ microvessels were found in the varicose veins, suggesting that an increase occurs in the vasa vasorum in the tissue.
      Because of the long-term stasis of the venous blood and the continuous consumption of oxygen by the vasa vasorum, the oxygen replacement rate of the blood in the lumen will decrease. In addition, the increase in the hydrostatic pressure in the lumen will compress the vasa vasorum, affecting the blood supply of the tube wall.
      • Lim C.S.
      • Kiriakidis S.
      • Paleolog E.M.
      • Alun H.D.
      Increased activation of the hypoxia-inducible factor pathway in varicose veins.
      These two factors work together to further reduce the partial oxygen pressure in the venous blood. This environment might promote the expression of nine hypoxia-inducing factors (HIFs), which, in turn, will lead to angiogenesis. Lim et al
      • Lim C.S.
      • Kiriakidis S.
      • Paleolog E.M.
      • Alun H.D.
      Increased activation of the hypoxia-inducible factor pathway in varicose veins.
      showed that the expression of HIF-1 and HIF-2 in varicose veins was greater than that in normal veins, which also supports the existence of this hypoxic state.
      IHC staining showed that staining of α-SMA was somewhat deeper in normal veins (strongly positive) and slightly lighter in varicose veins (positive). This might be related to the phenotype transformation of VSMCs in varicose veins. The increased pressure in blood vessels and the high expression of PGE2 can both reduce the expression of contractile markers and cause cells to transform to the synthetic phenotype, which directly affects the contractile capacity of VSMCs and reduces the tension of the venous wall.
      • Eschrich J.
      • Meyer R.
      • Kuk H.
      • Andreas H.W.
      • Thomas N.
      • Sebastian D.
      • et al.
      Varicose remodeling of veins is suppressed by 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors.
      ,
      • Adedoyin O.O.
      • Loftin C.D.
      Microsomal prostaglandin E synthase-1 expression by aortic smooth muscle cells attenuates the differentiated phenotype.
      When PGE2 was added to aortic VSMCs for incubation, it was found that many contractile-related molecular markers were downregulated, including α-actin, smooth muscle 22α, SMC differentiation-specific antigen, and myosin heavy chain.
      • Adedoyin O.O.
      • Loftin C.D.
      Microsomal prostaglandin E synthase-1 expression by aortic smooth muscle cells attenuates the differentiated phenotype.
      The decrease in these markers reflected the reduced vasoconstriction.
      We also found that PGE2 promoted the production of TGF-β1 and MMP-1 in HUVECs. TGF-β1 can promote cell secretion of the ECM and enhance the activity of the tissue inhibitor of MMPs, and MMP-1 can directly degrade the ECM.
      • Serralheiro P.
      • Novais A.
      • Cairrão E.
      • Maia C.
      • Costa Almeida C.M.
      • Verde I.
      Variability of MMP/TIMP and TGF-beta1 receptors throughout the clinical progression of chronic venous disease.
      Many factors affect the synthesis and degradation of the ECM. Our results only indicated that PGE2 has an important effect on this synthesis and degradation network, and endothelial cells might be involved in the remodeling of the ECM through the paracrine pathway. Subsequent experiments might directly measure the influence of PGE2 on the expression of collagen and other molecules in HUVECs and VSMCs to determine the specific changes in the ECM.
      In addition to the irreversible ECM remodeling, dilated and tortuous saphenous veins will result from the reversible reduction of vascular wall tension.
      • Raffetto J.D.
      • Khalil R.A.
      Matrix metalloproteinases and their inhibitors in vascular remodeling and vascular disease.
      Gas transmitters, including NO, carbon monoxide (CO), and hydrogen sulfate (H2S), can be produced by endothelial cells and act on the smooth muscle layer of the vascular media by diffusion, subsequently reducing the tension of the vascular wall and causing vasodilation.
      • Giles T.D.
      • Sander G.E.
      • Nossaman B.D.
      • Kadowitz P.J.
      Impaired vasodilation in the pathogenesis of hypertension: focus on nitric oxide, endothelial-derived hyperpolarizing factors, and prostaglandins.
      The key synthase of these gas molecules was detected to indirectly determine their expression.
      NO is able to inhibit the contraction of VSMCs and lower vessel tension, causing blood stasis in lower extremity varicose veins. As members of the endothelium-derived hyperpolarization factors, not only NO, but also CO and H2S, can stimulate vessel dilation.
      • Giles T.D.
      • Sander G.E.
      • Nossaman B.D.
      • Kadowitz P.J.
