Improving germline transmission efficiency in chimeric chickens using a multi-stage injection approach

Although different strategies have been developed to generate transgenic poultry, low efficiency of germline transgene transmission has remained a challenge in poultry transgenesis. Herein, we developed an efficient germline transgenesis method using a lentiviral vector system in chickens through multiple injections of transgenes into embryos at different stages of development. The embryo chorioallantoic membrane (CAM) vasculature was successfully used as a novel route of gene transfer into germline tissues. Compared to the other routes of viral vector administration, the embryo’s bloodstream at Hamburger-Hamilton (HH) stages 14–15 achieved the highest rate of germline transmission (GT), 7.7%. Single injection of viral vectors into the CAM vasculature resulted in a GT efficiency of 2.7%, which was significantly higher than the 0.4% obtained by injection into embryos at the blastoderm stage. Double injection of viral vectors into the bloodstream at HH stages 14–15 and through CAM was the most efficient method for producing germline chimeras, giving a GT rate of 13.6%. The authors suggest that the new method described in this study could be efficiently used to produce transgenic poultry in virus-mediated gene transfer systems.

of transgenes [5], the need to wait for two generations before having ubiquitously 48 expressing transgenic birds [3], insertional mutagenesis due to the lack of precise control over  Numerous studies have attempted to produce genetically-modified transgenic birds by 55 transducing blastoderm embryos at Eyal-Giladi and Kochav (EGK) stage X [8] using 56 retroviruses. However, the transduction efficiency of G 0 founders has been reported to be low 57 with a limited rate of GT [9][10][11][12][13]. To further improve the efficiency of this system, some research 58 groups have used lentiviral vectors [14][15][16][17][18][19][20][21]. The GT efficiency using lentiviral vectors was low 59 in these reports, ranging between 0.6 and 4.0%, although a relatively higher GT (average 17.8%) 60 has also been reported by a research group following injection of lentiviral vectors into the 61 subgerminal cavity of newly laid eggs [22]. 62 To optimize the GT, some researchers have attempted to determine the ideal 63 developmental stage for viral vector injection in chicken embryos. For example, Kawabe et al. 64 [23] reported that the highest rate of transduction of germline cells could be obtained when the  The biological titration of lentiviral vectors expressing EGFP was carried out by flow 106 cytometry as previously described [36]. Lentiviral stocks with titers of 1×10 9 transducing units 107 per milliliter (TU/ml) were used for experimentation.

108
Chicken embryo injection and hatching 6 109 Fertilized chicken eggs were obtained from a local poultry farm at the peak production 110 period. Each embryo was injected thrice (Group BAC) at different stages of its development and 111 through different routes as follows: at stage X, into the sub-germinal cavity of blastoderm 112 embryo, at HH stages 14-15, into the dorsal aorta of the embryos, and at HH stage 37 (on the 113 11 th day of incubation), into the CAM vasculature. For the first injection, fresh and fertile eggs 114 were placed vertically with their blunt end down at room temperature (RT) for 4 hours before   were injected through the transparent eggshell membrane into the blood vessels using a glass 169 needle (inner diameter: 80 µm). Before injection, the plunger was aspirated slightly to ensure 170 that the needle was successfully inserted into a blood vessel (Fig 1h). After the eggshell window 171 was sealed with hot glue (Fig 1i), the eggs were returned to the incubator. At the end of the 18 th 172 day of incubation, egg turning was stopped and the eggs were placed in hatching trays at 37.2 °C 173 and 65 % relative humidity until hatching (Fig 1j).     (Table 1), the lowest frequency was 276 observed when the injection was performed into the blastoderm (P < 0.05), whereas no 277 significant difference was observed between the dorsal aorta and CAM route groups (P = 0.67).

278
Both the double injection into the dorsal aorta and CAM route (AC group) and the triple 279 injection (BAC group) achieved higher germline-transgenic frequencies in the G 0 generation, 280 compared to the single injection groups (P < 0.05), while no significant difference was found 281 between the double-and triple-injection groups (P > 0.41).

