Molecular prevalence of Bartonella, Babesia, and hemotropic Mycoplasma species in dogs with hemangiosarcoma from across the United States

Hemangiosarcoma (HSA), a locally invasive and highly metastatic endothelial cell neoplasm, accounts for two-thirds of all cardiac and splenic neoplasms in dogs. Bartonella spp. infection has been reported in association with neoplastic and non-neoplastic vasoproliferative lesions in animals and humans.

The objective of this study was to determine the prevalence of Bartonella spp. in conjunction with two other hemotropic pathogens, Babesia spp. and hemotropic Mycoplasma spp., in tissues and blood samples from 110 dogs with histopathologically diagnosed HSA from throughout the United States.

This was a retrospective, observational study using clinical specimens from 110 dogs with HSA banked by the biospecimen repository of the Canine Comparative Oncology and Genomics Consortium. Samples provided for this study from each dog included:fresh frozen HSA tumor tissue (available from n = 100 of the 110 dogs), fresh frozen non-tumor tissue (n = 104), and whole blood and serum samples (n = 108 and 107 respectively). . Blood and tissues were tested by qPCR for Bartonella, hemotropic Mycoplasma, and Babesia spp. DNA; serum was tested for Bartonella spp. antibodies.

Bartonella spp. DNA was amplified and sequenced from 73% of dogs with HSA (80/110). In contrast, hemotropic Mycoplasma spp. DNA was amplified from a significantly smaller proportion (5%, p<0.0001) and Babesia spp. DNA was not amplified from any dog. Of the 100 HSA tumor samples submitted, 34% were Bartonella PCR positive (32% of splenic tumors, 57% of cardiac tumors, and 17% of other tumor locations). Of 104 non-tumor tissues, 63% were Bartonella PCR positive (56% of spleen samples, 93% of cardiac samples, and 63% of skin/subcutaneous samples). Of dogs with Bartonella positive HSA tumor, 76% were also positive in non-tumor tissue. Bartonella spp. DNA was not PCR amplified from whole blood.

This study documented a high prevalence of Bartonella spp. DNA in dogs with HSA from geographically diverse regions of the United States. While 73% of all tissue samples from these dogs were PCR positive for Bartonella DNA, none of the blood samples were, indicating that whole blood samples do not reflect tissue presence of this pathogen. Future studies are needed to further investigate the role of Bartonella spp. in the development of HSA.

Methods
Study design and sample sources

This was a retrospective, observational, descriptive study of 110 dogs with HSA. Specimens used for this study were previously collected and banked by the Canine Comparative Oncology and Genomics Consortium (CCOGC) based on previously published standard operating procedures.[34]. Briefly, the CCOGC collected samples from eight participating veterinary university teaching hospitals starting in 2006 with the goal of creating a repository of clinical samples from dogs with common naturally occurring cancers. Prior to sample submission to CCOGC, dogs were given a definitive diagnosis of neoplasia based on histopathology performed by a board-certified veterinary pathologist at the diagnostic laboratory of each participating university. Tissue samples for the CCOGC biospecimen repository were obtained from the primary tumor via surgical biopsy or post-mortem collection (HSA tumor tissue). As an internal control, tissue samples from each dog were also obtained from adjacent grossly normal tissue in the same organ, or if no grossly normal tissue in the affected organ was apparent, skin biopsies were obtained (non-tumor tissue). Tissue submission (HSA tumor and non-tumor) from each dog required provision of both formalin-fixed (subsequently stored in ethanol) and non-fixed specimens snap-frozen in liquid nitrogen (subsequently stored at -80° C). Dogs also had serum and whole blood collected at the time of tissue sampling for submission to the CCOGC.

