This protocol describes microinjection and in vivo electroporation for regionally restricted CRISPR-mediated genome editing in the mouse oviduct.
Germline genetically engineered mouse models (G-GEMMs) have provided valuable insight into in vivo gene function in development, homeostasis, and disease. However, the time and cost associated with colony creation and maintenance are high. Recent advances in CRISPR-mediated genome editing have allowed the generation of somatic GEMMs (S-GEMMs) by directly targeting the cell/tissue/organ of interest.
The oviduct, or fallopian tube in humans, is considered the tissue-of-origin of the most common ovarian cancer, high-grade serous ovarian carcinomas (HGSCs). HGSCs initiate in the region of the fallopian tube distal to the uterus, located adjacent to the ovary, but not the proximal fallopian tube. However, traditional mouse models of HGSC target the entire oviduct, and thus do not recapitulate the human condition. We present a method of DNA, RNA, or ribonucleoprotein (RNP) solution microinjection into the oviduct lumen and in vivo electroporation to target mucosal epithelial cells in restricted regions along the oviduct. There are several advantages of this method for cancer modeling, such as 1) high adaptability in targeting the area/tissue/organ and region of electroporation, 2) high flexibility in targeted cell types (cellular pliancy) when used in combination with specific promoters for Cas9 expression, 3) high flexibility in the number of electroporated cells (relatively low frequency), 4) no specific mouse line is required (immunocompetent disease modeling), 5) high flexibility in gene mutation combination, and 6) possibility of tracking electroporated cells when used in combination with a Cre reporter line. Thus, this cost-effective method recapitulates human cancer initiation.
The fallopian tube, called the oviduct in mice, is a tubular structure that connects the uterus to the ovary. It plays an essential role in mammalian reproduction, providing the environment for internal fertilization and preimplantation development1,2. Despite its importance, little is known about its function and homeostasis, partly due to the development of in vitro fertilization techniques circumventing any infertility issue related to this organ3. However, it has been recognized that precancerous lesions of high-grade serous ovarian carcinoma (HGSC), an aggressive histotype of ovarian cancer that accounts for around 75% of ovarian carcinomas and 85% of related deaths4, are restricted to the distal fallopian tube epithelium5,6,7,8. This indicates that not all cells in our body are equally susceptible to oncogenic insults, but rather only unique/susceptible cells in each tissue/organ become the cell-of-origin in cancer — termed cellular pliancy9. Along these lines, it has been shown that the epithelial cells of the distal fallopian tube, located adjacent to the ovary, are distinct from the rest of the tube10,11. Thus, traditional mouse models of HGSC, that target all cells in the oviduct, do not recapitulate the human condition. In a recent study, we used a combination of CRISPR-mediated genome editing, in vivo oviduct electroporation, and Cre-based lineage tracing to successfully induce HGSC by mutating four tumor suppressor genes in the distal mouse oviduct12,13. This manuscript presents a step-by-step protocol describing this procedure of microinjection and in vivo electroporation to target the mouse oviduct mucosal epithelium.
This method has several advantages. It can be adapted to target other tissues/organs, including organ parenchyma14. Although other in vivo gene delivery approaches like lentiviral and adenoviral systems can be used to achieve similar tissue/organ-specific targeting, the area of targeting is more easily adjusted using different sizes of tweezer-type electrodes for electroporation-based delivery. Depending on the concentration of DNA/RNA/ribonucleoproteins (RNPs), electroporation parameters, and size of electrodes, the number of electroporated cells can be altered. Further, specific cell types can be targeted when used in combination with promotors for Cas9 expression, without the absolute need for Germline genetically engineered mouse models (G-GEMMs). In addition, unlike viral delivery systems, electroporation allows for multiple plasmid delivery into single cells and less constraint in the insert DNA size15. In vivo screening of gene mutations can also be performed with relative ease due to this high flexibility. Further, electroporated cells can be tracked or traced when this method is used in combination with Cre reporter lines such as Tdtomato or Confetti16,17.
