Intestinal nematode infection confers a benefit to a non-declining frog species, while a fungal parasitic infection has sublethal impacts on reproductive investment

Emerging infectious disease is a major cause of wildlife decline around the world. Understanding the impacts of disease even in non-declining populations is important for understanding population-level health and resilience to other emerging threats. In this study, we explored the sublethal impacts of the fungal pathogen Batrachochytrium dendrobatidis, Bd, and a novel intestinal nematode on the non-declining Australian frog, Litoria lesueuri. We collected male animals that were either infected with the fungal parasite, infected with a nematode parasite, infected with both parasites, or uninfected and brought them into the lab, and monitored their morphology and fungal pathogen load over a 7-week trial. At the end of the experiment, we dissected the animals, collected the testes, and identified their nematode prevalence and burden. We morphologically and molecularly characterised the intestinal nematode as belonging to the genus Parathelandros within the order Oxyuridae. We found that this Parathelandros sp. infection was beneficial to the adult frogs, where infected animals were large, and had larger forearm width (an important trait for mating) when we accounted for animal size. The exact mechanisms of this improved condition are unknown and require additional research. However, in the Bd infected animals, we found an overall negative impact of infection, including reduced forearm width and sperm production. Bd infection is prevalent in this species, and there are tangible sublethal impacts of Bd infection, indicating that this species is affected even if mortality due to disease is low.

Methods
Field collection and husbandry

Male L. lesueuri (n=32) were collected from the wild at two sites in Lerderderg State Park (O’Briens Crossing, elevation 453m, -37.496026, 144.360978; Mackenzies Flat, elevation 33m, -37.615696, 144.424809) in the austral summer of December 2022. Litoria lesueuri are sexually dimorphic, and males are distinguished by their characteristic yellow breeding colouration and their smaller size. The animals were not treated for Bd infection on arrival in the lab, and any naturally acquired infection from the wild was allowed to develop during the experiment. 

Once captured in the field, individuals were maintained individually for biosecurity to ensure pathogens were not spread. Frogs were housed in individual tanks (230× 150×130 mm) on a gravel and moss substrate. Animals were maintained in the lab at an average temperature of 18°C, with a range between 15 and 23°C. Enclosures were flushed daily with filtered water, and frogs were fed small crickets ad libitum twice per week. All animals were weighed (digital scale to the nearest 0.01g), measured (snout-to-vent length, SVL, using dial callipers to the nearest 0.1mm), and swabbed for Bd infection (See methodology below) in the two days following their collection from the wild, and measurements were then made weekly except for week six. The width of the right forearm and the width of the nuptial pad were measured to the nearest 0.01mm using digital callipers. We took three repeat readings per frog for the nuptial pad and arm width. 

Any animals showing signs of moderate to severe chytridiomycosis (loss of righting reflex, severe erythema of groin and venter, excess skin slough) were euthanised with an overdose of MS222 (Tricaine methanesulfonate, Sigma-Aldrich), and decapitation was used as a secondary kill procedure. The experiment ended after 7 weeks, when all remaining animals were euthanised. Testes were removed, and the length of both testes was measured using digital callipers (0.01mm). The gonads were placed in formalin for histological processing. The whole specimen was thwaslaced in buffered formalin. 

Testing for Bd infection

We standardised the swabbing procedure by performing five repeat swab strokes on the middle of the venter, side of the venter, each thigh, and limb of each frog. The swab was gently rotated with each stroke to capture the greatest amount of fungal DNA from the skin. Swabs were stored at -20°C until processing. The DNA from the skin swabs was extracted in 50µL Prepman Ultra (Applied Biosystems®, Life Technologies Pty Ltd) and 30–40mg of 0.5mm silica beads (Biospec). We homogenised the samples (using a cell homogeniser) for 2min at 1400 oscillations per sec, then incubated the samples at 95°C to lyse the cells for 10min and collected and diluted the supernatant 6:100 in ultra-pure water before directly analysing for pathogen presence and quantity using qPCR (Brannelly et al. 2020). The remaining extracted DNA was stored at -20°C. With every extraction performed, one B. dendrobatidis positive control sample (zoospores from culture) and one negative control (swab only) were extracted.