      Impaired vasodilation in the pathogenesis of hypertension: focus on nitric oxide, endothelial-derived hyperpolarizing factors, and prostaglandins.
      In endothelial cells, NO is mainly synthesized by eNOS and iNOS, with neuronal NO synthase mainly expressed in the nervous system.
      • Forstermann U.
      • Sessa W.C.
      Nitric oxide synthases: regulation and function.
      H2S is mainly produced by three isozymes, CSE, cystathionine-β-synthase (CBS), and 3-mercaptopyruvate sulfurtransferase, which are tissue specific. The nervous system mainly expresses CBS and 3-mercaptopyruvate sulfurtransferase, and the cardiovascular system mainly expresses CSE.
      • Kanagy N.L.
      • Szabo C.
      • Papapetropoulos A.
      Vascular biology of hydrogen sulfide.
      ,
      • Sun H.L.
      • Li C.D.
      The role of gas signaling molecule hydrogen sulfide in regulating apoptosis and autophagy.
      The ileum mucosa, gastric mucosa, liver, and kidney express both CBS and CSE.
      • Sun H.L.
      • Li C.D.
      The role of gas signaling molecule hydrogen sulfide in regulating apoptosis and autophagy.
      CO in vivo is mainly produced by HO-1 and HO-2 in the process of catalysis of heme to bilirubin, in which HO-2 is constitutively expressed in cells, and its expression is seldom regulated by outside factors.
      • Lo S.
      • Palma S.D.
      • Yusuf H.
      • McCombe A.W.
      Constitutive (HO-2) and inducible (HO-1) haem oxygenase in pleomorphic adenomas of the human parotid: an immunocytochemical study.
      Our results showed that PGE2 promoted the expression of iNOS in HUVECs. The expression of iNOS can be induced by many inflammatory stimuli, including lipopolysaccharide and cytokines. However, the expression of eNOS was not affected by PGE2 in our study. Because eNOS is always constitutively expressed in vivo, its expression would not be affected. NO produced by endothelial cells could upregulate the content of cyclic guanosine monophosphate in VSMCs through diffusion, significantly reducing the tone of the venous wall and aggravating blood deposition.
      • Forstermann U.
      • Closs E.I.
      • Pollock J.S.
      • Schwarz P.
      • Gath I.
      • Kleinert H.
      Nitric oxide synthase isozymes: characterization, purification, molecular cloning, and functions.
      CO and H2S are members of the endothelium-derived hyperpolarization factor family. As important products secreted by endothelial cells, they can regulate vascular tension by affecting the membrane potential.
      • Barbara B.
      • Enrica C.
      • Daria S.
      • Stefano L.
      • Flavio C.
      Hydrogen sulfide in the mouse ductus arteriosus: a naturally occurring relaxant with potential EDHF function.
      However, no significant changes were found in the expression of CSE and HO-1, indicating that the PGE2 in endothelial cells exerted very limited influences on these two gas molecules. Moreover, Gomez et al
      • Gomez I.
      • Ozen G.
      • Deschildre C.
      • Amgoud Y.
      • Boubaya L.
      • Gorenne I.
      • et al.
      Reverse regulatory pathway (H2S/PGE2/MMP) in human aortic aneurysm and saphenous vein varicosity.
      found that CSE expression in varicose veins is higher than that in normal veins. However, what they detected contained both endothelium and VSMCs. It was difficult to determine whether the CSE level in endothelial cells is upregulated.
      • Gomez I.
      • Ozen G.
      • Deschildre C.
      • Amgoud Y.
      • Boubaya L.
      • Gorenne I.
      • et al.
      Reverse regulatory pathway (H2S/PGE2/MMP) in human aortic aneurysm and saphenous vein varicosity.
      The present study had some limitations. First, vein specimens can be divided into many sections, including the distal, middle, and proximal parts. Different sections of a vessel might generate different outcomes.
      • Naoum J.J.
      • Hunter G.C.
      Pathogenesis of varicose veins and implications for clinical management.
      Moreover, the major part of the saphenous vein consists of SMCs and not endothelial cells. Future studies should use human umbilical SMCs to explore the function of PGE2 in the tunica media. Furthermore, the various harvesting techniques, the timing of initiating the protocols, and the elapsed time in the operating room (the specimens were not processed until the surgery had been completed) could potentially affect the results. In addition, the chronic inflammatory condition of the donors' normal veins could have affected the expression of many cytokines and change the extracellular structure to a greater or lesser degree. This could have resulted in some unexpected outcomes. Finally, the sample size in the present study was relatively small, and Western blotting should be performed to observe the differences in the protein expression levels.