For this study, samples from dogs diagnosed with HSA were provided by the CCOGC repository to the investigators. Four sample types were provided for pathogen testing: fresh frozen HSA tumor tissue, fresh frozen non-tumor tissue, whole blood, and serum. Two sample types were provided for histopathological confirmation of tissue type and tumor presence included: formalin-fixed HSA tumor tissue and formalin-fixed non-tumor tissue. Any dog that had a diagnosis of HSA (as determined by the CCOGC SOP), and had an adequate amount of fresh frozen tissue stored in the biorepository at the time of the investigators’ request for specimens, was included. This study, therefore, included 110 dogs with samples collected between May 2008 and November 2011.

Because of previous sample requests or lack of submission of all requested samples to the CCOGC, there were samples missing when provided to the investigators. Of the 110 dogs, 91 had a complete set of the 4 sample types for pathogen testing (fresh frozen HSA tumor tissue, fresh frozen non-tumor tissue, whole blood, serum) provided to the investigators. No dog was missing more than one sample type, and all dogs had at least one fresh-frozen tissue sample (HSA tumor, non-tumor tissue, or both) available for testing. For this reason, dogs with missing samples for pathogen testing were not excluded from the study. Similarly, of the 110 dogs only 37 had a complete set of the 2 sample types for histopathological confirmation (formalin-fixed HSA tumor tissue, formalin-fixed non-tumor tissue) provided to the investigators. Of the 110 dogs, 37 had formalin-fixed HSA tumor tissue and 93 had formalin-fixed non-tumor tissue provided. There were no formalin-fixed samples of HSA tumor or non-tumor tissue from cardiac tumors available for independent histopathologic review. Because a large proportion of dogs were missing formalin-fixed samples for independent histopathological confirmation, the subgroups of tissues with independent histopathologic confirmation of tissue type and tumor presence was analyzed separately.

The CCOGC provided demographic information for each dog, including age (years), breed, weight (kg) and sex and neuter status. The date and geographic location (university teaching hospital) of sample collection was also provided. The anatomic location of tumor and non-tumor tissue samples for each dog was also provided.

Independent confirmation of demographic and histologic data

When possible, information provided by the CCOGC was independently confirmed by the investigators for this study. Histopathology reports providing the diagnosis of hemangiosarcoma wwere provided for 102 dogs; these were reviewed by one author (EL) to confirm hemangiosarcoma diagnosis and tumor location.

For dogs with formalin-fixed biopsy samples provided (see above), biopsies were independently evaluated by a board-certified veterinary pathologist (KL) to confirm the tissue and tumor origin of the biopsy. Formalin fixed tissues were submitted to the NCSU College of Veterinary Medicine histology laboratory (Raleigh NC) and tissues were embedded in paraffin blocks (FFPE blocks). Slides containing 5 um sections were prepared from the FFPE blocks and stained with hematoxylin and eosin (H&E). Samples were categorized by organ type and HSA tumor or non-tumor tissue based on H&E staining. If the tissue of origin was not able to be determined from the biopsy (5 tumor samples and 2 non-tumor tissues), or if no formalin-fixed sample was provided (68 tumor samples and 14 non-tumor tissues), the tissue of origin was categorized by the information provided by the CCOGC and the original diagnostic histopathology reports. Non-tumor tissues from skin or various subcutaneous tissues (including hair, skin, adipose, skeletal muscle, or mammary gland, or any combination of these tissues) were categorized as skin/SQ for analysis.

Pathogen detection methods

DNA was extracted from EDTA anti-coagulated blood and fresh frozen tissue samples using a Qiagen DNeasy® Blood and Tissue kit (Qiagen, Valencia, CA) following the manufacturer’s protocols. For each tissue sample (HSA tumor and non-tumor), a 25 mg piece of tissue was excised from the entire sample using a new, prepackaged sterile scalpel. DNA yield and quality was assessed by spectrophotometry (Nanodrop, Wilmington, DE).

