兽研简要总结
慢性无症状性睾丸炎（CAO）是雄性犬不育最常见的原因，目前缺乏有效的治疗选择。前列腺素内过氧化物合酶2（PTGS2）在炎症中发挥重要作用，并与StAR共同参与睾酮的合成。为阐明PTGS2在CAO的发生、维持和进展中的作用，本研究在11例犬CAO样本中检测了睾丸PTGS2和StAR在基因和蛋白水平的表达，并与5例精子正常对照（CG）进行比较。评估了睾丸生精区和产激素区中PTGS2和StAR的免疫阳性面积百分比（PIA）及平均灰度值（MGS）。在CAO中，基因和蛋白水平的表达均显著升高。仅在产激素区中StAR的MGS无显著变化。此外，除PTGS2基因表达外，所有参数均显示生精损伤程度越高，表达变化越显著。本研究表明PTGS2表达升高与CAO之间存在关联。需要进一步研究以探讨选择性PTGS2抑制剂的潜在临床获益。
关键词：不育，犬，炎症，慢性无症状性睾丸炎，PTGS2，StAR
1. 引言
由无精子症（即精液中完全缺乏精子）引起的不育占不育雄性犬最常见临床发现的34.8%[1]。与人类相同[2]，犬无精子症主要与精子发生障碍有关[3]。在不育人类[4,5,6]和犬[7,8]患者的睾丸活检中常可检测到免疫细胞浸润，主要为淋巴细胞，近期由此形成了慢性无症状性睾丸炎的常见诊断[6,7]。
尽管慢性无症状性睾丸炎（CAO）的诱发原因仍不清楚，但在犬睾丸中已鉴定出早期和晚期两种形式，伴有不同程度的精子发生改变[7]。浆细胞和T细胞侵入睾丸实质并穿过受损的血睾屏障[9]，导致精子发生受损和持续性炎症[7,8,10]。这种持续进行的炎症很可能是目前CAO导致永久性不育且无任何可用治疗选择的原因，尽管精原干细胞仍然存在——至少在某些曲细精管中[11]。
前列腺素在炎症和疼痛中发挥重要作用[12,13]，而前列腺素内过氧化物合酶（PTGS）2作为关键酶[14,15,16]，已被证实在实验性诱导自身免疫性睾丸炎（EAO）大鼠睾丸[17]及精子发生紊乱的不育人类患者[18,19]中表达。在多种其他物种的健康睾丸组织中检测到PTGS2，如叙利亚仓鼠[20]、小鼠[21]、大鼠[22,23]以及犬[24,25]，这表明前列腺素不仅与炎症相关，还参与多种物种的睾丸发育、类固醇生成和精子发生[20,21,22,23]。具体而言，已证明PTGS2和前列腺素可调节类固醇生成急性调节蛋白（StAR）——类固醇生成的限速步骤[26,27]——在小鼠Leydig细胞培养中的转录[28]。小鼠和犬Leydig细胞中PTGS2与StAR的共表达进一步支持了这一假说[25,28,29,30]。除Leydig细胞外，叙利亚仓鼠[20]、小鼠[21]、大鼠[22]、犬[24,25]和人[18]睾丸中的Sertoli细胞及部分管周细胞也表达PTGS2。此外，在混合性萎缩的不育人类患者睾丸活检中，肥大细胞表达PTGS2[31]。此外，与精子发生正常的患者[31,33,34,35]相比，精子发生紊乱、生殖细胞停滞综合征、唯Sertoli细胞综合征和混合性萎缩的不育男性中肥大细胞数量显著增加[32]。肥大细胞可能不仅通过PTGS2驱动的炎症，还通过类胰蛋白酶[36]促进精子发生障碍[35]，类胰蛋白酶是一种诱导纤维化的丝氨酸蛋白酶[32,37]，而纤维化在精子发生障碍中发挥重要作用[38,39]。在健康犬睾丸中，肥大细胞极少被检测到[40]；然而，由于犬CAO与纤维化增加相关[7]，我们推测表达类胰蛋白酶的肥大细胞参与该病的发病机制。总之，PTGS2作为炎症的关键酶，被怀疑参与CAO的病理过程。其对类固醇生成的总体影响已在小鼠Leydig细胞体外实验中得到证实[28]，但尚未在CAO背景下进行研究。此外，PTGS2也由肥大细胞表达，因此我们旨在鉴定和表征这些细胞。
在此背景下，本研究旨在检测不同程度生精损伤的CAO犬睾丸活检中PTGS2的表达，以阐明PTGS2在CAO病理生理学中的作用。为确定与健康组织相比PTGS2表达升高是否与CAO样本中肥大细胞的存在相关，进行了类胰蛋白酶染色。为进一步明确PTGS2对类固醇生成的总体影响，还评估了StAR的表达。
2. 材料与方法
2.1. 研究设计与样本采集
本研究纳入十六只性成熟且临床健康的雄性犬：其中十一只患非梗阻性无精子症（经重复精液样本确认），因不育病史就诊于诊所接受进一步诊断。简言之，临床和男科检查包括精液分析、生殖道超声、精浆碱性磷酸酶测定、精液细菌学检查、犬布鲁氏菌血清学检测和内分泌分析，确诊为非梗阻性无精子症，无睾丸前性或感染性病因指征[7]。为诊断目的，在全身麻醉下以剪刀活检方式获取双侧睾丸活检组织。组织学评估显示所有相关犬（n = 11）均存在CAO[7,8]。为进一步区分，按先前描述[7]将生精成熟阻滞分为“早期阻滞”（n = 5），即生精过程主要停滞于精原细胞水平，以及“晚期阻滞”（n = 6），即存在进一步发育的生殖细胞阶段。CAO犬平均年龄为6.5 ± 1.7岁（范围：4.5–9.5岁，补充表S1），平均体重为15.7 ± 9.6 kg（范围：5.0–18.0 kg），属于九个不同品种（柯利犬：n = 3；凯恩梗、迷你贵宾犬、拉布拉多寻回犬、冰岛牧羊犬、威尔士柯基彭布罗克犬、图莱亚尔棉毛犬、比格犬和杰克罗素梗：各n = 1）。
其余五只犬作为健康对照组（CG）。从这些犬采集了所有参数（活力、存活率、形态学和总精子计数）均在参考范围内的正常精子精液样本[41]。CG犬（n = 5）平均年龄为4.5 ± 4.0岁（范围：0.9–9.9岁，补充表S1），平均体重为7.6 ± 3.1 kg（范围：3.6–11.5 kg）。这些动物属于四个不同品种（各n = 1：哈瓦那犬、马尔济斯犬、波士顿梗、吉娃娃）；一只犬为混种犬。CG的睾丸组织样本在手术去势后采集，该去势手术系应主人要求因非医学原因实施。
所有主人均同意将睾丸组织样本用于研究目的。由于仅使用剩余组织，无需伦理批准。右睾丸组织样本浸入RNAlater®（Ambion Biotechnologie，Wiesbaden，德国）并保存于−80 °C用于mRNA提取。双侧睾丸样本（CAO）或右睾丸样本（CG）按先前描述[42]固定于Bouin液并石蜡包埋用于组织学和免疫组化。同一批犬用于基因和蛋白表达分析。
2.2. RNA提取、逆转录和定量逆转录聚合酶链反应（RT-qPCR）
使用基于TRIzol的方案（Thermo Scientific，Waltham，MA，USA）从CAO犬（n = 11）和CG犬（n = 5）的右睾丸组织样本中分离总RNA。CAO组中一只犬因提取过程中RNA产量反复偏低而被排除于mRNA分析。使用200 ng/µL RNA和RevertAid First Strand cDNA Synthesis Kit（#K1622，Thermo Scientific）按制造商方案进行全长第一链cDNA合成，包括DNase处理和以无核酸酶水作为无模板对照。先前建立的PTGS2和StAR引物组以及两个参考基因GAPDH（甘油醛-3-磷酸脱氢酶）和ACTB（β-肌动蛋白）的引物组购自Microsynth（Balgach，瑞士；表1）。先前已通过扩增子测序确保引物在犬睾丸中的特异性结合[25,29]。对于RT-qPCR，将3 µL cDNA（1:10稀释）加入8 µL FastStart Essential DNA Green Master（Roche Diagnostics，Mannheim，德国）、各1 µL正向和反向引物（10 pmol）以及2 µL无核酸酶水中。RT-qPCR条件为95 °C 10 min，随后35个循环的95 °C 10 s、60 °C 10 s、72 °C 10 s，并加熔解曲线。每次检测均包含无模板对照。所有样本使用LightCycler®96实时PCR系统（软件版本1.5.0，版本1.5.0.39，Roche Diagnostics）进行三重复检测。引物效率使用由混合cDNA样本两倍稀释系列（1:2–1:128）三重复RT-qPCR运行得出的相对标准曲线计算。混合CG样本作为校准品纳入每批检测。为计算相对基因表达比率，按Pfaffl[43]进行效率校正的相对定量。由于GAPDH与ACTB相比mRNA表达变异最小，选择GAPDH作为计算用参考基因。
Table 1. Sequences of primers for RT-qPCR, amplicon length in base pairs (bp), efficiency, and accession number (for = forward, rev = reverse).
