Comprehensive study for SNARE involved in the post-Golgi transport in Drosophila photoreceptors

Polarized transport is essential for the construction of multiple plasma membrane domains within cells. Drosophilaphotoreceptors serve as excellent model systems for studying the mechanisms of polarized transport. We conducted a comprehensive SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptor) screening of the fly genome using RNAi knockdown and CRISPR/Cas9 somatic knockout combined with the CoinFLP system to identify SNAREs involved in post-Golgi trafficking. The results suggest that, in post-Golgi transport, no SNARE is exclusively responsible for transport to a single specific plasma membrane domain. However, each SNARE shows some preference for certain membrane domains: the loss of nSyb, Ykt6, and Snap24/25 results in severe defects in rhabdomere transport, while the loss of Syx1A and Snap29 leads to significant impairments in basolateral transport. Together with the function of Syx1A, Snap25, and nSyb in the fusion of synaptic vesicles with the synaptic plasma membrane, these results suggest that SNAREs are not the sole determinants for vesicles to specify their target subdomains in the plasma membrane. Furthermore, rhodopsin transport to the rhabdomere requires two kinds of R-SNAREs, Ykt6 and nSyb, suggesting that multiple sets of post-Golgi SNAREs are contributing in tandem or in cooperation, rather than in parallel.

Methods
Plasmids were constructed as follows. 

pU62-wsh1
pU62-wsh1 is a plasmid carrying a cassette that works as a "dominant white-eye" marker in Drosophila, consist of Drosophila U6-2 promoter and shRNA against white gene. This vector was constructed based on a vector, pAc-sgRNA-Cas9 (Bassett et al., 2014) by inserting a synthetic shRNA unit against white gene, HMS00017, predesigned in TRiP RNAi & CRISPR fly project (Zirin et al., 2020). Four oligonucleotides TTC-wsh1-F1, wsh1-R1p, wsh1-F2p and wsh1-Ap-GTT were ligated into pAc-sgRNA-Cas9 digested with BbsI, to obtain pU62-wsh1.

pBIDwsh-UASC
pBIDwsh-UASC is a phi31 transgenic vector based on pBID-UASC (Wang et al., 2012) but carrying "dominant white-eye" marker instead of miniature-white “red-eye” marker. From pU62-wsh1, U6-wshRNA cassette was amplified with primers Nh-U62 and gCas9puro-F, then digested with NheI and ApaI, and then cloned between NheI and ApaI sites of pBID-UASC to obtain pBID-wsh-UASC.

pBIDwsh-Act-coinFLP-Gal4
pBIDwsh-Act-coinFLP-Gal4 is a phi31 transgenic vector carrying "dominant white-eye" marker and coinFLP-Gal4 cassette under control of Act5C promoter. From original coinFLP vector, pAct-FRT-stop-FRT3-FRT-FRT3-Gal4 attB (Bosch et al., 2015), the DNA fragment including Act5C promoter and FRT-stop-FRT3-FRT-FRT3-Gal4 cassette was amplified with primers Sbf-BID-Act-F and Gal4-BID-Xh-R. Using Gibson assembly, the fragment was cloned between SbfI and BglII sites of pBIDwsh-UASC, replacing UASC promoter and multiple cloning sites, to obtain pBIDwsh-Act-coinFLP-Gal4.

