人博卡病毒(Human bocavirus，HBoV)属细小病毒科细小病毒亚科，博卡病毒属，是继细小病毒 B19 之后新发现的第二个对人类致病细小病毒科成员。自 2005 年HBoV1被发现以来，各国研究学者相继在人呼吸道、 粪便、血清、脑脊液等标本中检出4 个基因型别(HBoV 1-4)，在世界范围内广泛流行。HBoV1常见于儿童急性呼吸道感染，可引起轻症、重症不同程度的呼吸道感染，甚至有单一感染致死性病例报道。HBoV2是急性腹泻患儿粪便标本中的主要HBoV型别，检出率为0.8%～26.0%。推测可能是HBoV1与HBoV2的组织嗜性不同，导致HBoV1以呼吸道感染为主，而HBoV2以肠道感染为主。冷冻电镜和3D图像重建技术对HBoV1、HBoV3、HBoV4的VP3蛋白进行结构重建，发现两个重要结构域（VRVII、VRIX）差异可能造就了种属易感性差异，但另两个结构域（VRV、VRIII）差异可能导致不同基因型别的抗原性不同，产生组织嗜性差异。而决定病毒组织嗜性的关键不仅包括病毒衣壳上与宿主细胞结合的特定区域，同时与病毒互作的宿主蛋白也影响病毒对宿主的感染能力。为确定HBoV1与HBoV2的组织嗜性差异，开展了如下工作：

1.      为寻找决定HBoV1 与 HBoV2组织嗜性的关键区域，选择衣壳蛋白VP3作为靶标，以存在至少8个以上氨基酸差异、抗原肽指数高作为基本筛选标准，通过 MEGA7 进行氨基酸序列比对，进一步利用 ProtScale、Jpred4 和 SWISS-MODEL 预测差异区域的二级和三级结构，将具有显著结构差异的区域确定为组织嗜性相关区域。随后，根据差异区域（DR）序列合成多肽，免疫兔，获得抗DR兔血清（anti-DRs），以在大肠杆菌（E.coli）中表达的 VP3 作为抗原，应用Western blotting、ELISA和生物膜层干涉技术明确了anti-DR与不同基因型HBoV的识别和结合能力。最后，应用HBoV1 DNA阳性的急性呼吸道感染患儿鼻咽分泌物作为间接免疫荧光试验（IFA）的抗原，确定各抗体的型别特异性。结果显示，HBoV1与HBoV2 VP3有四个差异区域DR1-4，不仅在氨基酸水平，在二级结构与三级结构也存在显著差异，尤其是HBoV2 VP3上的DR2区域缺失了HBoV1 VP3中存在的四个保守氨基酸²⁰⁴GNAA²⁰⁷，这一结构差异导致该区域三级结构一个半环的缺失。由HBoV1或HBoV2的各四个区域（DR1-4）推导出的8个多肽免疫后获得anti-DRs，Western blotting、ELISA和生物膜层干涉技术的型别特异性检测结果显示仅有anti-DR2具有基因型别特异性，这一特异性结果在呼吸道阳性标本中应用IFA得到验证。综上所述，VP3的DR2（195-220aa位点）被确认为基因型特异性区域，这可能是HBoV1和HBoV2之间潜在的组织嗜性决定性因素之一。

2.      为寻找可能是组织嗜性决定因素的与病毒互作的宿主蛋白，进行了HBoV1与HBoV2与病毒互作的宿主差异性分析。首先在肠道细胞系Caco2中表达了HBoV1和HBoV2的主要衣壳蛋VP3；通过免疫共沉淀(Co-Immunoprecipitation，Co-IP)将与HBoV1 及HBoV2 VP3结合的Caco细胞上的蛋白通过Strep标签蛋白结合到Meg-Strep Beads上进行共沉淀，将互作蛋白进行胶内酶解，经Ziptip C18 固相萃取，使用 Orbitrap Exploris^TM 240 质谱仪分析鉴定多肽混合物，通过生物学软件 MaxQuant (version 2.0.1.0)检索质谱输出数据，用生物学软件OmicsBox（1.2.4）对差异定量蛋白和差异蛋白进行GO富集及相关通路分析。最终，本研究筛选出相互作用差异显著的内源靶蛋白50个，其中VAMP8与HBoV2 VP3的结合量是与HBoV1 VP3 结合量的10倍以上（P< 0.001），推测VAMP8可协助病毒进入宿主细胞，且Caco2内环境更适于HBoV2 VP3表达。BPIFA1与HBoV1 VP3结合量是与HBoV2 VP3结合量 的10倍以上（P< 0.001），且BPIFA1与HBoV1 VP3结合位点位于有型别结构差异的DR3区域。由于BPIFA1大量表达于气道上皮，推测HBoV1 VP3靶向BPIFA1感染呼吸道上皮细胞。

总之，本研究从病毒及宿主两方面进行了HBoV1与HBoV2组织嗜性差异决定性因素研究。HBoV1与HBoV2间主要衣壳蛋白VP3的差异区域DR1-4中，能够诱导具有型别特异性抗体产生的区域为DR2；通过与肠道细胞系进行蛋白互作分析发现的两个与VP3作用差异显著蛋白VAMP8 和 BPIFA1，可能分别是 HBoV2 与 HBoV1的内源性靶蛋白。上述结果为深入研究HBoV1与 HBoV2 组织嗜性差异奠定了基础。