| Gene | Oligonucleotide Sequence (5′–3′) | Amplicon Length (bp) | Efficiency | Accession Number |
|---|---|---|---|---|
| GAPDH | for: GGCCAAGAGGGTCATCATCTC rev: GGGGCCGTCCACGGTCTTCT | 228 | 1.96 | NM_001003142.1 |
| ACTB | for: GCTGTGCTGTCCCTGTATG rev: GCGTACCCCTCATAGATGG | 98 | 1.92 | NM_001195845.3 |
| PTGS2 | for: GGAGCATAACAGAGTGTGTGATGTG rev: AAGTATTAGCCTGCTCGTCTGGAAT | 88 | 2.21 | NM_001003354 |
| StAR | for: CGAGGCTCCACCTGTGTGT rev: CCTTTCTGCTCAGGCATCTC | 65 | 1.94 | NM_001097542 |
2.3. 免疫组织化学及PTGS2、StAR和类胰蛋白酶染色的评估
PTGS2的免疫组织化学按先前所述方法进行，犬睾丸中特异性抗体结合通过Western blotting验证[25]。石蜡包埋组织的连续切片（2 µm）（CAO：n = 11只犬，左右样本；CG：n = 5，仅右侧样本）经脱蜡、在柠檬酸盐缓冲液（pH = 6）中煮沸、与3%过氧化氢甲醇溶液孵育，并与10%马血清孵育以封闭非特异性结合（S-2000，Vector Laboratories，Newark，CA，USA）。封闭在ICC缓冲液（1.2 g Na2HPO4、0.2 g KH2PO4、0.2 g KCl、8.0 g NaCl、3 mL Triton加至1 L）中进行，之后切片覆盖PTGS2单克隆抗体（Clone CX229，Cayman Chemicals，Ann Arbor，MI，USA，稀释1:100，相当于0.005 mg/mL），4 °C过夜。次日，样本与二抗（BA-2000，Vector Laboratories，稀释1:200）孵育。免疫阳性信号按制造商说明（VECTASTAIN PK-6100 Rabbit IgG Elite ABC Kit和Vector Nova-RED Substrate Kit SK-4800，Vector Laboratories）通过免疫过氧化物酶法显色。玻片用苏木精复染。以相应蛋白浓度的无关小鼠IgG抗体（I-2000，Mouse IgG，Vector Laboratories）作为同型对照。
StAR免疫染色参照先前描述并稍作修改进行，其中特异性抗体结合也经Western blotting确认[29]。使用磷酸盐缓冲液（PBS）洗涤，而封闭使用PBS缓冲液中的3%牛血清白蛋白（VWR Life Science，Solon，OH，USA）和10%山羊血清（S-1000，Vector Laboratories）。玻片与针对鼠StAR肽的多克隆抗体（稀释1:2000；由D.M. Stocco教授惠赠[29,44]）于4 °C孵育过夜。次日，玻片用PBST（含0.1% Tween 20的PBS）洗涤两次、PBS洗涤一次，随后与SuperVision-2 HRP Enhancer（DCS，Innovative Diagnostik Systeme，Hamburg，Germany）孵育。免疫阳性信号用DAB（DCS）显色。此处以无关兔IgG抗体（I-1000，Rabbit IgG，Vector Laboratories）作为同型对照。
类胰蛋白酶免疫染色按PTGS2所述方法进行，但使用类胰蛋白酶单克隆抗体（Clone AA1，Dako Denmark，Glostrup，Denmark，稀释1:100）。犬肥大细胞中特异性抗体结合此前已通过流式细胞术、免疫荧光和免疫细胞化学确认[45,46]。次日，按StAR所述方法（SuperVision-2 HRP Enhancer、HRP Polymer和DAB）依制造商说明进行制备。以犬肥大细胞瘤作为阳性对照。连续切片中类胰蛋白酶免疫阳性信号的评估为描述性，并与PTGS2染色进行比较，使用Olympus BX41TF显微镜（Shinjuku，Japan）在200×和400×放大倍数下观察。PTGS2、StAR和类胰蛋白酶的所有染色均在单一批次中完成以避免偏倚。包含仅使用缓冲液的阴性对照。
对于切片的计算机辅助图像分析，使用了ImageJ Fiji软件[47]（https://fiji.sc，于2026年7月10日访问，版本2.18.0）。为了分离和区分各个解剖隔室内的特异性表达变化，分别针对间质（PTGS2和StAR）和肾小管（PTGS2）评估了免疫阳性信号。由于StAR仅在间质隔室中表达，因此对间质组织（PTGS2和StAR）和肾小管（PTGS2）的免疫阳性面积（PIA）和染色强度（平均灰度值，MGS）进行了定量[40]。对于间质隔室，使用Olympus cellSens Dimension软件（版本2.1）在400倍放大（Olympus BX41TF）下随机选取20张图像（CAO中每侧10张）。类似地，对于肾小管隔室，在200倍放大下随机选取30个肾小管（CAO中右侧/左侧各15个）拍摄图像。根据感兴趣区域的不同，将包含管周细胞和Leydig细胞的间质组织或生精小管（基底膜以内的区域）设为感兴趣区域，并裁剪掉血管和伪影。所有图像均转换为灰度图像，像素灰度值范围为0（白色）至255（黑色）。分别针对间质组织和肾小管设定定义免疫阳性染色的各自阈值，PTGS2和StAR分别设定。为评估MGS，分析了所有高于阈值的阳性像素。PIA通过应用各自阈值对灰度图像进行二值化后测定[40]。
2.4. 统计分析
使用GraphPad Prism 7软件（GraphPad Software，美国加利福尼亚州拉霍亚）和Microsoft Excel（版本16.109.2，Microsoft，美国华盛顿州雷德蒙德）进行统计分析。p < 0.05的值被认为具有统计学显著性。
主要目的是确定CAO组与CG组之间在PTGS2和StAR mRNA及蛋白表达方面的显著差异。在后续分析中，CAO组被细分为早期和晚期生精停滞，并进行了组间比较。基于激素下调睾丸和对照睾丸中StAR基因表达数据[29]的功效分析，用于在单因素方差分析（ANOVA）中比较三组，显著性水平（alpha）= 0.05（双尾），目标功效= 0.8，标准化效应量= 1.207，每组共同标准差= 0.4，得出每组所需样本量为n = 4，达到的功效为0.889。
Shapiro–Wilk检验未检测到所有组中PTGS2的mRNA表达比值和经对数转换的StAR数据偏离正态分布（p < 0.3）。因此，CAO组数据与CG组使用Welch校正的非配对t检验进行比较（StAR采用log10转换数据）。将CAO组细分为早期和晚期停滞后，使用单因素方差分析（StAR采用log10转换数据）比较各组，随后进行经多重检验校正的Tukey多重比较检验。
通过客观的计算机辅助分析，检查了PTGS2（肾小管和间质）和StAR（间质）蛋白表达水平的PIA和MGS。为排除CAO组内的侧别特异性效应，首先比较了左侧和右侧睾丸。Shapiro–Wilk检验表明，CAO组左右睾丸在所有PTGS2数据集（PIA和MGS，肾小管和间质隔室）以及间质中StAR PIA方面均未偏离正态分布（p < 0.2）。随后的配对t检验显示两侧睾丸之间无显著差异。同样，Wilcoxon符号秩检验未检测到左右睾丸之间间质StAR MGS的显著差异（根据Shapiro–Wilk检验为非正态分布）。因此，对个体CAO犬左右睾丸的PIA和MGS值取平均值。
首先，将所有患有CAO的犬与CG进行比较。在Shapiro–Wilk检验表明未偏离正态分布（p < 0.2）后，这些配对比较使用非配对t检验（PTGS2的间质PIA和MGS均经log10转换后）。随后，CAO组进一步分为早期和晚期生精停滞。根据Shapiro–Wilk检验，所有组均呈正态分布（PTGS2和StAR的间质PIA经log10转换后）。因此，使用单因素方差分析及随后的经多重检验校正的Tukey事后检验进行组间比较。
3. 结果
3.1. PTGS2 mRNA和蛋白表达
经非配对t检验确定，CAO组的PTGS2 mRNA表达比值显著高于CG组（表2）。细分为早期和晚期停滞组后，单因素方差分析识别出显著的组间差异（p = 0.0022），与CG相比，早期停滞组（p = 0.0061）和晚期停滞组（p = 0.0036）的基因表达均较高（图1）。
Table 2. PTGS2 mRNA (ratios) and protein expression (percentage of immunopositive area, PIA, and mean gray scale, MGS) in testicular specimens collected from 11 dogs with chronic asymptomatic orchitis (CAO) and five normospermic dogs in the control group (CG). PTGS2 mRNA ratios and all MGS results are presented as arithmetic mean ± standard deviation, whereas all PIA values are displayed as geometric mean (geometric standard deviation factor). CAO was compared with CG using unpaired t -tests; log10 transformation was applied to PIA values prior to analysis, and Welch’s correction was used for mRNA ratios.