5x longGMR promoter
The longGMR promoter (Wernet et al., 2003) consists of five copies of enhancer sequence in the upstream of Rh1 promoter and a basal promoter core of hsp70. Since the detail of the construct was not described, the region containing longGMR promoter was amplified using PCR with primers wdelR_Nhe and Gal4DBD-R2 from a transgenic fly caring P{longGMR-Gal4} (Bloomington Drosophila Stock Center #8605)..
pP-longGMR, pBID-longGMR and pBIDwsh-longGMR
The fragment containing longGMR promoter was digested with SphI and EcoRI and then ligated between SphI and EcoRI sites of pPdM-UAST (Yamashita et al., 2022), pBID-UASC, and pBIDwsh-UASC, to obtain pP-longGMR, pBID-longGMR, and pBIDwsh-longGMR, respectively.
pP-3xlongGMR, pBID-3xlongGMR and pBIDwsh-3xlongGMR
DNA fragments with various number of LGMR repeats were amplified from pP-longGMR, with primers Sph-longGMR and UAST-R4. The fragments were digested with SphI and EcoRI, then separated by size in agarose gel electrophoresis, and then ligated into pPdM-UAST, pBID-UASC, and pBIDwsh-UASC, to obtain pP-3xlongGMR, pBID-3xlongGMR, and pBIDwsh-3xlongGMR, respectively.
pP-longGMR-coinFLP-Gal4 and pBIDwsh-longGMR-coinFLP-Gal4
From pBIDwsh-Act-coinFLP-Gal4, DNA fragments including FRT-stop-FRT3-FRT-FRT3-Gal4 cassette were excised with BamHI and KpnI, and then ligated between BglII and KpnI sites in pP-longGMR, pP-3xlongGMR and pBIDwsh-longGMR to obtain pP-LongGMR-coinFLP-Gal4, pP-3xlongGMR-coinFLP-Gal4 and pBIDwsh-longGMR-coinFLP-Gal4.

pX-HFCas9
A mammalian CRISPR-Cas9 KO vector, pX-HFCas9, coding HiFiCas9R691A (Vakulskas et al., 2018) was constructed based on pSpCas9(BB)-2A-Puro PX459 V2.0 (Ran et al., 2013), by introducing R691A mutation in the Cas9 gene using Gibson assembly with primers HF-Cas9-F and HF-Cas9-R.
pBIDwsh-Act-coinFLP-HFCas9-H2B-2xNG and pP-3xlongGMR-coinFLP-HFCas9-H2B-2xNG
DNA fragments including HFCas9-T2A was amplified from pX-HF-Cas9 with Bg-Coin-Cas9 and ddT2A-R. DNA fragments including H2B-GFP was amplified from pcDNA3.1 miniSOG2 T2A H2B-EGFP (Makhijani et al., 2017) with ddT2A-F2 and H2B-EGFP-BID-K. Using Gibson assembly, fragments HFCas9-T2A and H2B-2xmNeonGreen were cloned between BglII and KpnI sites of pBIDwsh-Act-coin-Gal4 and pP-3xlongGMR-coinFLP-Gal4 to obtain pBIDwsh-Act-coinFLP-HFCas9-H2B-2xNG and pP-3xlongGMR-coinFLP-HFCas9-H2B-2xNG.

pCFD4w-Syx6
A phi31 vector pCFD4w was constructed by replacing promoter-gRNA scaffold of pCFD5w, a phi31 vector with white+ marker (gift from Michael Boutros, Addgene plasmid #112645), by those of pCFD4. DNA fragments including dU6:2-BbsI-gRNA scaffold were amplified with primers Sgf-U61F and dU6-wsh-fix and digested with EcoRI and NheI, then inserted between EcoRI and XbaI sites of pCFD5w. To construct pCFD4w-Syx6, oligonucleotides Syx6-gRNA3f and Syx6-gRNA3r were annealed and ligated with BbsI digested pCFD4w.

pUAST-myc-Ykt6, pUAST-HA-Snap24 and pUAST-HA-Snap25 
From third instar larvae cDNA, DNA fragment encoding Ykt6 were amplified using primers Xh-dYkt6 and dYkt6-Mlu, then digested with XhoI and MluI, and then ligated between XhoI and MluI sites in pMT-HA-NBgA-m, to generate pMT-HA-Ykt6. The fragment containing Ykt6 was excised with XhoI and XbaI and then inserted into pPdM-UAST-m2NBgAm to generate pUAST-myc-Ykt6. DNA fragment encoding snap24 and snap25 were amplified using primers Xh-Snap24, Snap24-Ap, Xh-Snap25 and Snap25-Ap, digested with XhoI and ApaI, and cloned into pMT-HA-NBgA-m to obtain pMT-HA-snap24 and pMT-HA-snap25. HA-snap24 and HA-snap25 fragments were excised with KpnI and MluI, and then inserted into pPdM-UAST-m2NBgAm to generate pUAST-HA-Snap24 and pUAST-HA-Snap25.