Animals were housed in static microisolation cages with filter tops, located in a dedicated room containing a type II biosafety cabinet. All animal work was performed in accordance with institutional guidelines and was approved by McGill University's Faculty of Medicine and Health Sciences Animal Care Committee (AUP #7843).
1. Microinjection needle preparation
Figure 1: Microinjection needle preparation. (A,B) Pulled capillary tube with a pointed end (A) that was snipped to create an opening (B). Please click here to view a larger version of this figure.
2. Preparing the anesthetic for intraperitoneal injection
3. Preparing the surgery area
4. Preparing the solution and needle for injection
Figure 2: Surgery area preparation. (A) Surgery area setup inside a clean bench. (B) Dissection microscope and micromanipulator setup for a right-handed individual. Please click here to view a larger version of this figure.
5. Exposing the female reproductive tract
6. In vivo oviduct injection and electroporation
Figure 3: Female reproductive tract exposure and microinjection. (A) Location of midline incision (shown as a yellow line) on the dorsal side of a Rosa-LSLtdTomato mouse. (B) Female reproductive tract exposure. The fat pad was clamped using a sterile bulldog clamp to anchor the tract and keep it exposed. (C,D) Representative images of in vivo oviduct injection. A microinjection needle was inserted into the distal ampulla (labelled AMP), and filtered 100% trypan blue was injected into the oviduct lumen while the tract was exposed (C). Representative image of a dissected oviduct, demonstrating that injected trypan blue solution distributes throughout the distal and proximal oviduct lumen (D). Please click here to view a larger version of this figure.
Figure 1 and Figure 2 depict the microinjection needle preparation and surgery area setup, respectively, for in vivo microinjection and electroporation of the mouse oviduct. During the surgery, the female reproductive tract was exposed through incisions made in the dorsal skin (Figure 3A) and body wall of an anesthetized Rosa-LSLtdTomato (Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J)17 mouse. A bulldog clamp was clamped onto the fat pad above the tract to keep it exposed and anchored (Figure 3B). Injection solution containing PCS2 CreNLS plasmids18 at 400 ng/µL concentration was injected into the ampulla (labelled AMP) of the coiled oviduct and allowed to disperse throughout the oviduct lumen (Figure 3C,D). Although PCS2 CreNLS plasmids were used to generate representative results in this manuscript, the described method can be used with easily interchangeable DNA/RNA/RNPs in the injection solution for CRISPR/Cas-mediated gene editing in the mouse oviduct. Then, 1 mm or 3 mm tweezer type electrodes were positioned, such that the distal oviduct was in between the electrodes for electroporation-based delivery.
Oviducts were harvested 4 days following the surgery and stretched out by removing the mesosalpinx to visualize electroporation specificity. Using either 1 mm or 3 mm tweezer-type electrodes, the targeted area, labelled with TdTomato, was restricted to the distal oviduct epithelium (Figure 4A,D). The size of this targeted area was further controlled by using electrodes of different sizes. Using 3 mm tweezer-type electrodes, we targeted a much larger area of the AMP (Figure 4B,C), as compared to targeting only the distal-most tip, the INF (Figure 4E,F), using 1 mm tweezer-type electrodes. These harvested oviducts were fixed, treated with a sucrose gradient, and sectioned using a cryostat for detailed analyses. Sections were counterstained with Hoescht 33342 and Phalloidin to stain the nuclei and cytoskeleton respectively, then they were imaged using a point-scanning confocal microscope. In the targeted region, electroporated cells, marked by TdTomato, were randomly distributed among non-electroporated (TdTomato-ve) cells (Figure 4G). Additionally, electroporated cells were restricted to the mucosal epithelium and not found in the underlying stromal or muscle layer (Figure 4G,H). Finally, to confirm CRISPR-Cas-mediated gene editing, TdTomato+ve cells can be isolated by fluorescence-activated cell sorting (FACS) for DNA isolation and MiSeq sequencing12.
