Forty-one patients with EPs who underwent hysteroscopic surgery at Dongying People’s Hospital during their menstrual augmentation period were enrolled in the study. The excised EP samples were divided into endometriosis (n = 23) and non-endometriosis (n = 18) groups, depending on the presence or absence of endometriosis. Patients with EPs, systemic inflammatory diseases, a history of hormonal therapy within 3 months prior to the operation, uterine malformations, uterine adhesions, endometrial dysplasia, malignant neoplasia, or uterine fibroids were excluded. One portion of the EP samples was fixed with 4% paraformaldehyde (#G1101-500ML; YuBioLab, Beijing, China) and subjected to paraffin embedding for immunohistochemical (IHC) analysis, while the other portion was used to isolate primary ESCs. The study was carried out in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Dongying People’s Hospital (Approval Number 2024 [019]), and written informed consent was obtained from all patients or their families/legal guardians prior to tissue sample collection.

Protein expression levels and cellular localization of PKM2, HIF-1$\alpha$, and transforming growth factor-beta 1 (TGF-$\beta$1) in EP samples derived from endometriosis and non-endometriosis groups were measured using IHC analysis. Briefly, 4-µm-thick sections were prepared from paraffin-fixed samples and mounted on silane-coated glass slides. Rehydration was performed in serial dilutions of ethanol using a dewaxing reagent (#ST975; Beyotime, Shanghai, China). Endogenous peroxidase activity was quenched using 1% hydrogen peroxide (#7722-84-1; Sigma Aldrich, St Louis, MO, USA) for 15 min. After washing, the sections were incubated with primary antibodies against PKM2 (#bs-0101R-1; Bioss, Beijing, China), HIF-1$\alpha$ (#ab51608; Abcam, Cambridge, MA, USA), and TGF-$\beta$1 (#ab215715; Abcam, Cambridge, MA, USA) at 4 °C overnight. Horseradish-peroxidase-conjugated anti-rabbit IgG (#ab97051; Abcam, Cambridge, UK) was added and the samples were incubated for 25 min. After washing, the sections were incubated in 3,3-diaminobenzidine (#A690009; Sangon, Shanghai, China) in phosphate-buffered saline (PBS) for 10 min and counterstained with hematoxylin to allow visualization of the immune complexes. Images were obtained using a microscope (Olympus, Tokyo, Japan).

All slides were scored by an independent pathologist who was not informed of the sample characteristics, and the fields of view were randomly selected at a magnification of $\times$400. IHC results of PKM2, HIF-1$\alpha$, and TGF-$\beta$1 were quantified using the Allred score [17], which is the sum of the scores of the proportion of positive cells (score range, 0–5) and the response intensity (score range 0–3). The percentage of positive cells was scored as follows: absence of positive cells, 0; 1% positive cells, 1; 2–10% positive cells, 2; 11–30% positive cells, 3; 31–66% positive cells, 4; and 67–100% positive cells, 5. The staining intensity was defined as follows: 0, negative (no staining); 1, weakly positive (yellow); 2, moderately positive (brown-yellow); and 3, strongly positive (brown). Five fields of view were randomly selected for each sample and the results were analyzed using the semi-quantitative method. To avoid false positives or negatives, the primary antibody was replaced by PBS as a negative control, and tissues with known positive expression of PKM2, HIF-1$\alpha$, and TGF-$\beta$1 were used as positive controls. The EP tissues in each group were divided into a negative group (scores of 0 or 2) and a positive group (scores $\ge$3).

Fresh EP samples collected from patients with EP with or without endometriosis under sterile conditions were minced finely and digested enzymatically with 5 mg of collagenase I (500 µg/mL; #17100017; Gibco™, Thermo Fisher Scientific, Waltham, MA, USA) and 1 mg deoxyribonuclease type I (100 µg/mL; #10325ES80; Yeasen, Shanghai, China) for 1 h at 37 °C. After centrifugation, the cells were suspended in DMEM/F12 culture medium containing 1 nM estradiol (#IE0210; Solarbio, Beijing, China), 0.2% insulin (#P3376; Beyotime), 1% L-glutamine (#ST1441-25g; Beyotime), 1% antibiotic solution (#SNA-001; Sunncell, Wuhan, China), and 10% dextran-coated charcoal-treated fetal bovine serum (#SH30068.03; HyClone, Logan, UT, USA). The purity of the isolated ESCs was confirmed $>$95% (P2-P3) by Immunocytochemistry (ICC) staining using antibodies against vimentin (stromal cell marker) (#FNab09409; Finetest, Wuhan, China) and cytokeratin (epithelia cell marker) (#bs-1712R; Bioss). Non-endometriotic patient-derived ESCs were designated as control cells and named non-endometriotic patient-derived ESCs (NESCs). ESCs and NESCs from P3-P4 were used for subsequent experiments.

