1. Reprogramming of iPSCs and differentiation of pSMCs
Skin biopsies were collected from three female patients (ages 40-70). Dermal fibroblasts were reprogrammed to iPSCs using a modified mRNA/miRNA reprogramming method. mRNA encoding Oct4, Klf4, Sox2, c-Myc, and Lin28 as well as miR-302/367 cluster were transduced [9].
The iPSCs at passage 2 (P2) were differentiated into pSMCs using the feeder-free vascular endothelial progenitor protocol previously described[10]. pSMCs were cultured on human placenta-derived collagen IV (Sigma-Aldrich, Missouri) and expanded in DMEM/F12 HEPES with 5% fetal bovine serum (FBS). pSMCs at P4 were used for injection.
2. Generation of the vaginal-injury rat model and cell injection into the vagina
Healthy female immunodeficient Rowett Nude rats (RNU, Charles River Laboratories, Hollister, CA) weighing 150-200g were used to create the vaginal-injury model as described in our published study [11]. Animals were maintained at the Stanford University Research Animal Facility in accordance with Stanford University’s Institutional Animal Care and Use Committee guidelines. The animals were anesthetized with the intraperitoneal injection of ketamine (30-100mg/kg) and xylazine for the surgery or with inhalation of 1-4% isoflurane for the cell injections.
The animals were euthanized with inhalation of carbon dioxide.
Briefly, an abdominal midline incision was made. Adnexal vascular supply was ligated and ovaries resected. To expose the anterior vagina, the urethra was circumferentially detached from the anterior vaginal wall and pubic bone by sharp dissection. To mimic the vaginal injury caused by prolapse surgeries, two cuts were made in the anterior vaginal muscular layer without penetrating into the vaginal lumen. This area was then clamped with a hemostat for 5 minutes. The abdominal incision was then closed. The rats were checked daily for a week after surgery, and then once a week until euthanasia. Rat cages were housed in the same room and shelf to minimize confounders.
Cell injection was performed at 5 weeks post-surgery to allow for inflammatory changes to subside. In the VSHAM group, we injected saline into both sides of anterior vagina wall. In the cell-injection groups, 2 million cells in 200ul sterile saline were injected (1 million/100ul /each side). Unfortunately, the injection procedure could not be performed in a randomized fashion because the cells for each group needed to be prepared in one batch as cell survival is time sensitive.
The rats were assigned sequentially into five experimental groups as they arrived from Charles River Laboratories. Randomization was not feasible because of the sporadic availability and supply of the RNU rats during the Pandemic. The five experimental groups were: (1) intact controls (control group, n=12). (2) vaginal-injury model injected with saline (VSHAM group, n=33). (3-5) vaginal-injury model injected with pSMCs derived from patient A, B, and C (cell-injection groups A, B, and C, n=14/ group). Patients A and C are in their 40’s and patient B is in her 70’s of age. Rats were monitored for 5 weeks after injection.
Sample size calculation: our primary outcome measure was the vaginal tissue contractile response to KCl stimulation at 40mM. Based on our preliminary data, we estimated the expected effect size to be 0.2 g/cm2 (one times the contraction size of VSHAM) and the standard deviation to be 0.2. This gives us a standard effect size of 1.00 which yields a sample size of 12 rats per group for power of 80% and two-tailed a of 0.10[12]. The higher number of rats in VSHAM group is due to the need for comparisons with sham rats for the multiple assays.
3. In vivo bioluminescence imaging (BLI) of transplanted pSMCs
A human luciferase-tagged iPSC line (Huf-5) was used to differentiate pSMCs. These luciferase-tagged pSMCs were injected bilaterally into the anterior vagina (1 million cells/100uL sterile saline/site) of control (n=4) and vaginal-injury rats (n=4) 5 weeks after surgery to track the human cells. D-luciferin (Biosynth, Itasca, IL) was injected intraperitoneally 375 mg/kg body weight before image acquisition on Day 0, Day 2, Week 1, Week 2, and Week 5. Transplanted cell survival in vivo was monitored via bioluminescence imaging with Lago X (Spectral Instruments Imaging, Tucson, AZ). The photons emitted from luciferase-expressing cells were collected with integration times of 2 minutes.
4. Detection of human gene sequence in harvested tissues from the BLI study using Alu-sequence-PCR Assay
The bladder, urethra, vagina, and uterus were harvested from the rats used for BLI cell tracking after euthanasia on Day 2, Week 1, Week 2, and Week 5. Genomic DNA was extracted from tissue homogenates following the manufacturer’s instruction (QIAGEN, Redwood, CA). The amount of human DNA in each sample was quantified by amplification of a human-specific DNA sequence Alu (Primer sequences are shown in Table 1). Brilliant SYBR Green PCR method was used to perform PCR using AriaMx (Agilent Technologies, Santa Clara, CA). All PCR reactions were performed in triplicate for fifty cycles. pSMC numbers per 25ng of DNA was calculated using a calibration curve with defined numbers of pSMCs.
5. Tissue collection and organ bath myography of the five experimental groups
Vagina, bladder dome, trigone and urethra were harvested 10 weeks post-surgery (5 weeks post-injection). Below is a description of the tissue section used for the different assays.