Each DNA sample was screened for the presence of Bartonella spp. DNA using conventional and qPCR, and Babesia spp. and hemotropic Mycoplasma spp. using qPCR. Bartonella qPCR was performed using primers targeting the 16S-23S intragenic transcribed spacer (ITS) region of Bartonella species as described previously[35], in conjunction with a BsppITS438 FAM-labeled hydrolysis probe (TaqMan, Applied Biosystems, Foster City, CA, USA). Bartonella qPCR was performed at two dilutions for each sample (using 1 uL and 5 uL of template DNA respectively). PCR screening for Babesia spp. and hemotropic Mycoplasma spp. were carried out as described previously.[22] Briefly, oligonucleotides Myco16S-322s (5’ GCCCATATTCCTACGGGAAGCAGCAGT 3’) and Myco16S-938as (5’ CTCCACCACTTGTTCAGGTCCCCGTC 3’) were used as forward and reverse primers respectively for hemotropic Mycoplasma spp. DNA amplification. Oligonucleotides Piro18S-144s (5’ GATAACCGTGSTAATTSTAGGGCTAATACATG 3’) and Piroplasma18S-722as (5’ GAATGCCCCCAACCGTTCCTATTAAC 3’) were used as forward and reverse primers respectively for Babesia spp. DNA amplification.

Amplification was performed in a 25-µl final volume reaction containing 12.5 µl of MyTaq Premix (Bioline), 0.2 µl of 100 µM of each forward primer, reverse primer (IDT® DNA Technology), 7.1 µl of molecular–grade water, and 5 µl of DNA from each sample tested.  PCR negative controls were prepared using 5 µl of DNA from blood of a healthy dog. Positive controls for PCR were prepared by using 5 µl of DNA from previously characterized positive dog (clinical cases).  Conventional PCR was performed in an Eppendorf Mastercycler EPgradient® under the following conditions: a single hot-start cycle at 95°C for 2 minutes followed by 55 cycles of denaturing at 94°C for 15 seconds, annealing at 68°C for 15 seconds, and extension at 72°C for 18 seconds.  Amplification was completed by an additional cycle at 72°C for 1 minute, and products were analyzed by 2% agarose gel electrophoresis with detection using ethidium bromide under ultraviolet light.

Validation of positive results was performed by Sanger sequencing of amplicons followed by chromatogram evaluation and sequence alignment using Contig-Express and Align X software (Vector NTI Suite 10.1, Invitrogen Corp, CA, USA). For bacterial species identification, DNA sequences were analyzed for nucleotide sequence homology at NCBI nucleotide database using BLAST version 2.0. A sample was considered Bartonella spp. PCR positive if one or more PCR tests (qPCR or conventional PCR) were positive (tests run in parallel). Stringent processing methods were used to avoid DNA carryover during tissue processing.[36] Specifically, tissue samples were processed independently using manual DNA extraction. For all batches of DNA extractions, between 2 and 4 blanks samples (water) were used as negative controls.  All negative controls for DNA extractions rendered negative results on all PCR assays. DNA carryover after PCR amplification was avoided by processing each sample in three separate laboratory rooms (one for sample sorting and DNA extraction, a second for PCR processing, and a third for PCR analysis post amplification), and strict use of personal protective equipment for sample handling by laboratory personnel.

For IFA testing, Bartonella antibodies were determined using 3 cell culture grown Bartonella spp. (Bartonella henselae, Bartonella vinsonii subsp. berkhoffii, and Bartonella koehlerae) as antigens and following standard immunofluorescent antibody assay (IFA) techniques.[37,38] Briefly, bacterial colony isolates were passed from agar plate grown cultures into permissive cell lines.  For each antigen, heavily infected cell cultures were spotted onto 30-well Teflon-coated slides (Cel-Line/Thermo Scientific), air-dried, acetone-fixed, and stored frozen. Fluorescein conjugated goat anti-dog IgG (KPL, SeraCare, Milford MA) was used to detect bacteria within cells using a fluorescent microscope (Carl Zeiss Microscopy, LLC, Thornwood NY). Serum samples were diluted in phosphate-buffered saline (PBS) solution containing normal goat serum, Tween-20, and powdered nonfat dry milk to block nonspecific antigen binding sites. Sera were first screened at dilutions of 1:16 to 1:64. All sera that are reactive at 1:64 were further tested with two-fold dilutions to 1:8192.