300
We quantitatively assessed the EGFP expression at both mRNA (Fig 2b) and protein (Fig   301   3) levels in the testes of G 0 newly hatched chicks. The RT-qPCR data revealed that the triple-302 injection group had the highest EGFP expression, which was significantly higher than that of the 303 double-and single-injection groups (P < 0.05) (Fig 2b). Among the double-injection groups, the 304 relative expression of EGFP was significantly lower in the BC group than BA and AC groups (P 305 < 0.05) (Fig 2b). Among double-and single-injection groups, only two groups (BC and A) 306 showed no significant difference (P > 0.05) (Fig 2b). No significant differences were observed 307 among the single-injection groups (P > 0.05) (Fig 2b). double injection groups than that of the single injection groups (P < 0.05) (Figs 3b and 3c).

328
Among the double injection groups, no significant difference was observed between BA and BC 329 groups (P > 0.05); however, the percentages of EGFP positive testicular cells in the 330 aforementioned groups were significantly lower than that of the AC group (P < 0.05) (Figs 3b 331 and 3c). We also detected no significant difference between groups A and C in terms of the 332 percentage of EGFP positive cells (P > 0.05), but the mean values for these two groups were 333 significantly higher than that of the group B (P < 0.05) (Figs 3b and 3c).

334
Western blot analysis and fluorescence microscopy 335 We assessed the EGFP and β-actin expression at the protein level by Western blot 336 analysis in the testicular tissues of G 0 hatched chicks (Fig 4a) in which their EGFP expression 17 337 was previously confirmed by the RT-qPCR (Fig 2b) and flow cytometry (Fig 3)  testis tissues taken from a G 0 chick in group C (Fig 4c). To assess the germline transmission (GT) efficiency, three germline chimeric roosters 356 from each group were randomly selected and mated with wild-type hens. DNA derived from the 357 blood of G 1 chicks, obtained by the above crossing, was assessed for the presence of EGFP using 358 PCR (Fig 5a), and the results were expressed as transgenic ratios in G 1 progeny ( Table 1). The  eggs at the desired developmental stages (Fig 1). In this modified approach, we used the original 387 eggshells instead of the surrogate eggshells [38] to maintain embryo development from laying 388 eggs to hatching. Up to our knowledge, this is the first report presenting this novel modified 389 approach for improving germline transmission efficiency in chicken chimeras.

390
The minimum hatchability of this technique was 52.9%, which is higher than that  which is considered as the target of gene transfer at later stages.

451
The results regarding the EGFP expression at the mRNA level were not totally in line 452 with the above findings. The relative expression of EGFP at the mRNA level was similar 453 between the groups with a single injection despite to the different injection routes. However, in 454 the double-and triple-injection groups, the relative expression of EGFP at the mRNA level was 455 significantly higher than that of the single injection groups (Fig 2b). These findings indicate that 456 although the mRNA expression level was higher in the BA, BC, and BAC groups, which 457 received an additional injection into the blastoderm, their GT efficiency did not increase 458 compared to A, C, and AC groups, respectively. This incompatibility may be related to the 459 differential ability of PGC in the blastoderm, compared to that of the circulating PGC at HH

463
It is also of note that the number of PGC at the blastoderm stage was reported to be significantly 464 lower compared to the later stages [59]. In addition, PGC at the blastoderm stage might be more 465 vulnerable to manipulation than those at the later stages. These two latter possibilities can also 466 account for the differential gene targeting ability of PGC in the blastoderm. Furthermore, the 467 increase in the EGFP mRNA expression could be related to other testicular cells as the efficiency 468 of GT is affected by the efficiency of gene targeting in PGC. Another possibility is that although 469 the PGC in the blastoderm were targeted successfully by the viral vectors, they were not able to 470 form functional gametes that can produce offspring. Similar results were also observed at the 471 protein level assessed by flow cytometry (Figs 3b and 3c). The percentage of testis samples 23 472 positive for EGFP also indicates that the injection into the blastoderm is not the best route to 473 obtain transgene-expressing testis cells with high repeatability (Fig 3a). The finding of low 474 germline transgenesis observed in this study was consistent with that of the previous studies sub-germinal cavity of blastoderm embryos. Accordingly, we believe that this new strategy has 495 the potential to become a versatile and effective tool for germline gene transfer.