Small interfering RNAs (siRNAs) targeting PKM2 (si-PKM2#1, sense: GGAAAGAACAUCAAGAUAATT, antisense: UUAUCUUGAUGUUCUUUCCTT; si-PKM2#2, sense: GGAAUGAACGUGGCUCGUUTT, antisense: AACGAGCCACGUUCAUUCCTT; or si-PKM2#3, sense: GGGUGAACUUGGCCAUGAATT, antisense: UUCAUGGCCAAGUUCACCCTT) were utilized to interfere with PKM2 expression in ESCs, with si-NC (sense: UUCUCCGAACGUGUCACGUTT, antisense: ACGUGACACGUUCGGAGAATT) used as a control. A PKM2-overexpressing (PKM2-OE) plasmid was constructed by inserting the cDNA sequence of PKM2 (NM_002654.6) into pcDNA3.1, with an empty vector as a control. siRNAs (10 mM) were transfected into ESCs using Lipofectamine RNAiMAX (Invitrogen, Carlsbad, CA, USA). The PKM2-OE plasmid was transfected using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions.

TRIzol reagent (#15596026; Thermo Fisher Scientific) was used to extract total RNA from ESCs or NESCs. The purity and concentration of the RNA were verified by measuring the absorbance ratio at 260/280 nm. Subsequently, cDNA was generated using M-MLV reverse transcriptase (#AE101-03; TransGen Biotech, Beijing, China). qPCR was performed using iTaq Universal SYBR Green Supermix (#1725121; Bio-Rad, Hercules, CA, USA). PCR primers were synthesized by TsingKe (Beijing, China). PKM2 expression was analyzed using the forward primer, 5^′-ATGTCGAAGCCCCATAGTGAA-3^′, and reverse primer, 5^′-TGGGTGGTGAATCAATGTCCA-3^′ and TGF-$\beta$1 expression was analyzed using the forward primer, 5^′-TACCTGAACCCGTGTTGCTCTC-3^′ and the reverse primer, 5^′-GTTGCTGAGGTATCGCCAGGAA-3^′. The transcript levels of PKM2 were normalized to those of the housekeeping gene $\beta$-actin (actin), which was measured using the forward primer, 5^′-CACCATTGGCAATGAGCGGTTC-3^′ and reverse primer, 5^′-AGGTCTTTGCGGATGTCCACGT-3^′. The 2^{-Δ⁢Δ⁢Cq} method was applied to calculate the relative expression levels.

Cell samples were lysed in ice-cold RIPA buffer (#R0010; Solarbio) supplemented with a protease inhibitor cocktail (#C600386; Sangon). Protein levels were quantified using a Pierce BCA kit (#23225; Thermo Fisher Scientific). Equal amounts of protein (approximately 50 µg) were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene fluoride membranes (0.45 µm; #88585; Thermo Fisher Scientific). Following blocking in 5% fat-free milk, the membranes were incubated overnight at 4 °C with primary antibodies against PKM2 (#bs-0101R-1; Bioss), TGF-$\beta$1 (#ab215715; Abcam), or actin (#bs-0061R; Bioss). After incubation, the membranes were incubated with a horseradish-peroxidase-labeled goat anti-rabbit secondary antibody (#ab97051; Abcam). Protein bands were detected using Dura Extended Duration Substrate (#34075; Thermo Fisher Scientific). Band intensity was evaluated using ImageJ software (NIH, Bethesda, MD, USA; version 1.5) and normalized to the band intensity of actin.

ESCs/NESCs were plated on 96-well plates and incubated with or without an anti-TGF-$\beta$1 antibody (1 µg/mL) for 48 h. To each well, 250 µL of CCK-8 solution (#C0039; Beyotime) was added. Two hours later, the absorbance at 450 nm was recorded using a plate reader (Biotek, Winooski, VT, USA; Elx808).