| Compartment | Analysis Level | CAO | CG | p -Value |
|---|---|---|---|---|
| mRNA (ratios) | 4.93 ± 1.72 | 1.15 ± 0.41 | <0.0001 | |
| tubuli | PIA | 1.05 (3.05) | 0.07 (3.33) | 0.0006 |
| MGS | 98.16 ± 0.76 | 97.36 ± 0.32 | 0.0413 | |
| interstitium | PIA | 1.89 (2.02) | 0.06 (2.56) | <0.0001 |
| MGS | 114.77 ± 0.86 | 113.07 ± 1.15 | 0.0051 |
[图：患有慢性无症状性睾丸炎（CAO）的犬的睾丸PTGS2表达，分为早期停滞型和晚期停滞型]
在所有样本的支持细胞和间质细胞胞质中均观察到PTGS2的特异性免疫阳性染色（图2）。CG组（图2c）显示免疫阳性信号在支持细胞胞质中均匀分布，从基底膜延伸至管腔。在CAO组（图2a、b）中，支持细胞中的PTGS2染色不规则，在整个样本中呈弥漫性分布。此外，两组中部分管周细胞也呈免疫阳性染色（图2）。
[图：患有慢性无症状性睾丸炎（CAO）的犬（a、b）和精子正常犬的睾丸样本中PTGS2的免疫染色]
与mRNA水平相似，CAO组样本中曲细精管内和间质区的PTGS2 PIA和MGS均高于CG组（表2）。同样，各亚组（早期停滞、晚期停滞、CG）在曲细精管（p = 0.0084）和间质区（p = 0.0007，均采用方差分析）的PTGS2 PIA方面存在显著差异，早期和晚期生精停滞组的PIA均高于CG组（图1b、c）。然而，在亚组比较中，仅早期停滞组间质区的MGS显著高于CG组（方差分析，p = 0.0091；图1e）。
3.2. StAR mRNA和蛋白表达
CAO组的StAR mRNA表达比值显著高于CG组（表3）。在比较两个生精停滞阶段时也观察到这一现象（方差分析，p = 0.0025；图3a），尽管晚期停滞组与CG组之间的比值无显著差异（图3a）。
Table 3. StAR mRNA (ratios) and protein expression (percentage of immunopositive area, PIA, and mean gray scale, MGS) in testicular specimens collected from dogs with chronic asymptomatic orchitis (CAO) and normospermic dogs in the control group (CG). MGS results are presented as arithmetic mean ± standard deviation, whereas StAR mRNA ratios and PIA values are displayed as geometric mean (geometric standard deviation factor). The CAO group was compared with the CG using unpaired t -tests; log10 transformation was applied to StAR mRNA ratios and PIA values prior to analysis, and Welch’s correction was used for mRNA ratios.
| Compartment | Analysis Level | CAO | CG | p -Value |
|---|---|---|---|---|
| mRNA (ratio) | 2.92 (2.87) | 0.78 (1.31) | 0.0034 | |
| interstitium | PIA | 1.79 (1.72) | 0.35 (2.92) | 0.0010 |
| MGS | 99.36 ± 0.60 | 99.68 ± 0.75 | 0.3770 |
[图：患有慢性无症状性睾丸炎（CAO）的犬的睾丸StAR表达，分为早期和晚期生精停滞型]
在CAO组和CG组样本中，均观察到StAR在间质细胞和部分管周细胞胞质中的特异性免疫阳性染色（图4）。计算机辅助客观评估显示，CAO组间质区的PIA显著高于CG组（p = 0.0010，表3）。相比之下，两组之间的MGS无差异。同样，当将早期和晚期停滞样本与CG组进行比较时，各组间PIA存在显著差异（方差分析，p = 0.0285），但MGS无显著差异。在Tukey事后检验中，仅早期停滞组的PIA显著高于CG组（图3b）。
[图：慢性无症状性睾丸炎（CAO）犬睾丸样本中StAR的免疫染色（a，b）以及来自正常]
3.3. 类胰蛋白酶蛋白表达
类胰蛋白酶的特异性免疫阳性染色仅在CAO样本中检测到，主要位于间质区室，而在CG中缺失（图5）。鉴定出两种类型的免疫阳性细胞：（1）具有典型圆形外观的肥大细胞，特征性地定位于间质区域；（2）更细长的肥大细胞，通常位于小管壁附近。总体而言，CAO组中类胰蛋白酶阳性肥大细胞的数量较低。对PTGS2和类胰蛋白酶染色的连续切片进行比较发现，大多数类胰蛋白酶阳性细胞同时对PTGS2呈免疫阳性染色。
[图：慢性无症状性睾丸炎（CAO）犬睾丸样本中类胰蛋白酶的免疫染色（a，b）以及来自]
4. 讨论
我们假设PTGS2参与犬CAO的病理生理过程。由于PTGS2是前列腺素合成的关键酶[14,15]，我们推断CAO中PTGS2可能较取自精子正常犬患者的正常健康睾丸上调。我们目前的数据明确支持我们的假设，因为CAO样本中PTGS2 mRNA表达显著高于健康犬睾丸组织。此外，PTGS2 PIA的改变在早期停滞型CAO中更为明显，表明其与生精停滞阶段相关。类胰蛋白酶染色证实PTGS2表达性肥大细胞仅存在于CAO组睾丸中。CAO组睾丸中StAR基因和蛋白表达水平未降低这一事实表明，用于类固醇激素合成的胆固醇供应并未减少。
尽管在评估mRNA表达水平时必须考虑到区室分布的改变，即间质区室显著增加和小管区室减少[10]，但EAO大鼠中报道的PTGS2表达增加与我们的结果高度一致[17]。同样，CAO样本中小管内以及间质区域的PTGS2-PIA均升高，支持PTGS2参与睾丸炎症改变的假设。关于细胞定位，在CAO组和正常健康睾丸中均鉴定出针对PTGS2的特异性免疫阳性信号，位于支持细胞、管周细胞和间质细胞中。这与先前对健康犬睾丸组织的描述一致[25]，也已在叙利亚仓鼠[20]、大鼠[22,30]和小鼠[21]的健康睾丸以及不育男性[18,48]中有过描述。此外，经类胰蛋白酶染色证实，大多数肥大细胞对PTGS2呈免疫阳性染色。然而，这些细胞仅在CAO患犬中鉴定到，而在健康对照中未发现，从而支持其在对生精功能受损中的作用，正如先前在不育男性中所显示的那样[31,33,34]。与人类病例中的发现[33,34]相似，鉴定出两种形态学上不同的组织型类胰蛋白酶阳性肥大细胞，即圆形和细长形肥大细胞，尽管其确切功能和作用迄今仍不明确。特别是，与生精功能正常的患者相比，不育男性患者中管周肥大细胞增加[38,39]，这与我们在CAO患犬中的发现相符。关于犬睾丸中肥大细胞的研究较为稀少，且结果仍略有矛盾[40,49]。具体而言，Anton等人1998年使用甲苯胺蓝和阿尔新蓝染色报告未发现肥大细胞[49]，而我们课题组先前使用甲苯胺蓝染色鉴定到少量肥大细胞[40]。由于肥大细胞“特征谱”，即蛋白酶表达水平，在不同解剖部位之间存在差异[50]，且可能受该部位病理状态（健康 versus 炎症）的影响，使用肥大细胞特异性抗体和传统染色进行的比较研究可能有助于进一步表征CAO中的犬睾丸肥大细胞。由于我们的研究在CAO中鉴定到表达PTGS2的肥大细胞，未来研究应进一步探讨其在CAO病理生理中的作用，并确定特异性PTGS2抑制剂是否能减少其存在或纤维化的进展。
除细胞定位外，我们还通过客观的计算机辅助分析定量蛋白表达，分别阐明了PTGS2在间质和管状区域的独立作用。基于细胞表达模式，可以得出结论：间质区域的表达主要代表Leydig细胞和管周细胞中的PTGS2表达，而管状区域的表达代表Sertoli细胞的PTGS2表达。与正常健康睾丸相比，CAO样本间质区域的PIA和MGS均显著升高，支持了CAO患犬因染色细胞数量增加和免疫阳性染色增强而导致PTGS2表达升高的假说。这一观察与PTGS2基因表达结果高度一致。在EAO大鼠中，PTGS2基因表达与疾病的严重程度和进展相关[17]；同样，PTGS2蛋白表达（PIA和MGS）与精子发生受损程度呈正相关。由于PTGS2与睾丸纤维化相关[32]，CAO中纤维化增加和基底膜增厚[7,10]可能与PTGS2表达升高有关，从而在精子发生障碍中发挥作用。然而，患者仅在CAO临床发病后才就诊以确认诊断。由于该疾病临床进展无症状、伦理考量以及犬主的不配合，无法对患病犬进行连续活检。这些活检对于确定PTGS2是否参与CAO的发生、维持和/或进展至关重要。