The diluted DNA extract samples were analysed using standard qPCR (Rotogene, Qiagen) methodology (Boyle et al. 2004) to identify and quantify Bd DNA in the samples with minor modifications. We ran our reactions at 15µL volume with lo-ROX 2x mastermix (SensiFast, Bioline), including BSA to reduce PCR inhibition, plus 5 µL of template DNA per reaction well. We ran each sample in singlicate for 40 cycles (Roto-Gene Q 2.3.5 software). In each qPCR reaction plate, we included a set of seven standards of known Bd concentrations (Pisces Molecular) made using plasmid DNA of the Bd ITS region to confirm the Bd infection status and load from the skin swab samples. On every qPCR reaction plate, we included a no-template control (5µL of ultra-pure water to replace the template DNA). 

We calculated infection load as Bd ITS DNA copies present on the whole swab. We considered the sample to be positive if >2 ITS DNA copies were present in the reaction well. Animals were considered infected if they had at least one positive swab sample throughout the experiment. Animals were considered cleared of infection if they had at least two consecutive negative swab results and remained negative to the end of the experiment. In some cases, animals came in with a negative swab result, but over the weeks in the lab they r, they returned positive swab samples (n=8). We considered these individuals as positive for infection because they likely came into the experiment with a low/undetectable infection load, which gradually grew over time in captivity. While qPCR as a diagnostic tool is sensitive, low infection loads are often missed (Brannelly et al. 2020; Hollanders and Royle 2022). We believe these animals had a low infection when they entered the facility, rather than contamination, because our biosecurity protocols were rigorous and individuals were maintained in individual enclosures.

Testes preparation and morphology 

There is a developing understanding of how reproductive effort is being influenced by chytridiomycosis in amphibians, evaluated via proxies for gametogenesis such as gonad size and morphology, as well as secondary sex characteristics including colour and calling behaviour. Testis size in amphibians is closely tied to reproductive success, with an increase in size typically corresponding to greater sperm production (McCallum and Trauth 2007; Brannelly et al. 2016, 2021), where increases in seminiferous tubule size (or reduced tubule density) indicate greater reproductive output – potentially due to an increase in intra-tubule space accommodating more mature spermatozoa (McCallum and Trauth 2007; Brannelly et al. 2016, 2021). Germinal epithelium depth of seminiferous tubules has been used to indicate changes in reproductive effort in amphibians, where increased germinal epithelium depth correlates with increased effort by increasing the space for spermatogenesis to occur (McCallum and Trauth 2007; Brannelly et al. 2016, 2021). During the testes and sperm staging procedures, we were blinded to the experimental pathogen exposure status of the sample. 

The formalin-fixed left testis (n=30) was analysed for testis morphology and spermatogenesis. Testes were dehydrated in a graded series of ethanol, cleared with xylene, and embedded in paraffin. Two sequential 5µm transections of tissue were taken from six random locations (total of 12 histosections) along the length of the testis between the pole and centre. The sections were mounted on glass slides, stained with haematoxylin and eosin stain, and cover-slipped. Sequential histosections (one histosection randomly chosen from the pairs of sequential histosections) were examined for testis morphology for each animal. Photographs were taken of the whole histosection to calculate the histosection area and count the number of seminiferous tubules at 40 magnification. In each histosection analysed, the two largest seminiferous tubules were identified, and a photograph at 200 magnification, where tubule area was measured, and b, the maximum and minimum germinal epithelium depth was measured within each of the largest tubules. The photos were analysed using ImageJ.