极性运输对于细胞内多种质膜结构域的构建至关重要。果蝇光感受器是研究极性运输机制的优秀模型系统。本研究通过RNA干扰（RNA interference, RNAi）敲低、CRISPR/Cas9体细胞敲除结合CoinFLP系统，对果蝇全基因组开展了全面的SNARE（可溶性N-乙基马来酰亚胺敏感因子附着蛋白受体，soluble N-ethylmaleimide-sensitive factor attachment protein receptor）筛选，以鉴定参与高尔基后运输的SNARE蛋白。

研究结果表明，在高尔基后运输过程中，并无单一SNARE蛋白专门负责向某一特定质膜结构域的运输，但每种SNARE蛋白对特定膜结构域均表现出一定偏好性：nSyb、Ykt6以及Snap24/25的缺失会导致感杆束（rhabdomere）运输出现严重缺陷，而Syx1A与Snap29的缺失则会使基底外侧运输受到显著损伤。结合Syx1A、Snap25以及nSyb在突触囊泡与突触质膜融合过程中的功能，上述结果提示，SNARE蛋白并非囊泡在质膜上靶向特定亚结构域的唯一决定因素。此外，视紫红质向感杆束的运输需要Ykt6与nSyb这两种R-SNARE蛋白，表明多组高尔基后SNARE蛋白是以串联或协同而非平行的方式发挥作用。

实验方法
质粒构建流程如下：

pU62-wsh1
pU62-wsh1为携带果蝇“显性白眼”标记盒的质粒，该标记盒由果蝇U6-2启动子与针对white基因的短发夹RNA（short hairpin RNA, shRNA）组成。本载体以pAc-sgRNA-Cas9（Bassett等，2014）为骨架，通过插入TRiP RNAi与CRISPR果蝇项目（Zirin等，2020）中预先设计的针对white基因的合成shRNA单元HMS00017构建得到。将寡核苷酸TTC-wsh1-F1、wsh1-R1p、wsh1-F2p与wsh1-Ap-GTT连接至经BbsI酶切的pAc-sgRNA-Cas9载体中，即可获得pU62-wsh1。

pBIDwsh-UASC
pBIDwsh-UASC为基于pBID-UASC（Wang等，2012）构建的phi31转基因载体，但其标记替换为“显性白眼”标记而非微型白眼“红眼”标记。从pU62-wsh1中，以引物Nh-U62与gCas9puro-F扩增得到U6-wshRNA盒，经NheI与ApaI酶切后，克隆至pBID-UASC的NheI与ApaI酶切位点之间，即可获得pBIDwsh-UASC。

pBIDwsh-Act-coinFLP-Gal4
pBIDwsh-Act-coinFLP-Gal4为携带“显性白眼”标记与受Act5C启动子调控的coinFLP-Gal4盒的phi31转基因载体。从原始coinFLP载体pAct-FRT-stop-FRT3-FRT-FRT3-Gal4 attB（Bosch等，2015）中，以引物Sbf-BID-Act-F与Gal4-BID-Xh-R扩增得到包含Act5C启动子与FRT-stop-FRT3-FRT-FRT3-Gal4盒的DNA片段。通过Gibson组装，将该片段克隆至pBIDwsh-UASC的SbfI与BglII酶切位点之间，替换原有的UASC启动子与多克隆位点，即可获得pBIDwsh-Act-coinFLP-Gal4。

5x长GMR启动子（5x longGMR promoter）
长GMR启动子（Wernet等，2003）由Rh1启动子上游的5个增强子拷贝序列与hsp70的基础启动子核心区域组成。由于该构建体的具体细节未被公开报道，我们以携带P{longGMR-Gal4}的转基因果蝇（Bloomington果蝇种质中心 #8605）为模板，使用引物wdelR_Nhe与Gal4DBD-R2通过PCR扩增得到包含长GMR启动子的区域。

pP-longGMR、pBID-longGMR与pBIDwsh-longGMR
将包含长GMR启动子的片段经SphI与EcoRI酶切后，分别连接至pPdM-UAST（Yamashita等，2022）、pBID-UASC与pBIDwsh-UASC的SphI与EcoRI酶切位点之间，即可分别获得pP-longGMR、pBID-longGMR与pBIDwsh-longGMR。