除PTGS2在炎症中的作用外，我们旨在进一步研究最近提出的PTGS2对犬睾丸类固醇生成的影响（或反之）[25]。在既往研究中，我们表明PTGS2[25]和StAR[29]的表达水平在移除GnRH激动剂植入物后类固醇生成重建期间最高。使用连续切片，我们发现CAO组样本中StAR和PTGS2共表达显著增加，在Leydig细胞和管周细胞中检测到免疫阳性信号。然而，这与小鼠Leydig细胞培养中观察到的PTGS2升高对StAR和类固醇生成的间接抑制作用相反[28]，该抑制作用在炎症期间和衰老过程中[30]出现，导致睾酮浓度降低。StAR的PIA升高尤其值得关注。由于Leydig细胞数量在青春期后或与CAO相关发生变化的可能性不大，这可能反而提示Leydig细胞大小的改变。然而，CAO组中StAR未被下调这一事实支持了以下结论：在犬CAO中，精子发生受损而类固醇生成未受损。此外，CAO患者的血液睾酮浓度未见显著改变。因此，只能得出结论：睾丸类固醇生成的第一步，即StAR作为转导体的一部分提供胆固醇，未被下调。要研究完整的类固醇生成通路，需要通过RNA测序、蛋白质组学和代谢组学进行分析[51]。
我们的研究存在若干局限性：首先，结果基于小样本量获得，需在进一步研究中验证。所研究的犬属于不同品种，这影响了体重并导致各队列间体重变异性较大。此外，CG犬略年轻于CAO组犬。然而，这未必是偏倚，因为的确切发病时间仍不清楚。由于CAO的无症状性质，该疾病最初是通过交配失败而被发现的。进一步研究应纳入年龄和品种匹配的对照。最后，我们未量化肥大细胞总数。因此，无法识别精子发生早期停滞与晚期停滞之间可能存在的差异。疾病过程中缺乏连续活检，无法确定PTGS2是否参与CAO的起始、进展或维持。
尽管如此，我们关于炎症相关PTGS2表达升高的发现表明，给予PTGS2抑制剂可能是抑制睾丸炎症的合适治疗途径。在健康睾丸中，关于非甾体抗炎药治疗对精子发生影响的报道结果相互矛盾[52,53,54]。具体而言，口服乙酰水杨酸或吲哚美辛以及皮下注射双氯芬酸治疗可增强小鼠和犬的精子发生[52,53]；然而，腹腔注射乙酰水杨酸会降低大鼠间质细胞功能和精子计数[54]。选择性PTGS2抑制剂也报道了类似的有争议的作用，短期尼美舒利治疗未产生负面影响，而长期治疗则出现核碎裂、线粒体空泡化和未成熟巨精子细胞[21]。每日给予塞来昔布治疗五周未对大鼠睾丸重量或雄激素水平产生不良影响，也未导致睾丸组织学改变[55]。然而，观察到的睾丸间质液增加导致推测：在存在既存生育力低下的情况下可能发生损伤，但在急性炎症性疾病中则不会[55]。尽管如此，我们的结果表明，短期给予选择性PTGS2抑制剂可能有利于减轻炎症并可能预防CAO进展。该方法的一个主要局限性在于，早期开始治疗需要在早期阶段甚至可能在CAO表现之前检测和诊断睾丸损伤，才能取得成功。目前，此类生物标志物尚缺乏。
5. 结论
总体而言，我们的研究证实了PTGS2对CAO的重要性，因为CAO患者的PTGS2基因和蛋白表达水平显著升高，支持其参与炎症过程。此外，仅在受CAO影响的睾丸中鉴定出共表达PTGS2和类胰蛋白酶肥大细胞。CAO中StAR表达未降低，确保了睾丸类固醇生成所需的胆固醇供应不受干扰，尽管完整的后续合成级联反应以及PTGS2对其的影响仍有待进一步阐明。由于这些结果提示选择性PTGS2抑制剂可能为CAO治疗提供一种治疗选择，因此迫切需要进一步研究其潜在益处。
缩写
本手稿中使用以下缩写：
| ACTB | β-Actin |
| ANOVA | Analysis of Variance |
| CAO | Chronic Asymptomatic Orchitis |
| CG | Control Group |
| EAO | Experimentally Induced Autoimmune Orchitis |
| GAPDH | Glyceraldehyde-3-Phosphate Dehydrogenase |
| MGS | Mean Gray Scale |
| bp | Base Pairs |
| PBS | Phosphate-Buffered Saline |
| PIA | Percentage of Immunopositive Area |
| PTGS2 | Prostaglandin Endoperoxide Synthase 2 |
| RT-qPCR | Reverse Transcription Quantitative Polymerase Chain Reaction |
| StAR | Steroidogenic Acute Regulatory Protein |
补充材料
以下支持信息可从以下网址下载：https://www.mdpi.com/article/10.3390/vetsci13101110/s1。表S1：患有慢性无症状性睾丸炎（CAO）的个体患者和精液样本正常且精子发生未受影响的对照组（CG）犬的基线特征（品种、年龄（岁）、体重（kg））。还列出了精子发生状态（正常、早期停滞或晚期停滞）以及免疫细胞浸润（局灶性或弥漫性）。有关CAO犬的更多详细信息见[7]。
[内容请查看原文]
作者贡献
概念化，S.G.-P.；方法学，J.M.、H.K.、E.-M.P.和S.G.-P.；形式分析，J.M.和S.G.-P.；研究，J.M.、H.K.、E.-M.P.和S.G.-P.；写作——初稿准备，J.M.和S.G.-P.；写作——审查和编辑，J.M.、H.K.、E.-M.P.和S.G.-P.；监督，S.G.-P.；资金获取，S.G.-P.。所有作者均已阅读并同意手稿的出版版本。
机构审查委员会声明
该动物研究免于伦理审查和批准，因为根据丹麦相关主管机构（丹麦动物实验监察局）的规定，仅出于诊断目的使用剩余样本进行的采样无需批准。
知情同意声明
已获得动物主人的书面知情同意，同意其动物的剩余样本参与后续研究。
数据可用性声明
本研究中呈现的数据可应通讯作者要求提供。由于隐私原因，未提供进一步的临床数据。
利益冲突
作者声明无利益冲突。
资助声明
J.M.获得了Gesellschaft für kynologische Forschung（德国犬学研究会）的奖学金。
脚注
[内容请查看原文]
参考文献
[内容请查看原文]
关联数据
补充材料
[内容请查看原文]
数据可用性声明
本研究中呈现的数据可应通讯作者要求提供。由于隐私原因，未提供进一步的临床数据。
查看英文原文▾
Simple Summary
Chronic asymptomatic orchitis (CAO) is the most common reason for infertility in male dogs and currently lacks effective treatment options. Prostaglandin endoperoxide synthase 2 (PTGS2) plays an important role in inflammation and, together with StAR, in the synthesis of testosterone. To elucidate the involvement of PTGS2 in the development, maintenance, and progression of CAO, testicular PTGS2 and StAR expression were examined at the gene and protein levels in 11 canine CAO samples and compared to five normospermic controls (CG). The percentage of the immunopositive area (PIA) for PTGS2 and StAR and the mean gray scale (MGS) were evaluated in the sperm- and hormone-producing compartments of the testes. In CAO, expression at both the gene and protein levels was significantly increased. Only the MGS of StAR was not significantly altered in the hormone-producing compartment. Furthermore, for all parameters except PTGS2 gene expression, higher degrees of spermatogenic impairment were accompanied by more prominent expression changes. This study demonstrates an association between increased PTGS2 expression and CAO. Further research is needed to investigate the potential clinical benefits of selective PTGS2 inhibitors.