(1) Vagina
a) Proximal portion: one half was fixed in paraffin and used for H&E, elastin, and h-Caldesmon staining, and the other half was used for RT-qPCR.
b) Middle portion: used for organ bath myography first, followed by ELISA.
c) Distal portion: stored in -80℃ freezer.
(2) Bladder Dome and Trigone
a) Bladder dome and trigone were evaluated with organ bath myography first, followed by histology, RT-qPCR and ELISA.
(3) Urethra
a) Proximal portion: fixed by OCT (optimal cutting temperature) compound and used for elastin staining.
b) Middle portion: RT-qPCR
c) Distal portion: ELISA
The organ bath myography methodology was described in detail in our previous publication[11]. Briefly, the middle portion of the vagina was mounted circumferentially, and the bladder dome and trigone were mounted longitudinally. The contractile responses were monitored using a custom-made isometric force transducer, and signals were recorded using Lab Chart 7 (AD Instrument, Colorado Springs, Colorado). Firstly, we assessed tissue contraction using potassium chloride (KCl) solution: 40mM KCl for vagina and 160mM KCl for bladder tissues. After the tissues reverted to their resting tension by washing with Krebs buffer, we assessed the contraction by using carbachol (Sigma-Aldrich, Missouri), a nonselective muscarinic receptor agonist, in increasing concentrations of 0.625µM, 1.25µM, 2.5µM, 5µM, 10µM, 20µM. Strips were washed again before verifying true carbachol-stimulated response by adding 1µM atropine, and carbachol (20uM) 5 minutes post atropine. Tissues were washed and viability checked again using initial KCl concentrations. Tissue contractility data were normalized to tissue area, expressed as tension per unit of tissue area (g/cm2).
6. Reverse-Transcription Quantitative Polymerase Chain Reaction (RT-qPCR)
RNA extraction and RT-qPCR were performed as in our published manuscript[11]. RT-qPCR was used to evaluate mRNA expression of elastin, collagen I, and collagen III. Primer sequences are shown in Table 1. RT-qPCR was performed in duplicates on Aria Mx Real-Time PCR, using Brilliant SYBR Green PCR Master Mix as described previously[8]. GAPDH was used as an endogenous reference.
7. Enzyme-linked immunosorbent assay (ELISA)
Protein expression of elastin, collagen I, and III was quantified in duplicate with ELISA kits (Lifespan Biosciences, Seattle, WA) as per manufacturer’s instructions. Optical absorbance was measured with a spectrophotometer, SpectraMax M3 (Molecular Devices, San Jose, CA). Quantification of target proteins was calculated based on its standard curve, then normalized to the concentrations of the protein (mg/ml) in the samples.
8. Elastin staining and qualitative scoring of elastin in vaginal and urethral tissues
Proximal vaginal tissues were embedded in paraffin and proximal urethral tissues were embedded in OCT. Elastin staining on paraffin and OCT were performed as described previously [13, 14]. Briefly, elastin fibers were stained in Weigert’s Resorcin-Fuchsin solution (Electron Microscopy Sciences, Hatfield, PA). Cell nuclei were stained with Weigert’s iron hematoxylin working solution (Poly Scientific R&D Corporation, Bay Shore, NY). After washing in running water, the slides were placed in van Gieson’s solution for 3–5 min for collagen fiber staining.
The slides were scored by four people for elastin fiber length (1= short, 2 =moderate, 3 = long), thickness (1 = thin, 2 = moderate, 3 = thick), and density (1 = sparse, 2 = moderate density, 3 = dense). All four examiners were blinded to the group designation.
9. h-Caldesmon (h-CALD) staining and vaginal smooth muscle layer quantification
We performed h-CALD, a marker for SMC’s, immunohistochemical staining to evaluate SMC. The slides of proximal vaginal tissue were deparaffinized, retrieved and blocked. The mouse anti-h-CALD antibody (1:100, Santa Cruz Biotechnology, Santa Cruz, CA) was incubated at 4℃ overnight, and then incubated with the second antibody (horse anti-mouse-biotin,1:50 Themo Fisher Scientific, Waltham, MA). The ABC kit was used to develop the red color (Vector Laboratories, Burlingame, CA).
Because of the surgical procedures, some parts of the vagina were much thinner than other parts, so we measured the thinner and thicker parts separately. The thinner part of the vagina is likely to be the surgically injured area. Six measurements were taken on the thinner part and another 6 taken on the thicker part. Zen software (blue edition, version 3.4, ZEISS, White Plains, NY) was used to measure the thickness. This was performed by two separate people. Thus, the average of 12 measurements (from the examiners) was reported for each part of the vagina (thinner and thicker).
10. Data Analysis
Statistical analysis was performed using JMP software version 17 (SAS Institute, Inc., Cary, NC) The results are expressed as the mean±SD. The nonparametric Wilcoxon test was applied for statistical comparisons between groups. All data points were included in the analyses. Statistical significance was set at p<0.05.
The work has been reported in line with the ARRIVE guidelines 2.0