Figure 4: Validation of successful in vivo injection and electroporation of oviduct epithelial cells, 4 days post-surgery. (A–C) Electroporation using 3 mm sized tweezer-type electrodes. The oviduct was uncoiled by removing the mesosalpinx for better visualization of electroporation specificity (A). Electroporation area, identified by TdTomato expression, was restricted to the distal oviduct, labelled AMP (B,C). (D–F) Electroporation using 1 mm sized tweezer-type electrodes. The oviduct was uncoiled by removing the mesosalpinx for better visualization of electroporation specificity (D). The electroporation area was restricted to the infundibulum (labelled INF), the distal-most tip of the oviduct (E,F). (G,H) Transverse section of the distal oviduct 4 days following electroporation with PCS2 CreNLS plasmids. Electroporated cells labelled with TdTomato (G) were restricted to the epithelial monolayer (H). Please click here to view a larger version of this figure.
Crucial steps in this detailed protocol are the microinjection of DNA/RNA/RNP solution into the oviduct lumen and control of the electroporation strength and area. DNA/RNA/RNP solution leakage during microinjection may cause transfection of undesired areas/cells. For consistent and efficient electroporation, it is preferable to fill the oviduct lumen with the solution (Figure 3C,D). This is because the area of electroporation is mainly controlled by electrode size and placement. Weak electroporation reduces the number of electroporated cells, and harsh electroporation may cause the electroporation of undesired cells or disrupt tissue structure. The number of electroporated cells can be varied by adjusting the DNA/RNA/RNP solution concentration or electroporation parameters. However, since high voltages/heat production during electroporation could damage the tissue, these parameters should be tested prior to use on live/anesthetized mice. Finally, due to the demanding nature of this procedure, it is recommended to practice tract exposure and microinjection on dead mice prior to performing this surgery on live/anesthetized mice.
It is recognized that multiple genes are involved in cancer initiation, thus, recapitulating this event requires multi-allelic modification of several genes. Additionally, it is also known that not all cells are equally susceptible to oncogenic mutations9. Cancer modeling, therefore, requires specific Cre mouse lines to achieve tight control of oncogenic insults. However, their availability and specificity causes various challenges, including effects in non-targeted tissues and lethality19. Further, traditional G-GEMMs incur high costs for maintenance and require time for mouse line generation. The development of CRISPR/Cas9 genome editing technology and improvement of gene delivery into specific somatic cells has allowed us to overcome these issues. In this protocol, we present a DNA/RNA/RNP delivery method for somatic genome manipulation that can be used for cancer modeling, without the absolute need for specific mouse lines12. By injecting DNA/RNA/RNP solutions into the lumen and using different sizes of tweezer-type electrodes for electroporation-based delivery, restricted areas of the mouse oviduct mucosal epithelium are targeted (Figure 4A–F). This is especially useful in modeling the initiation of HGSCs that originate from the distal end of the fallopian tube.
The presented microinjection and in vivo electroporation method is highly versatile for targeting tissues/organs with a lumen. Organ parenchyma is also targetable using this method14. Viral delivery approaches, like lentiviral and adenoviral systems, can also be used for similar purposes. However, the advantages of electroporation over viral delivery approaches are: 1) multiple plasmid delivery into single cells, 2) no restriction on the insert size15, and 3) easy control of the area and timing of delivery. In addition, targeting specificity can be improved by using lineage specific promoters. This is sometimes difficult in viral systems due to the limit on viral packaging.
As is the nature of electroporation, electroporated epithelial cells are randomly interspersed with healthy, unedited cells (Figure 4G). However, this mosaicism in genetically modified and unmodified cells could be advantageous to study cancer initiation, as it recapitulates the sporadic nature of early cancer initiation within an immunocompetent microenvironment. The frequency of electroporated cells can be adjusted by varying the DNA/RNA/RNP solution concentrations and/or electroporation parameters. Using CRISPR-Cas-mediated gene editing in combination with the presented protocol, it is easy to screen multiple genes and generate heterogeneous patterns of mutations/allelic combinations in targeted genes; this is particularly useful in modeling cancers that present with a considerable number of genomic alterations like HGSCs20,21. Further, by sequencing targeted genes during cancer progression, it is possible to follow the in vivo clonal evolution of tumors and metastases in immunocompetent mice12.