Keywords: infertility, dog, inflammation, chronic asymptomatic orchitis, PTGS2, StAR
1. Introduction
Infertility due to azoospermia, the complete absence of spermatozoa in the ejaculate, represents, at 34.8%, the most frequent clinical finding in infertile male dogs [1]. As in humans [2], canine azoospermia is mainly related to a disruption of spermatogenesis [3]. Immune cell infiltrations, mainly of lymphocytes, are regularly detected in testicular biopsies from infertile human [4,5,6] and canine patients [7,8], recently resulting in the common diagnosis of chronic asymptomatic orchitis [6,7].
Although the cause for induction of chronic asymptomatic orchitis (CAO) still remains unknown, in the canine testis, an early and a late form with varying degrees of altered spermatogenesis have already been identified [7]. Plasma cells and T cells invade the testicular parenchyma and cross the disrupted blood testicular barrier [9], leading to impaired spermatogenesis and sustained inflammation [7,8,10]. This ongoing inflammation is likely the reason why currently CAO results in permanent infertility without any available treatment options, although spermatogonial stem cells remain present—at least in some tubules [11].
Prostaglandins play an important role in inflammation and pain [12,13], and the prostaglandin endoperoxide synthase (PTGS) 2, the key enzyme [14,15,16], was shown to be expressed in rat testis with experimentally induced autoimmune orchitis (EAO) [17] and infertile human patients with disturbed spermatogenesis [18,19]. The detection of PTGS2 in the healthy testicular tissue of various other species, such as Syrian hamster [20], mouse [21], rat [22,23], as well as dog [24,25], indicates that prostaglandins are, however, not only associated with inflammation but also involved in testicular development, steroidogenesis, and spermatogenesis in various species [20,21,22,23]. Namely, PTGS2 and prostaglandins have been shown to regulate the transcription of the steroidogenic acute regulatory protein (StAR), the bottleneck of steroidogenesis [26,27], in murine Leydig cell cultures [28]. This hypothesis is strengthened by the co-expression of PTGS2 and StAR in murine and canine Leydig cells [25,28,29,30]. Beyond Leydig cells, Sertoli cells and some peritubular cells express PTGS2 in the Syrian hamster [20], mouse [21], rat [22], dog [24,25], and human [18] testis. In addition, mast cells expressed PTGS2 in testicular biopsies of infertile human patients with mixed atrophy [31]. Further, the number of mast cells was significantly increased in infertile men with disturbed spermatogenesis, germ cell arrest syndrome, Sertoli-cell-only-syndrome, and mixed atrophy [32] compared to patients with normal spermatogenesis [31,33,34,35]. Mast cells likely contribute to disruption of spermatogenesis [35] not only through PTGS2-driven inflammation but also via tryptase [36], a serine protease that induces fibrosis [32,37], which plays an important role in spermatogenesis disorders [38,39]. In healthy canine testis, mast cells are rarely detected [40]; however, as canine CAO is associated with increased fibrosis [7], we postulate that tryptase-expressing mast cells are involved in the pathogenesis of the disease. In summary, PTGS2 as the key enzyme of inflammation is suspected to be involved in CAO pathology. Its general influence on steroidogenesis has been demonstrated [28] in vitro in mouse Leydig cells but has not yet been investigated in the context of CAO. Furthermore, PTGS2 is also expressed by mast cells, which is why we aimed to identify and characterize these cells.
Against this background, the purpose of this study was to investigate PTGS2 expression in CAO-affected canine testicular biopsies with varying degrees of spermatogenic impairment to elucidate the role of PTGS2 in CAO pathophysiology. To determine whether increased PTGS2 expression compared to healthy tissues is associated with the presence of mast cells in CAO samples, tryptase staining was conducted. To further specify the overall influence of PTGS2 on steroidogenesis, StAR expression was also evaluated.
2. Materials and Methods
2.1. Study Design and Sample Collection
Sixteen sexually mature clinically healthy male dogs were included in this study: Eleven dogs suffered from non-obstructive azoospermia (confirmed by repeated semen samples) and were presented with a history of infertility at the clinic for further diagnostics. Briefly, clinical and andrological examinations including semen analysis and ultrasound of the genital tract, determination of alkaline phosphatase in seminal plasma, bacteriological examination of the ejaculate, serological testing for Brucella canis, and endocrine analysis confirmed the diagnosis of non-obstructive azoospermia with no indication of a pretesticular or infectious origin [7]. Testicular biopsies of both testes were obtained for diagnostic purposes under general anesthesia as scissors biopsies. Histological assessment revealed CAO in all respective dogs (n = 11) [7,8]. For further differentiation, the spermatogenic maturation arrest was characterized as previously described [7], namely as “early arrest” (n = 5), in which spermatogenesis is arrested mainly at the spermatogonial level, and as “late arrest” (n = 6), with further developed germ cell stages present. Dogs with CAO had a mean age of 6.5 ± 1.7 years (range: 4.5–9.5 years, Supplementary Table S1), a mean body weight of 15.7 ± 9.6 kg (range: 5.0–18.0 kg) and belonged to nine different breeds (Collie: n = 3; Cairn Terrier, Miniature Poodle, Labrador Retriever, Icelandic Sheepdog, Welsh Corgi Pembroke, Coton de Tulear, Beagle, and Jack Russel Terrier: n = 1 each).
The remaining five dogs were considered the healthy control group (CG). Normospermic semen samples with all parameters (motility, viability, morphology, and total sperm count) within the reference range [41] had been collected from these dogs. Dogs in the CG (n = 5) were a mean of 4.5 ± 4.0 years old (range: 0.9–9.9 years, Supplementary Table S1) and had a mean body weight of 7.6 ± 3.1 kg (range: 3.6–11.5 kg). The animals belonged to four different breeds (n = 1 each: Havanese, Maltese, Boston Terrier, Chihuahua); one dog was mixed breed. Testicular tissue samples from the CG had been collected after surgical castration, which was performed at the owner’s request due to non-medical reasons.
All owners agreed to the use of testicular tissue samples for research purposes. As only leftover tissue was used, no ethics approval was needed. Tissue samples of the right testis were immersed in RNAlater® (Ambion Biotechnologie, Wiesbaden, Germany) and stored at −80 °C for mRNA extraction. Samples of both testes (CAO) or of the right testis (CG) were fixed in Bouin’s solution and embedded in paraffin for histology and immunohistochemistry as previously described [42]. The same dogs were used for gene and protein expression analyses.
2.2. RNA Extraction, Reverse Transcription and Quantitative Reverse Transcriptase Polymerase Chain Reaction (RT-qPCR)
Total RNA was isolated from the right testicular tissue samples of dogs with CAO (n = 11) and CG dogs (n = 5) using a TRIzol-based protocol (Thermo Scientific, Waltham, MA, USA). One dog from the CAO group was excluded from the mRNA analysis due to repeatedly low RNA yields during extraction. Full-length first-strand cDNA synthesis was performed by using 200 ng/µL RNA and the RevertAid First Strand cDNA Synthesis Kit (#K1622, Thermo Scientific), according to the manufacturer’s protocol including DNase treatment and nuclease-free water as a non-template control. Previously established primer sets for PTGS2 and StAR, as well as for the two reference genes GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and ACTB (β-actin), were purchased from Microsynth (Balgach, Switzerland; Table 1). Specific binding of the primers in the canine testis was previously ensured by sequencing the amplicons [25,29]. For RT-qPCR, 3 µL of cDNA (dilution 1:10) was added to 8 µL of FastStart Essential DNA Green Master (Roche Diagnostics, Mannheim, Germany), 1 µL each of the forward and reverse primers (10 pmol), and 2 µL of nuclease-free water. The RT-qPCR conditions were 95 °C for 10 min, followed by 35 cycles of 95 °C for 10 s, 60 °C for 10 s, 72 °C for 10 s, and a melting curve. Non-template controls were included in every assay. All samples were run in triplicate using a LightCycler®96 real-time PCR system (Software release 1.5.0, Version 1.5.0.39, Roche Diagnostics). Primer efficiencies were calculated using a relative standard curve derived from a triplicate RT-qPCR run of a two-fold dilution series (1:2–1:128) of pooled cDNA samples. Pooled CG samples were included in each batch as a calibrator. For calculation of the relative gene expression ratio, an efficiency-corrected relative quantification according to Pfaffl [43] was performed. As GAPDH showed the slightest variation in mRNA expression compared to ACTB, GAPDH was chosen as the reference gene for calculations.
