Animals and tissue collection
Donors, recipients, surgical controls and unoperated female poults from a broad breasted white female parent-stock line were supplied by Hybrid Turkeys (Kitchener, Ontario Canada) at 1-day posthatch (dph). All animals were housed in the University of Guelph’s Animal Care Facility, and all protocols were approved by the University of Guelph’s Animal Care Committee (Animal Utilization Protocol # 3921). Donor poults were given food and water ad libitum and housed in groups on wood shavings. Donor poults were euthanized via manual cervical dislocation at 8 ± 1 dph. After euthanasia, the donor poult’s ovary was removed and dissected into 2 ´ 2 mm pieces and held in holding media (PBS with 20% Fetal Bovine Serum) at room temperature for no longer than 20 mins before being transplanted. The optimal age of donor ovarian tissue was chosen based on a previous study, with this age resulting in better germ cell and follicle morphology post-surgery [8]. Recipients and surgical controls were given only water and were kept in groups without shavings prior to surgery at 1 or 2 dph. Fasting ensured that the digestive track did not obscure the view of the ovary during surgery. Since poults still possess and rely upon the remnant yolk sac for nutrition at that age, fasting is not an issue. After surgery, recipients and surgical controls were immediately offered food and water ad libitum and were housed in the same manner as donors. An unoperated group served as controls and were treated the same as the recipient and surgical controls except they did not undergo surgery. Recipients, surgical controls and unoperated poults were all euthanized via manual cervical dislocation at 7 or 8 dph, 6 days post-surgery in the case of the recipients and surgical controls.
Experimental groups
Four surgical procedures were compared, and all involved allotransplants of fresh tissue placed orthotopically. In the first three groups, either 33%, 66% or 100% of the left ovary was removed before receiving the donor ovarian tissue (n=8/group). In these groups, the dorsal portion of the left abdominal air sac membrane was fully opened, thus preventing it from covering or serving as an anchor for the transplants. In the fourth group, in addition to 100% ovary removal, the dorsal portion of the left abdominal air sac membrane was left partially intact over the transplants (n=6 poults). Surgical controls had varying portions of the ovary removed but did not receive transplants (n=2/group). Lastly, unoperated poults (n=8) served as an additional control group.
Transplantation of ovarian tissue
Orthotopic transplantation of ovarian tissue between breeds has been reported in chicken, quail, and turkeys [9,11,13], while inter-species transplantation has also been reported for ducks [14]. Based on these previous studies, a modified procedure was used here. Firstly, recipient poults 1 to 2 dph were moved to an intensive care unit (ICU; TLC-40, Brinsea Products Ltd., Florida, USA), set at 37.5 oC, to maintain body temperature before surgery. The poults were then weighed and given an intramuscular injection of butorphanol (2 mg/kg bodyweight [BW]) and returned to the ICU for 10 mins. An intramuscular injection of xylazine (2 mg/kg BW) was then administered. The poult was then placed in a supine position on a silicone pad on top of a heating plate (15x10 cm, World Precision Instruments, Florida, USA), which was connected to an ATC2000 small animal temperature controller (World Precision Instruments) fitted with a mouse rectal probe (RET-3, World Precision Instruments) which was inserted into the cloaca of the bird. This equipment monitored poult temperature and adjusted the heating plate as required. A rodent facemask, connected to a Bain circuit, was then placed over the beak, with the oxygen flow rate set at 0.8 L/min. For induction, the isoflurane vapor was set at 1.5%, and the anesthesia was maintained with the setting of the vaporiser ranging between 1 and 4%. The respiration and heart rate, and rectal temperature were monitored and recorded manually. Before and during surgery, the pedal reflex response of pinching the skin between the toes was used to assess the depth of anesthesia. Once the poult was fully anesthetized, feathers on the left side of the abdomen were shaved with a small electric shaver (Fig. 1A). The skin was then cleaned with Germi-stat gel (Cardinal Health Canada, Ontario, Canada) followed by 70% ethanol. A 2 cm incision across the chick’s abdomen (Fig. 1B) just below the ribs was then made using micro scissors (Vannas 8 cm with 5 mm blade, World Precision Instruments, Florida, USA). First, only the skin, rectus, and oblique abdominal muscles were cut, then, the peritoneal membrane was opened exposing the abdominal viscera (Fig. 1C). The yolk sac, which was located caudally within the abdominal cavity, was then gently pulled out to prevent rupturing (Fig. 1D). The connection between the yolk sac and intestine was ligated and then cut to remove the yolk sac. The abdominal viscera were then gently pushed down into the cavity the yolk sac previously