血管肉瘤（Hemangiosarcoma, HSA）是一种局部侵袭性且高转移性的内皮细胞肿瘤，约占犬心脏和脾脏肿瘤的三分之二。巴尔通体属（Bartonella spp.）感染已被报道与动物和人类的肿瘤性及非肿瘤性血管增生性病变相关。

本研究旨在对美国各地经组织病理学确诊为HSA的110只犬的组织及血液样本进行检测，明确巴尔通体属与另外两种嗜血性病原体——巴贝斯虫属（Babesia spp.）和嗜血支原体属（hemotropic Mycoplasma spp.）的检出率。

本研究为回顾性观察性研究，使用犬比较肿瘤学与基因组学联盟（Canine Comparative Oncology and Genomics Consortium, CCOGC）生物样本库中储存的110只HSA患犬的临床标本。本研究获取的每只犬的样本包括：新鲜冷冻HSA肿瘤组织（110只犬中100只可提供）、新鲜冷冻非肿瘤组织（104只）、全血样本（108份）及血清样本（107份）。采用定量聚合酶链反应（quantitative polymerase chain reaction, qPCR）检测血液和组织中的巴尔通体、嗜血支原体及巴贝斯虫属DNA；同时采用血清学方法检测巴尔通体属抗体。

73%的HSA患犬（80/110）的样本中扩增出巴尔通体属DNA并完成测序。与之相比，嗜血支原体属DNA的扩增比例显著更低（5%，p<0.0001），而未在任何犬只样本中扩增出巴贝斯虫属DNA。在提交的100份HSA肿瘤样本中，34%呈巴尔通体PCR阳性（脾脏肿瘤样本中占32%，心脏肿瘤样本中占57%，其余肿瘤部位样本中占17%）。在104份非肿瘤组织样本中，63%呈巴尔通体PCR阳性（脾脏样本中占56%，心脏样本中占93%，皮肤/皮下组织样本中占63%）。在巴尔通体PCR阳性的HSA肿瘤患犬中，76%的非肿瘤组织样本也呈阳性。全血样本中未通过PCR扩增出巴尔通体属DNA。

本研究证实，来自美国地理分布广泛区域的HSA患犬体内巴尔通体属DNA检出率较高。尽管这些犬只的所有组织样本中73%呈巴尔通体DNA PCR阳性，但全血样本均为阴性，这表明全血样本无法反映该病原体在组织中的存在情况。未来需开展进一步研究，以明确巴尔通体属在HSA发生发展中的作用。