Table 1. Sequences of primers for RT-qPCR, amplicon length in base pairs (bp), efficiency, and accession number (for = forward, rev = reverse).
| Gene | Oligonucleotide Sequence (5′–3′) | Amplicon Length (bp) | Efficiency | Accession Number |
|---|---|---|---|---|
| GAPDH | for: GGCCAAGAGGGTCATCATCTC rev: GGGGCCGTCCACGGTCTTCT | 228 | 1.96 | NM_001003142.1 |
| ACTB | for: GCTGTGCTGTCCCTGTATG rev: GCGTACCCCTCATAGATGG | 98 | 1.92 | NM_001195845.3 |
| PTGS2 | for: GGAGCATAACAGAGTGTGTGATGTG rev: AAGTATTAGCCTGCTCGTCTGGAAT | 88 | 2.21 | NM_001003354 |
| StAR | for: CGAGGCTCCACCTGTGTGT rev: CCTTTCTGCTCAGGCATCTC | 65 | 1.94 | NM_001097542 |
2.3. Immunohistochemistry and Evaluation of PTGS2, StAR and Tryptase Staining
Immunohistochemistry for PTGS2 was performed as previously described, and specific antibody binding in the canine testis was verified by Western blotting [25]. Serial sections (2 µm) from paraffin-embedded tissues (CAO: n = 11 dogs, left and right samples; CG: n = 5, only right sample) were deparaffinized, boiled in citrate buffer (pH = 6), incubated with 3% hydrogen peroxide in methanol, and incubated with 10% horse serum to block nonspecific binding (S-2000, Vector Laboratories, Newark, CA, USA). Blocking was performed in ICC buffer (1.2 g Na2HPO4, 0.2 g KH2PO4, 0.2 g KCl, 8.0 g NaCl, 3 mL Triton ad 1 L), after which, sections were covered with the PTGS2 monoclonal antibody (Clone CX229, Cayman Chemicals, Ann Arbor, MI, USA, dilution 1:100 corresponding to 0.005 mg/mL) overnight at 4 °C. On the following day, samples were incubated with the secondary antibody (BA-2000, Vector Laboratories, dilution 1:200). Immunopositive signals were visualized by an immunoperoxidase method according to the manufacturer’s instructions (VECTASTAIN PK-6100 Rabbit IgG Elite ABC Kit and Vector Nova-RED Substrate Kit SK-4800, Vector Laboratories). Slides were counterstained with hematoxylin. An irrelevant mouse IgG antibody (I-2000, Mouse IgG, Vector Laboratories) at the respective protein concentration served as the isotype control.
StAR immunostaining was performed with slight modifications similar to earlier descriptions, in which specific antibody binding was also confirmed by Western blotting [29]. Phosphate-buffered saline (PBS) was used for washing, while 3% bovine serum albumin (VWR Life Science, Solon, OH, USA) and 10% goat serum (S-1000, Vector Laboratories) in PBS buffer were used for blocking. The slides were incubated with a polyclonal antibody against a murine StAR peptide (dilution 1:2000; kindly donated by Prof. D.M. Stocco [29,44]) overnight at 4 °C. On the following day, slides were washed twice with PBST (PBS with 0.1% Tween 20) and once with PBS followed by incubation with SuperVision-2 HRP Enhancer (DCS, Innovative Diagnostik Systeme, Hamburg, Germany). Immunopositive signals were visualized with DAB (DCS). Here, an irrelevant rabbit IgG antibody (I-1000, Rabbit IgG, Vector Laboratories) served as the isotype control.
Tryptase immunostaining was conducted as described for PTGS2, but using a tryptase monoclonal antibody (Clone AA1, Dako Denmark, Glostrup, Denmark, dilution 1:100). Specific antibody binding in canine mast cells had previously been confirmed by flow cytometry, immunofluorescence, and immunocytochemistry [45,46]. On the following day, preparation was performed as described for StAR (SuperVision-2 HRP Enhancer, HRP Polymer and DAB), according to the manufacturer’s instructions. A canine mast cell tumor served as the positive control. Evaluation of immunopositive signals for tryptase in serial sections was descriptive and comparative to PTGS2 staining using the Olympus BX41TF microscope (Shinjuku, Japan) at 200× and 400× magnification. All stainings for PTGS2, StAR, and tryptase, were performed in single batches to avoid bias. Negative controls using buffer only were included.
For computer-assisted image analysis of the slides, the ImageJ Fiji software [47] (https://fiji.sc, accessed on 10 July 2026, version 2.18.0) was used. To separate and differentiate specific expression changes within individual compartments, immunopositive signals were evaluated specifically for the interstitium (PTGS2 and StAR) and the tubules (PTGS2). Since StAR was only expressed in the interstitial compartment, the immunopositive area (PIA) and the staining intensity (mean gray scale, MGS) of the interstitial tissue (PTGS2 and StAR) and tubules (PTGS2) were quantified [40]. For the interstitial compartment, 20 randomly chosen images (10 for each side in CAO) were taken at 400× magnification (Olympus BX41TF) using the Olympus cellSens Dimension software (version 2.1). Similarly, for the tubular compartment, images from 30 randomly selected tubules (15 right/left each in CAO) were taken at 200× magnification. Depending on the area of interest, either the interstitial tissue including peritubular cells and Leydig cells, or the seminiferous tubules (area within the basement membrane) were set as the region of interest, and blood vessels and artifacts were cropped. All images were converted to grayscale pictures, with pixel gray values ranging from zero (white) to 255 (black). Individual thresholds defining immunopositive staining were set for the interstitial tissue and tubules, separately for PTGS2 and StAR. To evaluate the MGS, all positive pixels above the threshold were analyzed. The PIA was determined by binarized grayscale images applying the individual thresholds [40].
2.4. Statistical Analysis
GraphPad Prism 7 software (GraphPad Software, La Jolla, CA, USA) and Microsoft Excel (Version 16.109.2, Microsoft, Redmond, WA, USA) were used for statistical analysis. Values at a level of p < 0.05 were considered statistically significant.
The primary objective was to identify significant differences between the CAO and CG groups regarding PTGS2 and StAR mRNA and protein expression. In a subsequent analysis, the CAO group was subdivided into early and late spermatogenic arrest, and groupwise comparisons were conducted. A power analysis based on gene expression data of StAR in hormonally downregulated and control testes [29] for comparing three groups in a one-way analysis of variance (ANOVA) with a significance level (alpha) = 0.05 (two-tailed), a target power = 0.8, a standardized effect size = 1.207, and a common standard deviation for each group = 0.4 resulted in a required sample size of n = 4 per group with an achieved power of 0.889.
The Shapiro–Wilk test detected no deviation from a normal distribution (p < 0.3) for the mRNA expression ratios of PTGS2 and log-transformed StAR data for all groups. Consequently, CAO group data were compared with the CG using an unpaired t-test with Welch’s correction (utilizing log10-transformed data in the case of StAR). After subdividing the CAO group into early and late arrest, groups were compared using a one-way ANOVA, (with log10-transformed data for StAR), followed by Tukey’s multiple comparisons test adjusted for multiple testing.
PTGS2 (tubular and interstitial) and StAR (interstitial) protein expression levels were examined concerning PIA and MGS via objective computer-assisted analysis. To rule out side-specific effects within the CAO group, left and right testes were initially compared. No deviations from normal distribution (p < 0.2) were indicated by the Shapiro–Wilk test for the left and right testes with the CAO group across all PTGS2 data sets (PIA and MGS, tubular and interstitial compartments), as well as for StAR PIA in the interstitium. Subsequent paired t-tests revealed no significant differences between both testes. Likewise, the Wilcoxon signed-rank test did not detect a significant difference in interstitial StAR MGS between the left and right testes (not normally distributed according to the Shapiro–Wilk test). Therefore, the PIA and MGS values of the left and right testes of the individual CAO dogs were averaged.
Initially, all dogs with CAO were compared to the CG. These pairwise comparisons were conducted using unpaired t-tests (after log10 transformation of both interstitial PIA and MGS for PTGS2) after the Shapiro–Wilk test had indicated no deviation from normal distribution (p < 0.2). Subsequently, the CAO group was further differentiated into early and late arrest of spermatogenesis. According to the Shapiro–Wilk test, all groups were normally distributed (after log10 transformation of interstitial PIA for both PTGS2 and StAR). Therefore, groupwise comparisons were conducted using a one-way ANOVA with a subsequent Tukey’s post hoc test adjusted for multiple testing.