Human bocavirus, an emerging virus, then named HBoV1, was firstly identified in nasopharyngeal aspirates from children with acute respiratory tract infection (ARTI) in 2005 by Allander et al. through new nucleic acid-based techniques, metagenomic detection systems for active virus hunting, which paved the way for virus discoveries. In close succession, three human bocaviruses, HBoV2, 3, and 4, were similarly detected in 2008-2010 in human stool samples when researchers searched for novel viruses causing gastroenteritis. In addition to stool and nasopharyngeal aspirates, the viruses have been found in serum and cerebrospinal fluid and in sewage and river water. It has been confirmed that HBoV1 is a common pathogen for acute respiratory tract infection while HBoV2 was the most common one among HBoV1-4 detected in stool specimens with acute diarrhea, which support the hypothesis that different tissue tropism exists between HBoV1 and HBoV2. The key points in determining the tissue tropism of a virus include the viral factor, the specific regions on the surface of virus capsid proteins that binds to the host cell, and the host factor, proteins that interacts with the virus. Using cryo-electron microscopy and 3D image reconstruction techniques to reconstruct the structures of VP3 proteins of HBoV1, HBoV3, and HBoV4, two critical domains (VRVII, VRIX) were suggested to have species heterotrophy characters, while other two domains (VRV, VRIII) different between HBoV1 and HBoV3/HBoV4 may lead to different antigenicity among different genotypes, resulting in different tissue tropism. Therefore, inorder to reveal the different tissue tropism between HBoV1 and HBoV2, the following research work were performed:

1. To search for the viral factors determing the different tissue tropism between HBoV1 and HBoV2, the divergent regions (DRs) located on capsid protein VP3 were defined through viral amino acid alignment and structure predictions. Peptides deduced from DRs were used as antigens to harvest the corresponding rabbit anti-DRs sera. Then the genotype specificities of these anti-DRs sera were tested using anti-DRs sera as antibodies, and HBoV1 and HBoV2 VP3 expressed in Escherichia coli as antigens in Western blotting, ELISA, and bio-layer interferometry assay, which were then evaluated using clinical specimens in the indirect-immunofluorescence assay (IFA). The results revealed that there were four DRs (DR1-4) located on VP3 with different secondary and tertiary structures between HBoV1 and HBoV2, especially the DR2 region on HBoV2 VP3 lacking the four conserved amino acid ²⁰⁴GNAA²⁰⁷ of HBoV1 VP3 DR2, which explained the deletion of a half-loop in the tertiary structure of this region. In reacting with VP3 of HBoV1 or HBoV2 in Western blotting and ELISA, high intra-genotype cross-reactions were shown in anti-HBoV1 or HBoV2 DR1, 3, and 4, while no intra-genotype cross-reactions were captured inthe anti-HBoV1 DR2 and anti-HBoV2 DR2. The genotype-specific binding capacity of anti-HBoV1 or HBoV2 DR2 sera with VP3 was confirmed by bio-layer interferometry assay and IFA in which only anti-HBoV1 DR2 showed positive results in reacting with HBoV1 positive respiratory specimens. In conclusion, the genotype specificity of anti-HBoV1 DR2 and anti-HBoV2 DR2 against DR2 located on VP3 of HBoV1 or HBoV2 implied that the DR2 (195-220aa) of VP3 was identified as a genotype specific region, which may be one of the potential determinants of tissue tropism between HBoV1 and HBoV2.

2. To capture the host factors determing the tissue tropism between HBoV1 and HBoV2, proteins in host cells with over 10 times binding capcity difference in the interaction of host cell proteins with VP3 of HBoV1 and HBoV2 were searched. The HBoV1 and HBoV2 VP3, which can form virus-like particles, were express in Caco2. The host Caco2 proteins interacting with HBoV1 and HBoV2 VP3 were bound to Meg-Strep Beads by Strep-tagged proteins in co-immunoprecipitation (Co-Immunoprecipitation, Co-IP). The interaction proteins were digested in gel, subjected to Ziptip C18 solid-phase extraction, and analyzed by Orbitrap ExplorisTM 240 mass spectrometer. The mass spectrometry output data was retrieved by the biological software MaxQuant (version 2.0.1.0), and the biological software OmicsBox (1.2.4) was used to perform GO enrichment and related pathway analysis for co-quantified proteins and differential proteins. In this study, 50 differentially interacting endogenous target proteins were identified. VAMP8 and BPIFA1 were enriched in the pathway of antibacterial and antiviral humoral immunity. The binding capacity of VAMP8 with HBoV2 VP3 is over 10 times more than that with HBoV1 VP3 (P=0.000). It is guessed that VAMP8 helps the virus to enter into the host cell, and the internal environment of Caco2 is more suitable for HBoV2 VP3. The binding capacity of BPIFA1 with HBoV1 VP3 is over 10 times more than that with HBoV2 VP3 (P=0.000). BPIFA1 is abundantly expressed in the airway epithelium cells. Moreover, the binding sites of BPIFA1 and HBoV1 VP3 are located in the DR3 region. Therefore, It is speculated that HBoV1 VP3 targets BPIFA1 to infect respiratory epithelial cells and activate the host immune pathway.

In conclusion, we studied the viral and host factors determing the different tissue tropism between HBoV1 and HBoV2. The highest divergent region among DR1-4 between HBoV1 and HBoV2, was DR2 located on VP3, while two endogenous target proteins, VAMP8 and BPIFA1, with significant different binding capacity with the VP3 of HBoV1 and HBoV2 were captured through protein interaction. These data provided highlights for further study of the different tissue tropism between HBoV1 and HBoV2.