occupied, allowing access to the ovary. At this point, a pair of front-lens-mounted PrismPro 8.0x surgical loupes (EVK 800, Surgitel, Michigan, USA) with a high intensity mini-LED mounted headlight were used to provide sufficient illumination and magnification. The ovary was reached by first pushing back the left hepatic lobe and gizzard, to expose the ventral aspect of the abdominal air sac (Fig. 1F). The air sac was opened via blunt dissection using micro forceps, exposing the dorsal portion of the membrane overlaying the ovary (Fig. 1G). For surgery groups 1 to 3, this part of the abdominal air sac was opened via blunt dissection again to expose the ovary (Fig. 1I). For surgery group 4, the dorsal part of the air sac was only partially opened below the ovary, keeping the portion overlaying the ovary intact. Varying portions of the ovary were then removed using micro forceps (Dumont 90o bend, World Precision Instruments, Florida, USA). Poults in group 1 had 33% of the left ovary removed, along its right side, whereas group 2 had 66% of their left ovary removed along its right and left side (Fig. 2A-B), and groups 3 and 4 had all their left ovary removed. In all groups two pieces of donor tissue were positioned orthotopically (Fig. 1L). For group 4, the transplants were covered using the abdominal air sac membrane. The abdominal viscera were then gently repositioned over the transplants, and the abdominal muscles and skin were sutured closed (4-0 Monocryl PC-5 19 mm 3/8c Multipass, Ethicon, New Jersey, USA). A simple interrupted pattern was used with sutures spaced ~ 2 mm apart. At the end of surgery, poults were given an intramuscular injection of ceftiofur hydrochloride (5 mg/bird, EXCENEL® RTU), and a neck tag for identification purposes. The poults were then moved back into the intensive care unit, to recover from the anesthesia. Once recovered, poults were given meloxicam via crop gavage (0.5 mg/bird, Metacam® Oral Suspension) for 3 days post-surgery to mitigate pain. Poults were weighed daily to confirm that they were eating and growing.
Transplant processing, staining, and imaging
Upon euthanasia of the recipients, surgical controls, and unoperated birds, the abdomen was opened and photographed using a dissecting microscope equipped with a camera (Leica EZ4 W; Optic-Tech Scientific Inc, Ontario, Canada). The recipient’s ovary and surrounding tissue were then removed and fixed in Bouin solution for 1hr at RT, followed by an additional 23 hr at 4 oC [15]. After fixation, the ovary was washed with PBS and transferred to a tissue cassette before being stored in 70% ethanol at 4 oC until processing. The ovary was then dehydrated and cleared with xylene and embedded in paraffin blocks as previously described for whole ovaries [15]. Transplants and recipient ovarian tissue were positioned in the paraffin blocks so that sectioning occurred parallel to the sagittal plane of the ovary. Ovaries were sectioned at a thickness of 5 µm using a Finesse ME microtome (ThermoShandon, Cheshire, UK). Every 8th section was collected through the transplants and recipient ovarian tissue, with five sections collected per slide and approximately 10 slides collected per ovary. Slides were stained with hematoxylin and eosin (H&E) to allow examination of the entire specimen and ensured that small amounts of recipient tissue or transplants were not overlooked. During sectioning, two additional slides were created, each with five randomly collected sections. These two slides were used for immunohistochemistry analyses. The H&E sections were imaged using a Leica DM 5000B light microscope (Leica, Wetzlar, Germany) equipped with a B-Series LED light source (ScopeLED, Richmond, CA, USA) for sequential red/green/blue imaging. Images were captured using a Hamamatsu Orca-Flash 4 camera (Hamamatsu Photonics, Hamamatsu City, Japan).
Immunohistochemistry for CD3 and MUM-1.
Immunohistochemistry for CD3 and MUM-1 antigens was performed at the University of Guelph Animal Health Laboratory (AHL), an American Association of Veterinary Laboratory Diagnosticians (AAVLD) accredited laboratory, using an automated staining instrument (Dako autostainer, Dako / Agilent, Ontario, Canada). Following manual deparaffinization and rehydration, the sections were treated with 3% hydrogen peroxide to quench endogenous peroxidase activity. For CD3, heat-induced epitope retrieval (HIER) was carried out using citrate buffer (pH 6) and a decloaking chamber (BioCare Medical, Concord, CA, USA). For MUM-1, HIER was performed using an EDTA buffer (pH 9) and a similar pressure cooker device (PT Link, Dako). Primary antibodies against CD3 (rabbit polyclonal, Dako) and MUM-1 (mouse monoclonal, clone MUM1p, Dako) were diluted 1:100 and 1:50, respectively, and incubated at room temperature (RT) for 30 mins. Slides were then incubated with a dual anti-mouse / anti-rabbit anti-IgG horseradish peroxidase–linked polymer (EnVision for CD3, EnVision FLEX for MUM-1, Dako) for 30 mins at RT, and reactions were visualized using Nova Red chromogen (Vector Laboratories, Burlington, ON, Canada). Slides were counterstained with hematoxylin.