To assess spermatogenesis activity within the testis, we analysed the number of cell clusters of each sperm stage within the testis. We chose four non-sequential histosections, and within each of those four histosections, we identified the two largest seminiferous tubules. At 400 magnification, we took photos that included the germinal epithelium of each of the two largest seminiferous tubules. We then identified the number of cell clusters within that field of view that belonged to each of the four distinct stages of spermatogenesis: spermatogonia, spermatocytes, spermatid, and spermatozoa (Brannelly et al. 2016, 2021). 

Parasitic nematode collection

The presence of gastrointestinal nematodes within the colony was identified during initial dissection. During dissection, nematode samples were taken from two male frogs, and these samples were preserved in 95% ethanol and stored at -20°C. Approximately 6 months after euthanasia (n= 32), dissection and fixation in formalin, we dissected each frog to remove the lungs and intestine. The lungs were placed on a glass slide and observed under a light microscope (Leica LED2000) to examine for the presence of lungworm. No lungworm was found in any of the samples. We then removed the large intestine and analysed the faeces for intestinal nematodes. We removed the faeces by making an incision at the distal end of the large intestine and applying pressure to compress the large intestine. We made one or two faecalsmears, depending on the volume of faeces removed from the frog. The faecal smears were observed under a light microscope for the presence of adult female nematodes. If adult nematodes were found, they were counted and photographed. Because the frogs were fixed in formalin, only animals with adult nematodes were considered infected with intestinal nematodes. While nematode eggs can be identified using a faecal floatation test, this is not possible once samples have been fixed in formalin. However, because we analysed all the faeces within the intestine, we can be confident that if an adult female nematode was present, then we detected it.

新发传染病是全球野生动物种群衰退的主要诱因。即便在未出现种群衰退的群体中，解析疾病的影响，对于理解种群整体健康水平以及其应对其他新发威胁的恢复力而言，均具有重要意义。本研究以种群未出现衰退的澳大利亚本土蛙类——莱氏雨滨蛙（Litoria lesueuri）为研究对象，探究了真菌病原体蛙壶菌（Batrachochytrium dendrobatidis，简称Bd）以及一种新型肠道线虫对其产生的亚致死效应。我们采集了四类雄性个体：仅感染真菌寄生虫、仅感染线虫寄生虫、同时感染两种寄生虫以及未感染任何寄生虫的个体，将其带回实验室，并在为期7周的实验周期内监测其形态学特征与真菌病原体载量。实验结束后，我们对所有个体实施解剖，采集其睾丸组织，并统计线虫的感染率与感染强度。我们通过形态学与分子生物学手段对该肠道线虫进行鉴定，确认其隶属于尖尾目（Oxyuridae）下的副特兰德鲁斯线虫属（Parathelandros）。研究发现，该副特兰德鲁斯线虫属未定种（Parathelandros sp.）的感染对成年蛙类具有益处：在校正个体体型的前提下，感染线虫的个体体型更大，前臂宽度（交配过程中的关键性状）也更宽。这种机体状态改善的确切机制目前尚不明确，有待进一步研究探明。然而，在感染Bd的个体中，我们观察到感染带来的整体负面影响，包括前臂宽度减小与精子生成量下降。该蛙类中Bd感染十分普遍，且Bd感染存在明确的亚致死效应，这表明即便因该疾病导致的死亡率较低，该物种仍会受到感染的影响。