      Conclusions

      PGE2 can affect, not only the remodeling of the ECM, but also reduce the wall tension of the great saphenous veins by promoting the expression of iNOS, thus aggravating blood stasis. Finally, the veins will become tortuous and dilated.

      Author contributions

      Conception and design: JW, JG, YL, QM, SL
      Analysis and interpretation: JW, JG, YL, QM, JF
      Data collection: JW, JG, YL, QM, JF
      Writing the article: JW, JG, YL, QM, SL
      Critical revision of the article: JF, SL
      Final approval of the article: JW, JG, YL, QM, JF, SL
      Statistical analysis: Not applicable
      Obtained funding: JW, JF, SL
      Overall responsibility: SL
      JW and JG contributed equally to this article and share co-first authorship.

      References

        • Martin A.
        • Odling-Smee W.
        Pressure changes in varicose veins.
        Lancet. 1976; 307: 768-770
        • Chang M.Y.
        • Chiang P.T.
        • Chung Y.C.
        • Ho S.Y.
        • Lin S.D.
        • Lin S.R.
        • et al.
        Apoptosis and angiogenesis in varicose veins using gene expression profiling.
        Fooyin J Health Sci. 2009; 1: 85-91
        • Naoum J.J.
        • Hunter G.C.
        Pathogenesis of varicose veins and implications for clinical management.
        Vascular. 2007; 15: 242-249
        • Raffetto J.D.
        Pathophysiology of chronic venous disease and venous ulcers.
        Surg Clin North Am. 2018; 98: 337-347
        • Gomez I.
        • Foudi N.
        • Longrois D.
        • Norel X.
        The role of prostaglandin E2 in human vascular inflammation.
        Prostaglandins Leukot Essent Fatty Acids. 2013; 89: 55-63
        • Lawrence R.A.
        • Jones R.L.
        Investigation of the prostaglandin E (EP-) receptor subtype mediating relaxation of the rabbit jugular vein.
        Br J Pharmacol. 1992; 105: 817-824
        • Foudi N.
        • Kotelevets L.
        • Gomez I.
        • Louedec L.
        • Longrois D.
        • Chastre E.
        Differential reactivity of human mammary artery and saphenous vein to prostaglandin E(2): implication for cardiovascular grafts.
        Br J Pharmacol. 2011; 163: 826-834
        • Serralheiro P.
        • Novais A.
        • Cairrão E.
        • Maia C.
        • Costa Almeida C.M.
        • Verde I.
        Variability of MMP/TIMP and TGF-beta1 receptors throughout the clinical progression of chronic venous disease.
        Int J Mol Sci. 2017; 19: 6
        • Vo B.T.
        • Morton Jr., D.
        • Komaragiri S.
        • Millena A.C.
        • Leath C.
        • Khan S.A.
        TGF-β effects on prostate cancer cell migration and invasion are mediated by PGE2 through activation of PI3K/AKT/mTOR pathway.
        Endocrinology. 2013; 154: 1768-1779
        • Jing L.
        • Yi-De Z.
        • Xiao-Lan C.
        • Xue-Ling Z.
        • Xu C.
        • Jian-Hua W.
        • et al.
        The protective effect of the EP2 receptor on TGF-β1 induced podocyte injury via the PI3K/Akt signaling pathway.
        PLoS One. 2018; 13: e0197158
        • Giles T.D.
        • Sander G.E.
        • Nossaman B.D.
        • Kadowitz P.J.
        Impaired vasodilation in the pathogenesis of hypertension: focus on nitric oxide, endothelial-derived hyperpolarizing factors, and prostaglandins.
        J Clin Hypertens. 2012; 14: 198-205
        • Forstermann U.
        • Sessa W.C.
        Nitric oxide synthases: regulation and function.
        Eur Heart J. 2012; 33: 829-837
        • Kanagy N.L.
        • Szabo C.
        • Papapetropoulos A.
        Vascular biology of hydrogen sulfide.
        Am J Cell Physiol. 2017; 312: C537-C549
        • Sun H.L.
        • Li C.D.
        The role of gas signaling molecule hydrogen sulfide in regulating apoptosis and autophagy.
        Chin Med J. 2013; 93: 2520-2522
        • Lo S.
        • Palma S.D.
        • Yusuf H.
        • McCombe A.W.