Tissue known to have high levels of T cells (thymus), and B cells (bursa of Fabricius) were used as positive controls for CD3 and MUM-1 IHC, respectively. These tissues were collected, fixed, and sectioned in the same way as for the transplants, although only a couple sections were collected, as needed. For negative controls, non-immune rabbit serum was substituted for anti-CD3 polyclonal antisera, and antibody diluent was substituted for MUM-1 antibody. Sections were imaged using the same microscope and camera setup as for H&E slides.
Identification, size, and pathological analyses of recipient and transplanted ovarian tissues
Remnant recipient ovarian tissue and transplants were first identified by comparing the surgical control birds to those that received donor ovarian tissue, using the gross dissecting microscope images. The size, shape, position, and color of the surgical control ovaries, remnant recipient tissue and transplants were all taken into consideration during the identification process. For birds that received donor tissue, ovarian tissue which was inconsistent with the ones from the surgical control birds were identified as possible transplants. Ovarian tissue which was consistent with the ones from the surgical control birds was classified as being remnant recipient ovarian tissue. Histological analysis of the potential grafts and remnant recipient tissue, using the H&E slides, was carried out to confirm their identity. To be confirmed as remnant ovarian tissue, the tissue required attachment via the ovary ligament to the abdominal cavity. With the medulla possessing normal lacunar channels, and a well-defined medulla-cortex border, with the cortex having a mixture of prefollicular germ cells, and primordial follicles [15]. Conversely, transplants were classified as tissue with or without prefollicular germ cells and primordial follicles, that also lacked attachment via the ovarian ligament. Transplants could also be surrounded by fibrous connective tissue, around the entire or a portion of the transplant (sign of transplantation). The surgical control ovary, remnant recipient tissue and transplants were then measured (diameter and height), using the dissecting microscope gross images. The measurements were then used to calculate the area occupied by surgical control ovaries, remnant recipient tissue or transplants.
Percent of positive CD3 and MUM-1 area
The amount of T and B cells infiltrating the grafts was determined to evaluate the immunological status of the transplants. T and B cell infiltration was determined by the amount of cytoplasmic and nuclear immunostained area for CD3 and MUM-1, respectively. This was performed using a computer assisted image analysis tool in ImageJ (Open-source image processing software, 2.0.0-rc-69/1.52p). For each IHC stain, one image from each category (remnant recipient tissue, and transplant) was analyzed per bird. The image was captured at 40x magnification resulting in an area of approximately 105,625 µm2 being analysed. For both IHC tests, the image colors were adjusted via the input levels using photoshop CC 2019 (Version 20, Adobe, California, USA). This was done to achieve a white background, changes were kept consistent with the two different IHC stains, thus improving the contrast between background and the stain. The color channel feature (color deconvolution) was used to separate the blue (Hematoxylin) and brown (Nova Red chromogen) color for each image. For the brown channel, the threshold feature was used to determine the percentage of immunostained (brown) area on each image, as expressed either by CD-3-positive cytoplasm or MUM-1-positive nuclear area, on a particular section. The total nuclear (blue) and cytoplasm (white) area on each image was determined on the blue channel. For CD3, the percentage of immunostained area was divided by the percentage of total cytoplasm to provide a percent of immunostained cytoplasmic area. For MUM-1, the percent of immunostained nuclei area was divided by the percent of total nuclei area, corresponding to the percent of immunostained nuclei area.
Statistical analysis
For bird survival, recipient ovary presences, and size values were combined between the birds that received transplants and their respective surgical controls, for analysis. Statistical analyses were performed using SPSS 25.0 for Mac (SPSS Inc., Chicago, IL). For each experimental group, the average ovarian size (surgical control, remnant recipient tissue or transplant) and percentage of CD3- or MUM-1-immunostained area were presented as means ± standard error of the mean (SEM). Normality and equal variance of the data were evaluated by residual plots and Levene’s tests, respectively. A one-way ANOVA followed by a post-hoc (Tukey) test was used to analyze the effect of surgery group. Significance was set at (P £ 0.05). For poult survival and presence of ovarian tissue or transplants within the surgical controls or recipients, the data was presented in count form with percentages. The data was analyzed via a generalized linear model (binary logistic), to determine if the surgery group affected the survival rate or presence of remnant ovarian tissue or transplants. After an effect (P £ 0.05) was determined, a pair-wise comparison was run to determine the varying groups.