## 材料与方法
### 野外采集与饲养管理
本研究于2022年12月的南半球夏季，在莱德德格州立公园（Lerderderg State Park）的两个采样点——奥布莱恩渡口（O’Briens Crossing，海拔453米，坐标-37.496026, 144.360978）与麦肯齐平原（Mackenzies Flat，海拔33米，坐标-37.615696, 144.424809），共采集了32只雄性莱氏雨滨蛙。莱氏雨滨蛙存在性别二态性：雄性个体具有标志性的繁殖期黄色体色，且体型更小。个体被带回实验室后，未针对Bd感染进行任何处理，野外自然获得的感染将在实验期间持续发展。野外捕获后，所有个体均单独饲养以保障生物安全，防止病原体扩散。蛙类饲养于单个饲养箱（230×150×130 mm）内，箱底铺设碎石与苔藓基质。实验室饲养环境的平均温度为18℃，温度波动范围为15℃至23℃。饲养箱每日用过滤水冲洗一次，每周投喂两次小型蟋蟀，确保个体可以自由取食。个体从野外被捕获后的两天内，我们使用精度达0.01g的电子秤称量其体重，使用数显游标卡尺测量其吻肛长（snout-to-vent length，SVL，精度达0.1mm），并通过棉拭子采样检测Bd感染情况（详见下述方法）；此后除第6周外，每周均进行一次上述测量。使用数显游标卡尺测量右前臂宽度与婚垫宽度，精度达0.01mm；每只蛙的婚垫与前臂宽度均测量三次。若个体表现出中度至重度壶菌病症状（如失去翻正反射、腹股沟与腹侧严重红斑、皮肤大量脱落），则使用过量MS222（间氨基苯甲酸乙酯甲磺酸盐，Tricaine methanesulfonate，Sigma-Aldrich品牌）实施安乐死，并以断头作为辅助致死手段。实验持续7周后结束，对所有存活个体实施安乐死。摘除其睾丸，使用数显游标卡尺（精度0.01mm）测量双侧睾丸的长度。将性腺置于福尔马林中固定，用于后续组织学处理；整个标本均置于缓冲福尔马林中保存。

### Bd感染检测
我们统一了棉拭子采样流程：对每只蛙的腹侧中部、腹侧两侧、每一侧大腿与肢体各擦拭五次，每次擦拭时轻轻转动拭子，以尽可能多地采集皮肤表面的真菌DNA。采集后的拭子置于-20℃冰箱保存，直至后续处理。使用50μL Prepman Ultra试剂（Applied Biosystems®，Life Technologies Pty Ltd）与30~40mg粒径0.5mm的硅胶微珠（Biospec公司）提取皮肤拭子中的DNA。使用细胞匀浆器以1400次/秒的振荡频率对样本匀浆2分钟，随后将样本置于95℃下孵育10分钟以裂解细胞；收集上清液，用超纯水按6:100的比例稀释后，通过实时荧光定量PCR（qPCR）直接检测病原体的存在与载量（Brannelly等，2020）。剩余的提取DNA置于-20℃冰箱保存。每批次提取过程中，均同步设置阳性对照样本（来自培养物的蛙壶菌游动孢子）与阴性对照样本（仅采集拭子，未接触蛙类皮肤）。稀释后的DNA提取物采用经过小幅优化的标准qPCR方法（Rotogene仪器，Qiagen公司；Boyle等，2004）进行分析，以鉴定并定量样本中的Bd DNA。反应体系总体积为15μL，包含lo-ROX 2×预混液（SensiFast，Bioline公司）、用于降低PCR抑制效应的牛血清白蛋白（BSA），以及每孔5μL的模板DNA。每个样本均单次反应，扩增循环数为40次（使用Roto-Gene Q 2.3.5软件）。每块qPCR反应板均设置7个已知浓度的Bd标准品（Pisces Molecular公司），标准品基于Bd的ITS区域质粒DNA制备，用于校准皮肤拭子样本的Bd感染状态与载量。每块反应板均设置无模板对照（以5μL超纯水替代模板DNA）。我们以整个拭子上的Bd ITS DNA拷贝数作为感染载量的计算依据。若反应孔中ITS DNA拷贝数大于2，则判定样本为阳性。若个体在整个实验过程中至少有一次拭子检测结果为阳性，则判定为感染个体；若个体至少连续两次拭子检测结果为阴性，且直至实验结束均保持阴性，则判定为已清除感染。部分个体在捕获时的拭子检测结果为阴性，但在实验室饲养数周后转为阳性（共8只）。我们将这类个体判定为感染个体，因为它们极有可能在进入实验时携带低载量或无法被检测到的Bd感染，在人工饲养环境下随时间推移逐渐增殖至可检测水平。尽管qPCR作为诊断工具具有较高灵敏度，但低载量感染常难以被检测到（Brannelly等，2020；Hollanders与Royle，2022）。由于本研究的生物安全规程十分严格，且所有个体均单独饲养，因此我们认为这些个体在进入实验室时已存在低载量感染，而非交叉污染导致的结果。