## 方法

### 研究设计与样本来源
本研究为回顾性观察描述性研究，纳入110只经确诊的HSA患犬。本研究使用的标本由犬比较肿瘤学与基因组学联盟（Canine Comparative Oncology and Genomics Consortium, CCOGC）依据此前发布的标准操作流程[34]预先收集并储存。简言之，CCOGC自2006年起从8家合作兽医学院教学医院收集样本，旨在建立常见自然发生癌症患犬的临床样本库。在向CCOGC提交样本前，每所合作大学的诊断实验室均由认证兽医病理学家通过组织病理学检查对患犬作出明确的肿瘤诊断。CCOGC生物样本库的组织样本通过手术活检或尸检获取原发性肿瘤组织（即HSA肿瘤组织）。作为内部对照，同时从患犬同一器官的邻近肉眼正常组织获取样本；若受累器官无肉眼正常组织，则获取皮肤活检样本作为非肿瘤组织。每只犬的组织样本（HSA肿瘤组织与非肿瘤组织）需同时提供福尔马林固定（后续置于乙醇中储存）及经液氮快速冷冻、-80℃保存的未固定样本。患犬同时在组织采样时采集血清与全血样本提交至CCOGC。
本研究中，CCOGC生物样本库向研究者提供了确诊HSA患犬的样本。用于病原体检测的样本类型共4种：新鲜冷冻HSA肿瘤组织、新鲜冷冻非肿瘤组织、全血及血清；用于组织类型及肿瘤存在性组织病理学确认的样本类型共2种：福尔马林固定HSA肿瘤组织及福尔马林固定非肿瘤组织。所有经CCOGC标准操作流程（SOP）确诊为HSA，且在研究者提出样本申请时生物样本库中存有足量新鲜冷冻组织的患犬均被纳入本研究。因此，本研究纳入的样本采集时间为2008年5月至2011年11月。
由于此前的样本申请需求或未向CCOGC提交全部所需样本，研究者收到的样本存在缺失情况。110只犬中，91只获得了全部4种病原体检测所需的样本（新鲜冷冻HSA肿瘤组织、新鲜冷冻非肿瘤组织、全血及血清），无犬仅缺失1种以上样本类型，且所有犬至少可提供1种新鲜冷冻组织样本（HSA肿瘤组织、非肿瘤组织或二者兼具），因此样本缺失的患犬未被排除出本研究。同理，110只犬中仅37只获得了全部2种组织病理学确认所需的样本（福尔马林固定HSA肿瘤组织及福尔马林固定非肿瘤组织）；其中37只犬可提供福尔马林固定HSA肿瘤组织，93只犬可提供福尔马林固定非肿瘤组织。无心脏肿瘤的福尔马林固定HSA肿瘤或非肿瘤组织样本可用于独立组织病理学复核。由于多数犬只缺少用于独立组织病理学确认的福尔马林固定样本，因此将对具有独立组织病理学确认的组织类型及肿瘤存在性的亚组进行单独分析。
CCOGC提供了每只犬的人口统计学信息，包括年龄（岁）、品种、体重（kg）、性别及绝育状态，同时提供了样本采集日期及地理来源（合作兽医学院教学医院），以及每只犬的肿瘤与非肿瘤组织样本的解剖部位。

### 人口统计学与组织病理学数据的独立复核
本研究中研究者尽可能对CCOGC提供的信息进行独立复核。102只犬的组织病理学报告已提供，由作者之一（EL）复核以确认血管肉瘤诊断及肿瘤部位。
对于可提供福尔马林固定活检样本的患犬，由认证兽医病理学家（KL）独立评估活检样本以确认活检组织的类型及肿瘤起源。福尔马林固定组织被提交至北卡罗来纳州立大学兽医学院病理学实验室（北卡罗来纳州罗利市），并包埋为石蜡块（formalin-fixed paraffin-embedded, FFPE）。从FFPE块中制备5μm厚切片，经苏木精-伊红（hematoxylin and eosin, H&E）染色。基于H&E染色结果对样本按器官类型及HSA肿瘤/非肿瘤组织进行分类。若无法通过活检确定组织来源（5份肿瘤样本及2份非肿瘤样本），或未提供福尔马林固定样本（68份肿瘤样本及14份非肿瘤样本），则依据CCOGC提供的信息及原始诊断组织病理学报告确定组织来源。源自皮肤或各类皮下组织（包括毛发、皮肤、脂肪组织、骨骼肌或乳腺，或上述组织的任意组合）的非肿瘤组织被归类为皮肤/SQ以供分析。