ESCs/NESCs (1 $\times$ 10⁵) were seeded in 12-well plates and incubated until a subconfluent monolayer was formed. A sterile pipette tip (200 µL) was used to make a scratch-wound in the confluent monolayers. The cells were further cultured in Dulbecco’s modified Eagle medium (DMEM)/F12, with or without an anti-TGF-$\beta$1 antibody for 24 h. Images were obtained using an inverted microscope and analyzed using ImageJ software.

ESCs/NESCs (1 $\times$ 10⁵) suspended in serum-free DMEM/F12 were seeded into the upper chambers of 24-well Transwell plates pre-coated with Matrigel (#356234; Corning, New York, NY, USA) and diluted at a ratio of 1:3. To the lower chamber, 10% fetal-bovine-serum-supplemented DMEM/F12 (600 µL), with or without an anti-TGF-$\beta$1 antibody, was added. Non-invasive cells were removed from the upper chambers after 24 h of incubation, and the remaining cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet (#E607309; Sangon) for 30 min. Observation and photography were performed under an inverted microscope, and counting was performed using ImageJ software.

Transfected and non-transfected ESCs/NESCs (1 $\times$ 10⁵) were incubated for 24 h under different treatments, followed by collection of the culture media. Quantification of glucose and lactate levels in the cell culture medium was performed using a glucose assay kit (#GAGO20-1KT; Sigma Aldrich, St Louis, MO, USA) or a lactate assay kit (#ab65330; Abcam), respectively, according to the manufacturer’s instructions.

TGF-$\beta$1 levels in the supernatants of ESCs/NESCs were determined using an enzyme-linked immunosorbent assay (ELISA) kit (#PT880; Beyotime) according to the manufacturer’s instructions.

Cells coated on glass covers were allowed to grow overnight to prepare the slides, which were then fixed with 4% paraformaldehyde. After permeabilization with 0.2% Triton X-100 (#9002-93-1; Solarbio), the cells were blocked with 1% bovine serum albumin (#9048-46-8; Solarbio) and then incubated with a primary antibody against HIF-1$\alpha$ (#ab51608; Abcam) for 12 h at 4 °C. They were then washed with PBS and incubated with a fluorescently labeled secondary antibody (#ab150079; Abcam) for 1 h in the dark. Cell nuclei were stained with 4^′,6-diamidino-2-phenylindole (#E607303; Sangon) at a concentration of 1.43 µM (blue). The cells were observed and imaged using a fluorescence microscope (Olympus).

Transfected and non-transfected ESCs/NESCs were transiently transfected with the pGL3-HIF-1$\alpha$-promoter vector (0.5 µg) together with the Renilla luciferase plasmid phRL-TK (#E2231; Promega, Madison, WI, USA) using Fugene HD transfection reagent (#E2311; Promega) and incubated with or without an anti-TGF-$\beta$1 antibody. A double-luciferase reporter assay system (#E1910; Promega) was used to detect luciferase activity.

Data presented in this paper represent at least three independent experiments and are expressed as the mean $\pm$ standard deviation. Statistical analyses were performed using GraphPad Prism software (version 8.0; GraphPad, San Diego, CA, USA). The normality of the data was determined using the Shapiro-Wilk test. Comparisons between two groups were conducted using an unpaired Student’s t-test. Data from more than two groups were analyzed using one-way analysis of variance, followed by Tukey’s post-hoc test. The correlation of IHC scores among PKM2, HIF-1$\alpha$, and TGF-$\beta$1 in all EP samples was analyzed using Pearson’s correlation coefficient. p 0.05 was considered statistically significant.