pP-3xlongGMR、pBID-3xlongGMR与pBIDwsh-3xlongGMR
以pP-longGMR为模板，使用引物Sph-longGMR与UAST-R4扩增得到不同LGMR重复拷贝数的DNA片段。将这些片段经SphI与EcoRI酶切后，通过琼脂糖凝胶电泳按分子量分离，随后分别克隆至pPdM-UAST、pBID-UASC与pBIDwsh-UASC中，即可分别获得pP-3xlongGMR、pBID-3xlongGMR与pBIDwsh-3xlongGMR。

pP-longGMR-coinFLP-Gal4与pBIDwsh-longGMR-coinFLP-Gal4
从pBIDwsh-Act-coinFLP-Gal4中，使用BamHI与KpnI酶切得到包含FRT-stop-FRT3-FRT-FRT3-Gal4盒的DNA片段，随后将其连接至pP-longGMR、pP-3xlongGMR与pBIDwsh-longGMR的BglII与KpnI酶切位点之间，即可获得pP-LongGMR-coinFLP-Gal4、pP-3xlongGMR-coinFLP-Gal4与pBIDwsh-longGMR-coinFLP-Gal4。

pX-HFCas9
哺乳动物CRISPR-Cas9敲除载体pX-HFCas9编码HiFiCas9R691A（Vakulskas等，2018），其构建以pSpCas9(BB)-2A-Puro PX459 V2.0（Ran等，2013）为骨架，通过使用引物HF-Cas9-F与HF-Cas9-R进行Gibson组装，在Cas9基因中引入R691A突变得到。

pBIDwsh-Act-coinFLP-HFCas9-H2B-2xNG与pP-3xlongGMR-coinFLP-HFCas9-H2B-2xNG
以pX-HF-Cas9为模板，使用引物Bg-Coin-Cas9与ddT2A-R扩增得到包含HFCas9-T2A的DNA片段；以pcDNA3.1 miniSOG2 T2A H2B-EGFP（Makhijani等，2017）为模板，使用引物ddT2A-F2与H2B-EGFP-BID-K扩增得到包含H2B-GFP的DNA片段。通过Gibson组装，将HFCas9-T2A与H2B-2xmNeonGreen片段分别克隆至pBIDwsh-Act-coin-Gal4与pP-3xlongGMR-coinFLP-Gal4的BglII与KpnI酶切位点之间，即可获得pBIDwsh-Act-coinFLP-HFCas9-H2B-2xNG与pP-3xlongGMR-coinFLP-HFCas9-H2B-2xNG。

pCFD4w-Syx6
phi31载体pCFD4w的构建方法为：将带有white+标记的phi31载体pCFD5w（Michael Boutros馈赠，Addgene质粒#112645）的启动子-gRNA支架替换为pCFD4的对应序列。使用引物Sgf-U61F与dU6-wsh-fix扩增得到包含dU6:2-BbsI-gRNA支架的DNA片段，经EcoRI与NheI酶切后，插入至pCFD5w的EcoRI与XbaI酶切位点之间。为构建pCFD4w-Syx6，将寡核苷酸Syx6-gRNA3f与Syx6-gRNA3r退火后，连接至经BbsI酶切的pCFD4w载体中。

pUAST-myc-Ykt6、pUAST-HA-Snap24与pUAST-HA-Snap25
从三龄幼虫cDNA中，使用引物Xh-dYkt6与dYkt6-Mlu扩增得到编码Ykt6的DNA片段，经XhoI与MluI酶切后，连接至pMT-HA-NBgA-m的XhoI与MluI酶切位点之间，构建得到pMT-HA-Ykt6。将包含Ykt6的片段经XhoI与XbaI酶切后，插入至pPdM-UAST-m2NBgAm中，构建得到pUAST-myc-Ykt6。使用引物Xh-Snap24、Snap24-Ap、Xh-Snap25与Snap25-Ap扩增得到编码snap24与snap25的DNA片段，经XhoI与ApaI酶切后，克隆至pMT-HA-NBgA-m中，得到pMT-HA-snap24与pMT-HA-snap25。将HA-snap24与HA-snap25片段经KpnI与MluI酶切后，插入至pPdM-UAST-m2NBgAm中，构建得到pUAST-HA-Snap24与pUAST-HA-Snap25。