3. Results
3.1. PTGS2 mRNA and Protein Expression
The PTGS2 mRNA expression ratio was significantly higher in the CAO group than in the CG, as determined by the unpaired t-test (Table 2). After subdivision into the early- and late-arrest groups, a one-way ANOVA identified significant group differences (p = 0.0022), with higher gene expression in both the early-arrest (p = 0.0061) and late-arrest groups (p = 0.0036) compared with the CG (Figure 1).
Table 2. PTGS2 mRNA (ratios) and protein expression (percentage of immunopositive area, PIA, and mean gray scale, MGS) in testicular specimens collected from 11 dogs with chronic asymptomatic orchitis (CAO) and five normospermic dogs in the control group (CG). PTGS2 mRNA ratios and all MGS results are presented as arithmetic mean ± standard deviation, whereas all PIA values are displayed as geometric mean (geometric standard deviation factor). CAO was compared with CG using unpaired t -tests; log10 transformation was applied to PIA values prior to analysis, and Welch’s correction was used for mRNA ratios.
| Compartment | Analysis Level | CAO | CG | p -Value |
|---|---|---|---|---|
| mRNA (ratios) | 4.93 ± 1.72 | 1.15 ± 0.41 | <0.0001 | |
| tubuli | PIA | 1.05 (3.05) | 0.07 (3.33) | 0.0006 |
| MGS | 98.16 ± 0.76 | 97.36 ± 0.32 | 0.0413 | |
| interstitium | PIA | 1.89 (2.02) | 0.06 (2.56) | <0.0001 |
| MGS | 114.77 ± 0.86 | 113.07 ± 1.15 | 0.0051 |
[图：Testicular PTGS2 expression in dogs with chronic asymptomatic orchitis (CAO), divided into early-arrest and late-arrest ]
Specific immunopositive staining for PTGS2 was observed in the cytoplasm of Sertoli cells and Leydig cells in all samples (Figure 2). The CG (Figure 2c) showed a uniform distribution of the immunopositive signal in the cytoplasm of Sertoli cells extending from the basement membrane to the lumen. In the CAO group (Figure 2a,b), the PTGS2 staining in Sertoli cells was irregular and diffuse within the whole sample. Additionally, some peritubular cells stained immunopositive in both groups (Figure 2).
[图：Immunostaining for PTGS2 in testicular samples from dogs with chronic asymptomatic orchitis (CAO) ( a , b ) and from nor]
Similar to the mRNA levels, PTGS2 PIA and MGS within the tubules and in the interstitial compartment were larger in the CAO group samples compared to CG (Table 2). Likewise, the subgroups (early arrest, late arrest, CG), differed significantly regarding PTGS2 PIA in both the tubular (p = 0.0084) and the interstitial compartment (p = 0.0007, both by ANOVA), with early and late spermatogenic arrest showing a higher PIA compared with CG (Figure 1b,c). When comparing the subgroups, however, MGS was significantly higher only in the early-arrest group compared to the CG in the interstitial compartment (ANOVA, p = 0.0091; Figure 1e).
3.2. StAR mRNA and Protein Expression
The StAR mRNA expression ratio was significantly higher in the CAO group compared with the CG (Table 3). This was also apparent when comparing both stages of spermatogenic arrest (ANOVA, p = 0.0025; Figure 3a), although the ratio did not differ significantly between the late-arrest group and the CG (Figure 3a).
Table 3. StAR mRNA (ratios) and protein expression (percentage of immunopositive area, PIA, and mean gray scale, MGS) in testicular specimens collected from dogs with chronic asymptomatic orchitis (CAO) and normospermic dogs in the control group (CG). MGS results are presented as arithmetic mean ± standard deviation, whereas StAR mRNA ratios and PIA values are displayed as geometric mean (geometric standard deviation factor). The CAO group was compared with the CG using unpaired t -tests; log10 transformation was applied to StAR mRNA ratios and PIA values prior to analysis, and Welch’s correction was used for mRNA ratios.
| Compartment | Analysis Level | CAO | CG | p -Value |
|---|---|---|---|---|
| mRNA (ratio) | 2.92 (2.87) | 0.78 (1.31) | 0.0034 | |
| interstitium | PIA | 1.79 (1.72) | 0.35 (2.92) | 0.0010 |
| MGS | 99.36 ± 0.60 | 99.68 ± 0.75 | 0.3770 |
[图：Testicular StAR expression in dogs with chronic asymptomatic orchitis (CAO), divided into early- and late-arrest spermat]
Specific immunopositive staining for StAR was observed in the cytoplasm of Leydig cells and some peritubular cells in both the CAO and CG samples (Figure 4). Computer-assisted objective evaluation revealed a significantly higher PIA in the interstitial compartment in the CAO group compared to the CG (p = 0.0010, Table 3). In contrast, the MGS did not differ between the groups. Similarly, when comparing early- and late-arrest samples to the CG, a significant difference between groups was identified for PIA (ANOVA, p = 0.0285), but not for MGS. In Tukey’s post hoc test, PIA was significantly higher only in the early-arrest group compared with the CG (Figure 3b).
[图：Immunostaining for StAR in testicular samples from dogs with chronic asymptomatic orchitis (CAO) ( a , b ) and from norm]
3.3. Tryptase Protein Expression
Specific immunopositive staining for tryptase was only detected in the CAO samples, predominantly in the interstitial compartment, but was absent in the CG (Figure 5). Two types of immunopositive cells were identified: (1) mast cells with a typical rounded appearance, characteristically localized in the interstitial area, and (2) more elongated mast cells, typically located close to the tubular wall. Overall, the number of tryptase positive mast cells was low in the CAO group. Comparison of PTGS2- and tryptase-stained serial sections revealed that the majority of tryptase positive cells also stained immunopositive for PTGS2.
[图：Immunostaining for tryptase in testicular samples from dogs with chronic asymptomatic orchitis (CAO) ( a , b ) and from ]
4. Discussion
We hypothesized that PTGS2 is involved in the pathophysiology of CAO in dogs. Since PTGS2 is the key enzyme in prostaglandin synthesis [14,15], we reasoned that PTGS2 might be upregulated in CAO compared to normal healthy testes obtained from normospermic canine patients. Our present data clearly support our hypothesis, as PTGS2 mRNA expression was significantly higher in the CAO samples compared to the healthy canine testicular tissues. Moreover, the alteration in PTGS2 PIA was more pronounced in early-arrest CAO, indicating a correlation with the stage of spermatogenic arrest. Tryptase staining confirmed the presence of PTGS2-expressing mast cells solely in CAO group testes. The fact that StAR gene and protein expression levels were not decreased in CAO group testes indicates that the supply of cholesterol for steroid hormone synthesis is not reduced.
Even though alterations in the compartment distribution with a significantly increased interstitial and a decreased tubular compartment must be taken into consideration [10] when evaluating the mRNA expression levels, the increased PTGS2 expression reported in rats with EAO is in good agreement with our results [17]. Similarly, the PTGS2-PIA was increased within the tubuli as well as in the interstitial area of CAO samples, supporting the hypothesis that PTGS2 is involved in testicular inflammatory changes. Regarding the cellular localization, a specific immunopositive signal against PTGS2 was identified in Sertoli, peritubular, and Leydig cells in both the CAO group and normal healthy testes. This is consistent with previous descriptions of healthy canine testicular tissue [25] and has also been described for healthy testes of Syrian hamsters [20], rats [22,30], and mice [21], as well as for infertile men [18,48]. Additionally, as confirmed by tryptase staining, most mast cells stained immunopositive for PTGS2. However, these cells were only identified in CAO-affected dogs, not in healthy controls, thus supporting their role in impaired spermatogenesis, as previously shown in infertile men [31,33,34]. Similar to the findings in the human cases [33,34], two morphologically different tissue types of tryptase-positive mast cells were identified, round and elongated mast cells, though their exact function and role remain unspecified up until now. In particular, peritubular mast cells were increased in infertile human patients compared to those with normal spermatogenesis [38,39], which matches the findings in our CAO-affected dogs. Studies on mast cells in the canine testis are sparse, and the results remain slightly conflicting [40,49]. Specifically, Anton et al. 1998 reported an absence of mast cells using toluidine blue and alcian blue staining [49], whereas our group previously identified sparse numbers using toluidine blue staining [40]. As mast cell “profiles”, namely protease expression levels, differ between anatomical sites [50] and potentially the pathological state of the site (healthy versus inflamed), comparative studies using mast-cell-specific antibodies and traditional stains might be suitable to further characterize canine testicular mast cells in CAO. As our study identified PTGS2-expressing mast cells in CAO, future studies should further investigate their role in CAO pathophysiology and determine whether specific PTGS2 inhibitors might reduce their presence or the progression of fibrosis.