To explain the relationship between PKM2, HIF-1$\alpha$, and TGF-$\beta$1 in patients with EPs and endometriosis, we performed IHC staining of EP samples from patients with EPs, with or without endometriosis (Fig. 1A). IHC staining identified significantly higher protein levels of PKM2 (p 0.0001), HIF-1$\alpha$ (p 0.0001), and TGF-$\beta$1 (p 0.0001) in EP samples combined with endometriosis (n = 23) than in EP samples without endometriosis (n = 18), with PKM2 and HIF-1$\alpha$ primarily localized in nucleus and cytoplasm, but TGF-$\beta$1 preferentially localized in the cytoplasm (Fig. 1A–D). Additionally, a positive correlation on IHC scores in all EP samples was observed between HIF-1$\alpha$ and PKM2 (r = 0.8951, p 0.0001), HIF-1$\alpha$ and TGF-$\beta$1 (r = 0.7563, p 0.0001), as well as PKM2 and TGF-$\beta$1 (r = 0.7859, p = 0.002) (Fig. 1E–G). These results suggested that PKM2, HIF-1$\alpha$, and TGF-$\beta$1 may be involved in the pathogenesis of EPs combined with endometriosis.

Fig. 1.

Pyruvate kinase M2 (PKM2) levels are positively correlated with hypoxia-inducible factor-1 alpha (HIF-1$\alpha$) and transforming growth factor-beta 1 (TGF-$\beta$1) levels in patients with endometrial polyp (EP). (A) Representative images ($\times$200 and $\times$400) of immunohistochemical (IHC) staining for PKM2, HIF-1$\alpha$, and TGF-$\beta$1 in EP samples from patients with EPs with or without endometriosis. Scale bars: 100 µm. (B–D) Scatterplots showing IHC scores for PKM2, HIF-1$\alpha$, and TGF-$\beta$1 in patients with EP with (n = 23) and without (n = 18) endometriosis (^*⁣**p 0.001; unpaired Student’s t-test). (E–G) Correlation analysis of PKM2, HIF-1$\alpha$, and TGF-$\beta$1 IHC scores in all EP samples (n = 23). Bars represent the mean $\pm$ standard deviation (SD).

To analyze PKM2 function, we isolated ESCs and NESCs from EP samples obtained from patients with and without endometriosis. The purity of the ESCs and NESCs was determined by ICC staining using anti-vimentin and anti-cytokeratin antibodies. As illustrated in Fig. 2A, the purity of ESCs and NESCs exceeded 95% after passaging for 2–3 generations. Subsequently, PKM2 expression at the transcriptional and translational levels were assessed. The data also show higher mRNA (p = 0.0021) and protein (p = 0.0388) levels of PKM2 in ESCs than in NESCs (Fig. 2B,C). Taken together, these results suggested that high PKM2 levels may be related to EPs combined with endometriosis.

Fig. 2.

High levels of PKM2 are observed in primary ESCs. (A) Representative images of ICC staining for vimentin and cytokeratin in primary endometrial stromal cells (ESCs) and non-endometrial patient-derived ESCs (NESCs). Scale bars: 100 µm. (B,C) Relative mRNA and protein levels of PKM2 in primary ESCs and NESCs were detected by reverse transcription (RT)-quantitative polymerase chain reaction (qPCR) and western blotting, respectively (n = 3; ^*p 0.05 and ^**p 0.01; unpaired Student’s t-test). All bars represent the mean $\pm$ SD.

Considering the up-regulation of PKM2 in ESCs, we investigated the function of PKM2 by interfering with PKM2 expression in ESCs using si-PKM2#1, si-PKM2#2, or si-PKM2#3. All three siRNAs repressed PKM2 at both transcriptional and protein levels, and si-PKM2#2 (p = 0.0002 and p 0.0001; si-PKM2#1, p = 0.0078 and p = 0.0427; si-PKM2#3, p = 0.0158 and p = 0.0017), which had the best knockdown efficiency, was selected for subsequent analyses (Fig. 3A,B). We observed higher viability in ESCs in comparison to NESCs (p 0.0001), but the viability of ESCs was impaired after PKM2 knockdown (p = 0.0002), as evidenced by CCK-8 assays (Fig. 3C). Furthermore, ESCs possessed stronger migratory (p 0.0001) and invasive (p = 0.0022) abilities than NESCs; however, PKM2 down-regulation reduced the migratory (p 0.0001) and invasive (p = 0.0027) abilities of ESCs (Fig. 3D,E). As an important regulator of glycolytic enzymes, PKM2 promotes lactate production and metabolic reprogramming. Therefore, we investigated the effect of PKM2 on glycolysis in ESCs. The results showed a striking increase in glucose uptake (p 0.0001) and lactate production (p 0.0001) in ESCs versus NESCs; however, these features of ESCs were undermined upon PKM2 knockdown (glucose uptake, p 0.0001; lactate production, p 0.0001) (Fig. 3F,G). Previous reports have demonstrated the regulatory role of PKM2 in TGF-$\beta$1 signaling [18, 19], we determined the effect of PKM2 on TGF-$\beta$1 in ESCs. As expected, a greater amount of TGF-$\beta$1 was released from ESCs than NESCs (p 0.0001), but the silencing of PKM2 reduced the release of TGF-$\beta$1 from ESCs (p 0.0001) (Fig. 3H). Consistently, TGF-$\beta$1 mRNA and protein levels were strongly elevated in ESCs compared with those in NESCs (p = 0.0001 and p = 0.0006), yet PKM2 silencing repressed TGF-$\beta$1 mRNA and protein levels in ESCs (p = 0.0002 and p = 0.0004) (Fig. 3I,J). Collectively, these results showed that PKM2-dependent glycolysis affects the proliferation, migration, invasion, and TGF-$\beta$1 secretion of ESCs.