### 睾丸制备与形态学分析
目前学界对壶菌病如何影响两栖动物繁殖投入的认知正逐步深化，相关研究常通过性腺大小与形态等配子发生替代指标，以及体色、鸣叫行为等第二性征进行评估。两栖动物的睾丸大小与繁殖成功率密切相关：睾丸体积增大通常意味着精子生成量提升（McCallum与Trauth，2007；Brannelly等，2016、2021）。曲细精管体积增大（或曲细精管密度降低）则代表繁殖输出增加，这可能是因为曲细精管内部空间增大，可容纳更多成熟精子（McCallum与Trauth，2007；Brannelly等，2016、2021）。曲细精管的生精上皮厚度常被用于指示两栖动物繁殖投入的变化：生精上皮厚度增加可扩大精子发生的空间，从而提升繁殖投入（McCallum与Trauth，2007；Brannelly等，2016、2021）。在睾丸处理与精子分期的操作过程中，我们对样本的实验病原体暴露状态不知情，以避免偏倚。我们对30只个体的福尔马林固定左侧睾丸进行分析，以评估睾丸形态与精子发生情况。睾丸样本经梯度乙醇脱水、二甲苯透明后，置于石蜡中包埋。从睾丸的极部到中部，随机选取6个位置，每个位置切取两张连续的5μm厚组织切片，共获得12张组织学切片。将切片裱于载玻片上，使用苏木精-伊红染色后封片。针对每只蛙，选取一组连续切片中的一张进行睾丸形态学观察。在40倍物镜下拍摄整张组织切片的照片，用于计算切片面积并计数曲细精管的数量。在每张被分析的组织切片中，选取两条最大的曲细精管，在200倍物镜下拍摄照片，用于测量曲细精管的面积，并在每条最大曲细精管内测量生精上皮的最大与最小厚度。使用ImageJ软件对照片进行分析。为评估睾丸内的精子发生活性，我们对睾丸内各精子发生阶段的细胞簇数量进行统计。我们选取4张非连续的组织切片，在每张切片中选取两条最大的曲细精管。在400倍物镜下拍摄包含两条最大曲细精管生精上皮的照片，随后在该视野内计数四个不同精子发生阶段——精原细胞、精母细胞、精子细胞与精子——的细胞簇数量（Brannelly等，2016、2021）。

### 寄生线虫采集
在首次解剖过程中，我们发现实验种群中存在胃肠道线虫。解剖期间，我们从两只雄性蛙体内采集了线虫样本，将其保存在95%乙醇中，并置于-20℃冰箱保存。在所有个体安乐死并经福尔马林固定约6个月后（共32只），我们对每只蛙进行解剖，摘除其肺脏与肠道。将肺脏置于载玻片上，使用光学显微镜（Leica LED2000）观察是否存在肺线虫，但所有样本均未检出肺线虫。随后我们取出大肠，通过粪便分析检测肠道线虫。在大肠远端切口，通过按压挤压大肠以排出粪便，根据粪便量制作1~2张粪便涂片。将涂片置于光学显微镜下观察是否存在成年雌性线虫，若发现成虫则计数并拍摄照片。由于蛙类个体已用福尔马林固定，因此仅将体内发现成年线虫的个体判定为肠道线虫感染个体。尽管粪便浮聚法可用于检测线虫虫卵，但样本经福尔马林固定后无法使用该方法。不过，由于我们对大肠内的所有粪便均进行了分析，因此可以确信：若存在成年雌性线虫，我们必然能够检出。