### 病原体检测方法
采用Qiagen DNeasy®血液与组织试剂盒（Qiagen，加利福尼亚州瓦伦西亚市），按照制造商说明书从乙二胺四乙酸（ethylenediaminetetraacetic acid, EDTA）抗凝全血及新鲜冷冻组织样本中提取DNA。对于每份组织样本（HSA肿瘤组织与非肿瘤组织），使用全新预包装无菌手术刀从整个样本中切取25mg组织块。通过分光光度法（Nanodrop，特拉华州威尔明顿市）评估DNA的产量与质量。
采用常规PCR及定量PCR（qPCR）筛查每份DNA样本中的巴尔通体属DNA，采用qPCR筛查巴贝斯虫属及嗜血支原体属DNA。巴尔通体qPCR采用此前报道的靶向巴尔通体属16S-23S基因间转录间隔区（ITS）的引物[35]，结合BsppITS438 FAM标记水解探针（TaqMan, "Applied Biosystems", Foster City, CA, USA）进行。每份样本采用两种稀释度进行巴尔通体qPCR检测（分别加入1μL及5μL模板DNA）。巴贝斯虫属及嗜血支原体属的PCR筛查方法参照此前报道[22]。简言之，嗜血支原体属DNA扩增的正向引物为Myco16S-322s（5’ GCCCATATTCCTACGGGAAGCAGCAGT 3’），反向引物为Myco16S-938as（5’ CTCCACCACTTGTTCAGGTCCCCGTC 3’）；巴贝斯虫属DNA扩增的正向引物为Piro18S-144s（5’ GATAACCGTGSTAATTSTAGGGCTAATACATG 3’），反向引物为Piroplasma18S-722as（5’ GAATGCCCCCAACCGTTCCTATTAAC 3’）。
扩增反应总体积为25μL，包含12.5μL MyTaq预混液（Bioline）、0.2μL 100μM的正向及反向引物（IDT® DNA Technology）、7.1μL分子生物学级纯水，以及5μL待测样本DNA。PCR阴性对照采用5μL健康犬全血DNA制备；PCR阳性对照采用5μL此前已鉴定的阳性犬DNA（临床病例样本）制备。常规PCR在Eppendorf Mastercycler EPgradient®仪器上进行，反应条件为：95℃热启动2分钟，随后55个循环：94℃变性15秒、68℃退火15秒、72℃延伸18秒，最后72℃延伸1分钟完成扩增。扩增产物通过2%琼脂糖凝胶电泳分离，溴化乙锭染色后在紫外灯下检测。
阳性结果通过桑格测序对扩增产物进行验证，随后使用Contig-Express及Align X软件（"Vector NTI Suite 10.1", Invitrogen Corp, CA, USA）进行色谱图评估及序列比对。对于细菌种类鉴定，使用BLAST 2.0版本在NCBI核苷酸数据库中分析核苷酸序列同源性。若至少1项PCR检测（qPCR或常规PCR）呈阳性，则判定该样本为巴尔通体属PCR阳性。本研究采用严格的操作流程以避免组织处理过程中的DNA交叉污染[36]：具体而言，组织样本采用手动DNA提取方式独立处理；每批次DNA提取过程中设置2~4份空白对照（纯水）作为阴性对照。所有DNA提取的阴性对照在所有PCR检测中均呈阴性结果。为避免PCR扩增后的DNA交叉污染，将样本处理分为三个独立实验室区域（分别为样本分拣与DNA提取区、PCR处理区、扩增后PCR分析区），实验室人员处理样本时严格穿戴个人防护装备。
采用间接免疫荧光试验（IFA）检测巴尔通体抗体，使用3株细胞培养增殖的巴尔通体属菌株（汉赛巴尔通体、伯氏巴尔通体伯克霍夫亚种及科勒巴尔通体）作为抗原，遵循标准免疫荧光抗体试验技术[37,38]。简言之，将琼脂平板培养的菌落接种至允许其增殖的细胞系中。将重度感染的细胞培养物点样至30孔聚四氟乙烯涂层玻片（Cel-Line/赛默飞世尔科技），风干后丙酮固定并冷冻保存。使用荧光素标记的山羊抗犬IgG（KPL, SeraCare, Milford MA），通过荧光显微镜（Carl Zeiss Microscopy, LLC, Thornwood NY）检测细胞内的细菌。血清样本用含有正常山羊血清、吐温-20及脱脂奶粉的磷酸盐缓冲液（PBS）稀释，以阻断非特异性抗原结合位点。血清首先以1:16至1:64的稀释度进行初筛，所有在1:64稀释度下呈反应性的血清进一步进行两倍梯度稀释至1:8192进行检测。