Fig. 3.

PKM2-dependent glycolysis affects the proliferation, migration, and invasion of ESCs. (A,B) The PKM2-interference efficiency of si-PKM2#1, si-PKM2#, and si-PKM2#3 in ESCs was detected by reverse transcription (RT)-quantitative polymerase chain reaction (qPCR) and western blotting (n = 3; ns, p $>$ 0.05 vs. control, ^#p 0.05, ^#⁢#p 0.01, and ^#⁢#⁢#p 0.001 vs. the negative control small interfering RNAs (siRNAs) [si-NC]; one-way analysis of variance [ANOVA]). (C–E) The viability, migration, and invasion of NESCs, ESCs, and ESCs transfected with si-NC or si-PKM2 were determined by Cell Counting Kit-8 (CCK-8), wound-healing, and Transwell invasion assays (n = 3; ^**p 0.01 and ^*⁣**p 0.001 vs. NESCs, ns, p $>$ 0.05 vs. ESCs, and ^#⁢#p 0.01 and ^#⁢#⁢#p 0.001 vs. ESCs + si-NC; one-way ANOVA). Scale bars: 100 µm. (F,G) Glucose uptake and lactate production were measured in different subgroups of cells using respective kits (n = 3; ^*⁣**p 0.001 vs. NESCs, ns, p $>$ 0.05 vs. ESCs, and ^#⁢#⁢#p 0.001 vs. ESCs + si-NC; one-way ANOVA). (H) The amount of TGF-$\beta$1 released from different subgroups of cells was measured by enzyme-linked immunosorbent assay (ELISA; n = 3; ^*⁣**p 0.001 vs. NESCs, ns, p $>$ 0.05 vs. ESCs, and ^#⁢#⁢#p 0.001 vs. ESCs + si-NC; one-way ANOVA). (I,J) Relative mRNA and protein levels of TGF-$\beta$1 were determined by RT-qPCR and western blotting (n = 3; ^*⁣**p 0.001 vs. NESCs, ns, p $>$ 0.05 vs. ESCs, and ^#⁢#⁢#p 0.001 vs. ESCs + si-NC; one-way ANOVA). Bars represent the mean $\pm$ SD.

PKM2 has been reported to interact with HIF-1$\alpha$ and stimulate the HIF-1$\alpha$ transcriptional activation domain function. Therefore, we further determined whether PKM2 can mediate the transcriptional activity of HIF-1$\alpha$ in ESCs. IF staining showed higher levels of HIF-1$\alpha$ in ESCs relative to NESCs, whereas PKM2 down-regulation repressed HIF-1$\alpha$ expression in ESCs (Fig. 4A). To further validate this relationship, we constructed the luciferase reporter gene vector pGL3-HIF-1$\alpha$-promoter encompassing the promoter of the HIF-1$\alpha$ gene. Dual-luciferase reporter assays showed that the luciferase activity of the pGL3-HIF-1$\alpha$-promoter vector was stronger in ESCs than in NESCs (p = 0.0004) (Fig. 4B). However, the luciferase activity of the pGL3-HIF-1$\alpha$-promoter vector was repressed in ESCs co-transfected with si-PKM2 (p = 0.0027) (Fig. 4B). All results showed that PKM2 promoted the transcriptional activity of HIF-1$\alpha$ in ESCs.