The authors have nothing to disclose.
The authors thank Katie Teng and Dr. Matthew J. Ford for providing plasmids used to generate representative results and protocol optimization. K.H. was supported by Fonds de recherche du Québec – Santé (FRQS) doctoral grant, Donnor Foundation, Delta Kappa Gamma World Fellowship, Centre for Research in Reproduction and Development (CRRD), Hugh E. Burke, Rolande & Marcel Gosselin, and Alexander McFee graduate studentships.
0.22 µm sterile filter unit | LifeGene | SF0.22PES | |
1 mL syringe | Terumo Medical Corportation | SS-01T | |
1 mm tweezer-type electrodes | BEX CO., LTD | LF650P1 | |
3 mm tweezer-type electrodes | BEX CO., LTD | LF650P3 | |
30G x 1/2 Needle | BD | 305106 | Needle is attached to the 1 mL syringe when using injectable anesthetic or analgesic. |
Adson forceps | Fisher Scientific | 10-000-451 | |
Air-pressured syringe | BD | B302995 | Attached to the micromanipulator; as shown in Figure 2B |
Analgesic | – | – | Carprofen, dose: 5mg/kg. |
Antiseptic | Cardinal Health | OMEL0000017 | Baxedin: 2% chlorhexidine in 70% isopropyl alcohol |
Avertin (2,2,2-tribromoethanol) – Anesthetic | Sigma Aldrich | T48402-25G | |
Clamp mount micromanipulator | AD Instruments | MM-33 | |
Curved serrated forceps | Fisher Scientific | NC0696845 | |
Dissecting microscope | Zeiss | SteREO Discovery.V8 | Apochromat S, 0.63X, FWD 107mm. Attached KL 200 LED for top light, labelled as light source in Figure 2B. |
Eye lubricant | Alcon | – | Systane ointment (Lubricant eye ointment) |
Glass capillary needles | Sutter Instrument Co. | B100–50-10 | Outside diameter: 1 mm, inside diameter: 0.5 mm, length: 10 cm |
Hartman hemostats | Fisher Scientific | 50-822-711 | |
Heating pad/disc | – | – | SnuggleSafe Pet Bed Microwave Heating Pad was used in this protocol. Microwave for 2-5min, and test with back of gloved hand. |
Magnetic Mount for micromanipulator | AD Instruments | MB-B | Used to keep the clamp mount micromanipulator stable and upright during injection. |
Micro bulldog clamp | Fisher Scientific | 50-822-230 | 3 cm |
Micropipette puller | Sutter Instrument Co. | P-97 | P-97 Flaming/Brown type micropipette puller. Program used: P = 500, Heat = 576, PULL = 50, VEL = 80, DEL = 70 |
Petri dish | Fisher Scientific | 263991 | Autoclaved/sterile glass petridishes can also be used. |
Phosphate Buffered Saline (PBS) | Bio Basic Inc. | PD8117 | 10X PBS was diluted to 1X with DI water. Autoclave before use. |
Pulse generator/Electroporator | BEX CO., LTD | CUY21 EDIT II | Electroporation settings: 30 V, 3 pulses, 1 s interval, P. length = 50 ms, unipolar |
Rosa-LSLtdTomato mice | The Jackson Laboratory | 7914 | Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J |
Sharp-blunt dissection scissors | Fisher Scientific | 28251 | |
Sharp-pointed dissecting scissors | Fisher Scientific | SDI130 | |
Shaver | – | – | Hair removal cream can also be used. |
Silk braided sutures | Ethicon | 682G | 3/8 circle, gauge: 5-0, needle size: 18 mm, thread length: 75 cm. Staples can also be used. |
Tapered ultrafine tip forceps | Fisher Scientific | 12-000-122 | |
Thin absorbent paper | Kimberly-Clark Professional | 34120 | Kimwipes |
Trypan blue | STEMCELL Technologies | 7050 | Filtered using 0.22 µm sterile filter before use for microinjection. |