Beyond cellular localization, we also elucidated the role of PTGS2 separately in the interstitial and tubular compartments by quantifying protein expression via objective computer-assisted analysis. Due to the cellular expression patterns, it can be concluded that expression in the interstitial compartment mainly represents PTGS2 expression in Leydig and peritubular cells, whereas expression in the tubular compartment represents Sertoli cell PTGS2 expression. Both PIA and MGS were significantly higher in the interstitial compartment in CAO samples compared to normal healthy testes, supporting the hypothesis of a higher PTGS2 expression due to an increased number of stained cells and stronger immunopositive staining in CAO-affected dogs. This observation is in good agreement with the PTGS2 gene expression results. In EAO rats, PTGS2 gene expression was correlated with the severity and progression of the disease [17]; likewise, PTGS2 protein expression (PIA and MGS) was positively associated with the degree of spermatogenic impairment. As PTGS2 is associated with testicular fibrosis [32], increased fibrosis and thickening of the basement membrane in CAO [7,10] might be related to the increased PTGS2 expression, thereby playing a role in disruption of spermatogenesis. Nevertheless, patients are only presented for confirmation of CAO after its clinical onset. Due to the clinically asymptomatic progression of the disease, ethical concerns, as well as the reluctance of owners, serial biopsies of affected dogs were not accessible. These would be essential to determine whether PTGS2 is involved in the development, maintenance, and/or progression of CAO.
Beyond the role of PTGS2 in inflammation, we aimed to further investigate the recently postulated effect of PTGS2 on steroidogenesis (or vice versa) in the canine testis [25]. In previous studies, we showed that PTGS2 [25] and StAR [29] expression levels were highest during the re-establishment of steroidogenesis following removal of the GnRH agonist implant. Using serial sections, we identified a significantly increased StAR and PTGS2 co-expression in CAO group samples, with immunopositive signals detected in Leydig and peritubular cells. This is, however, in contrast to the indirect inhibitory effect of increased PTGS2 on StAR and steroidogenesis as observed in murine Leydig cell cultures [28] during inflammation and during aging [30], which results in low testosterone concentrations. The increased PIA for StAR is especially interesting. As it seems unlikely that the number of Leydig cells changes postpubertally or in relation to CAO, it might instead indicate a change in Leydig cell size. Nevertheless, the fact that StAR was not downregulated in the CAO group supports the conclusion that spermatogenesis, but not steroidogenesis, is impaired in CAO in dogs. Furthermore, blood testosterone concentrations are not significantly altered in patients with CAO. Consequently, it can only be concluded that the first step of testicular steroidogenesis, the provision of cholesterol by StAR as part of the transduceosome, is not downregulated. To investigate the complete steroidogenic pathway, analysis via RNA-sequencing, proteomics, and metabolomics is needed [51].
Our study has several limitations: First, the results were obtained using small sample sizes and need to be verified in further studies. The investigated dogs belonged to various breeds, which influenced body weight and led to a high variability in body weight across cohorts. Additionally, the CG dogs were slightly younger than the CAO group dogs. This is, however, not necessarily a bias, as the exact onset of the disease remains unknown. Due to the asymptomatic nature of CAO, the disease was first recognized through unsuccessful matings. Age- and breed-matched controls should be included in further studies. Lastly, we did not quantify the total number of mast cells. Therefore, possible differences between early- and late-arrest of spermatogenesis could not be identified. The lack of serial biopsies during the course of the disease prevents the determination of whether PTGS2 is involved in the initiation, progression, or maintenance of CAO.
Nevertheless, our findings regarding increased PTGS2 expression associated with inflammation indicate that the administration of PTGS2 inhibitors could be a suitable therapeutic approach to suppress the testicular inflammation. In the healthy testis, contradictory results concerning the effects of non-steroidal anti-inflammatory drug treatment on spermatogenesis have been reported [52,53,54]. Specifically, oral acetylsalicylic acid or indomethacin as well as subcutaneous diclofenac treatment enhanced murine and canine spermatogenesis [52,53]; however, intraperitoneal acetylsalicylic acid decreased rat Leydig cell function and sperm counts [54]. Similar controversial effects of selective PTGS2 inhibitors have been described with short-term nimesulide treatment exerting no negative effects, whereas long-term treatment revealed nuclear fragmentation, vacuolated mitochondria and immature giant spermatids [21]. Daily celecoxib treatment over five weeks did not adversely affect testis weight or androgen levels, nor did it result in histological changes in the testes of rats [55]. The observed increase in interstitial testicular fluid, however, led to the assumption that damage could occur in case of pre-existing subfertility, though not in acute inflammatory diseases [55]. Nevertheless, our results indicate that short-term administration of selective PTGS2 inhibitors may be beneficial for reducing inflammation and potentially preventing CAO progression. A major limitation of this approach is that an early start of treatment requires the detection and diagnosis of the testicular insult at an early stage, possibly even before the manifestation of CAO, in order to be successful. Currently, such biomarkers are lacking.
5. Conclusions
Overall, our study confirms the significance of PTGS2 for CAO, as the PTGS2 gene and protein expression levels were significantly higher in patients with CAO, supporting its involvement in inflammation. Furthermore, mast cells coexpressing PTGS2 and tryptase were identified solely in CAO-affected testes. StAR expression was not reduced in CAO, ensuring an undisturbed supply of cholesterol for testicular steroidogenesis, although the complete subsequent synthetic cascade and the influence of PTGS2 thereon remain to be further elucidated. As these results implicate that selective PTGS2 inhibitors might offer a therapeutic option for CAO treatment, further studies on their potential benefits are urgently needed.
Abbreviations
The following abbreviations are used in this manuscript:
| ACTB | β-Actin |
| ANOVA | Analysis of Variance |
| CAO | Chronic Asymptomatic Orchitis |
| CG | Control Group |
| EAO | Experimentally Induced Autoimmune Orchitis |
| GAPDH | Glyceraldehyde-3-Phosphate Dehydrogenase |
| MGS | Mean Gray Scale |
| bp | Base Pairs |
| PBS | Phosphate-Buffered Saline |
| PIA | Percentage of Immunopositive Area |
| PTGS2 | Prostaglandin Endoperoxide Synthase 2 |
| RT-qPCR | Reverse Transcription Quantitative Polymerase Chain Reaction |
| StAR | Steroidogenic Acute Regulatory Protein |
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/vetsci13101110/s1. Table S1: Baseline characteristics (breed, age in years, body weight in kg) of the individual patients suffering from chronic asymptomatic orchitis (CAO) and control group (CG) dogs with normospermic semen samples and unaffected spermatogenesis. The status of spermatogenesis (normal, early arrest, or late arrest) as well as immune cell infiltration (focal or generalized) are also presented. Additional details about the dogs with CAO are provided in [7].
[内容请查看原文]
Author Contributions
Conceptualization, S.G.-P.; methodology, J.M., H.K., E.-M.P., and S.G.-P.; formal analysis, J.M. and S.G.-P.; investigation, J.M., H.K., E.-M.P., and S.G.-P.; writing—original draft preparation, J.M. and S.G.-P.; writing—review and editing, J.M., H.K., E.-M.P., and S.G.-P.; supervision, S.G.-P.; funding acquisition, S.G.-P. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement
Ethical review and approval was waived for the animal study because, according to the respective Danish authority (Danish Animal Experiments Inspectorate), no approval is necessary for sampling conducted solely for diagnostic purposes using only left-over samples.
Informed Consent Statement
Written informed consent was obtained from the owners for the participation of their animal’s left-over samples in follow-up studies.
Data Availability Statement
The data presented in this study are available on request from the corresponding author. Further clinical data are not presented due to privacy reasons.
Conflicts of Interest
The authors declare no conflicts of interest.
Funding Statement
J.M. received a scholarship from the Gesellschaft für kynologische Forschung (Society for Cynological Research, Germany).
Footnotes
[内容请查看原文]
References
[内容请查看原文]
Associated Data
Supplementary Materials
[内容请查看原文]
Data Availability Statement
The data presented in this study are available on request from the corresponding author. Further clinical data are not presented due to privacy reasons.