        Constitutive (HO-2) and inducible (HO-1) haem oxygenase in pleomorphic adenomas of the human parotid: an immunocytochemical study.
        J Laryngol Otol. 2005; 119: 179-183
        • Takase S.
        • Pascarella L.
        • Lerond L.
        • Bergan J.J.
        • Nbein G.W.
        Venous hypertension, inflammation and valve remodeling.
        Eur J Vasc Endovasc Surg. 2004; 28: 484-493
        • Raffetto J.D.
        • Khalil R.A.
        Ca(2+)-dependent contraction by the saponoside escin in rat vena cava: implications in venotonic treatment of varicose veins.
        J Vasc Surg. 2011; 52: 489-496
        • Hagen M.D.
        • Johnson E.D.
        • Adelman A.
        What treatments are effective for varicose veins?.
        J Fam Pract. 2003; 52: 329-331
        • Kirsch D.
        • Dienes H.P.
        • Küchle R.
        • Duschner H.
        • Wahl W.
        • Böttger T.
        • et al.
        Changes in the extracellular matrix of the vein wall—the cause of primary varicosis?.
        Vasa. 2000; 29: 173-177
        • Buján J.
        • Gimeno M.J.
        • Jiménez J.A.
        • Kielty C.M.
        • Mecham R.P.
        • Juan M.B.
        Expression of elastic components in healthy and varicose veins.
        World J Surg. 2003; 27: 901-905
        • Eschrich J.
        • Meyer R.
        • Kuk H.
        • Andreas H.W.
        • Thomas N.
        • Sebastian D.
        • et al.
        Varicose remodeling of veins is suppressed by 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors.
        J Am Heart Assoc. 2016; 5: e002405
        • Lim C.S.
        • Kiriakidis S.
        • Paleolog E.M.
        • Alun H.D.
        Increased activation of the hypoxia-inducible factor pathway in varicose veins.
        J Vasc Surg. 2012; 55: 1427-1439
        • Adedoyin O.O.
        • Loftin C.D.
        Microsomal prostaglandin E synthase-1 expression by aortic smooth muscle cells attenuates the differentiated phenotype.
        J Cardiovasc Pharmacol. 2016; 68: 127-142
        • Raffetto J.D.
        • Khalil R.A.
        Matrix metalloproteinases and their inhibitors in vascular remodeling and vascular disease.
        Biochem Pharmacol. 2008; 75: 346-359
        • Forstermann U.
        • Closs E.I.
        • Pollock J.S.
        • Schwarz P.
        • Gath I.
        • Kleinert H.
        Nitric oxide synthase isozymes: characterization, purification, molecular cloning, and functions.
        Hypertension. 1994; 23: 1121-1131
        • Barbara B.
        • Enrica C.
        • Daria S.
        • Stefano L.
        • Flavio C.
        Hydrogen sulfide in the mouse ductus arteriosus: a naturally occurring relaxant with potential EDHF function.
        Am J Physiol Heart Circ Physiol. 2013; 304: H927-H934
        • Gomez I.
        • Ozen G.
        • Deschildre C.
        • Amgoud Y.
        • Boubaya L.
        • Gorenne I.
        • et al.
        Reverse regulatory pathway (H2S/PGE2/MMP) in human aortic aneurysm and saphenous vein varicosity.
        PLoS One. 2016; 11: e0158421
        • Aravind B.
        • Saunders B.
        • Navin T.
        • Sandison A.
        • Monaco C.
        • Paleolog E.M.
        • et al.
        Inhibitory effect of TIMP influences the morphology of varicose veins.
        Eur J Vasc Endovasc Surg. 2010; 40: 754-765
        • Sun C.
        • Feng S.B.
        • Cao Z.W.
        • Jun Jie B.
        • Hou H.
        Up-regulated expression of matrix metalloproteinases in endothelial cells mediates platelet microvesicle-induced angiogenesis.
        Cell Physiol Biochem. 2017; 41: 2319
        • Biagi G.
        • Lapilli A.
        • Zendron R.
        • Piccinni L.
        • Coccheri S.
        Prostanoid production in varicose veins: evidence for decreased prostacyclin with increased thromboxane A2 and prostaglandin E2 formation.
        Angiology. 1988; 39: 1036-1042
        • Gomez I.
        • Benyahia C.
        • Louedec L.
        • Leséche G.
        • Jacob M.-P.
        • Longrois D.
        • et al.
        Decreased PGE2 content reduces MMP-1 activity and consequently increases collagen density in human varicose vein.
        PLoS One. 2014; 9: e88021