Fig. 4.

PKM2 mediates the transcriptional activity of HIF-1$\alpha$. (A) Representative images of immunofluorescence staining for HIF-1$\alpha$ in NESCs, ESCs, and ESCs transfected with si-NC or si-PKM2. Scale bars: 100 µm. (B) Dual-luciferase reporter assays were used to determine the luciferase activity of the pGL3-HIF-1$\alpha$-promoter vector in NESCs, ESCs, and ESCs transfected with si-NC or si-PKM2 (n = 3; ^*⁣**p 0.001 vs. NESCs, ns, p $>$ 0.05 vs. ESCs, and ^#⁢#p 0.001 vs. ESCs + si-NC; one-way ANOVA). Bars represent the mean $\pm$ SD.

Since up-regulation of PKM2 promotes proliferation, migration and invasion of ESCs, we introduced PKM2 into NESCs to verify that PKM2 is a key factor in EPs combined with endometriosis. Transfection of the PKM2-OE vector strongly upregulated PKM2 mRNA and protein levels in NESCs (both p 0.0001) (Fig. 5A,B). PKM2 overexpression significantly increased TGF-$\beta$1 mRNA and protein levels (p 0.0001 and p = 0.0026) (Fig. 5C,D). Functional analyses demonstrated that PKM2 overexpression increased the viability, migration, and invasion of NESCs (all p 0.0001), but these effects were reversed by the introduction of an anti-TGF-$\beta$1 antibody (all p 0.0001) (Fig. 5E–G), Furthermore, the up-regulation of PKM2 markedly increased glucose uptake and lactate production by NESCs (both p 0.0001), but the addition of an anti-TGF-$\beta$1 antibody had no effect on PKM2-mediated increases in glucose uptake and lactate production (p = 0.9991 and p = 0.6620) (Fig. 5H,I). Importantly, PKM2 up-regulation enhanced the luciferase activity of the pGL3-HIF-1$\alpha$-promoter in NESCs (p 0.0001), but the addition of an anti-TGF-$\beta$1 antibody did not affect the luciferase activity of this promoter vector in NESCs overexpressing PKM2 (p = 0.1994), suggesting that TGF-$\beta$1 is a downstream effector molecule of the PKM2/HIF-1$\alpha$ axis in NESCs (Fig. 5J). Together, these data showed that PKM2/HIF-1$\alpha$ axis-dependent glycolysis contributes to NESC proliferation, migration, and invasion via TGF-$\beta$1.

Fig. 5.

The PKM2/HIF-1$\alpha$ axis enhances the viability, migration, and invasion of NESCs via TGF-$\beta$1. (A,B) The transfection efficiency of the PKM2- overexpression (OE) vector in NESCs was evaluated by RT-qPCR and western blotting (n = 3; ns, p $>$ 0.05 $>$ 0.05 vs. control and ^*⁣**p 0.001 vs. vector; one-way ANOVA). (C,D) Relative mRNA and protein levels of TGF-$\beta$1 in NESCs and NESCs transfected with empty vector or the PKM2-OE were determined by RT-qPCR and western blotting (n = 3; ns, p $>$ 0.05 0.05 vs. control and ^**p 0.01, ^*⁣**p 0.001 vs. vector; one-way ANOVA). (E–G) The viability, migration, and invasion of cells in different subgroups (ESCs, NESCs, NESCs + vector, NESCs + PKM2-OE, and NESCs + PKM2-OE + anti-TGF-$\beta$1 antibody [Ab]) were determined by CCK-8, wound healing, and Transwell invasion assays (scale bar = 100 µm) (n = 3; ns, p $>$ 0.05 and ^*⁣**p 0.001; one-way ANOVA). (H,I) Measurement of glucose uptake and lactate production in cells from different subgroups was performed using appropriate kits (n = 3; ns, p $>$ 0.05 and ^*⁣**p 0.001; one-way ANOVA). (J) Luciferase activity of the pGL3-HIF-1$\alpha$ promoter vector was determined using a dual-luciferase reporter assay in cells with different treatments (n = 3; ns, p $>$ 0.05 and ^*⁣**p 0.001; one-way ANOVA). Bars represent the mean $\pm$ SD.