Materials

Recombinant H2 relaxin was generously provided by Corthera Inc. (San Carlos, CA, USA; a subsidiary of Novartis AG, Basel, Switzerland). Human BM-MSCs were commercially purchased from the Tulane Centre for Stem Cell Research and Regenerative Medicine (Tulane University, New Orleans, LA, USA; Website: https://medicine.tulane.edu/stem-cell-research-regenerative-medicine). Prior to any experiments being conducted on this study, these BM-MSCs were previously characterised to display a normal karyotype, form colony-forming unit-fibroblasts (CFU-F), and show multi-lineage differentiation potential [34]. These BM-MSCs were also found to express the cell surface antigens CD73, CD90 and CD105 (confirming that they were MSCs), but lacked expression of CD14, CD19, CD34, CD45 (confirming that they were not of hemopoietic or endothelial origin) and MHC class II human leukocyte antigen–DR isotype (HLA-DR; confirming that they were immunoprivileged) [34]. Perindopril Erbumine was manufactured by MedChem Express (Monmouth, NJ, USA).

Human BM-MSC culture

Human bone marrow-derived mesenchymal stem cells (BM-MSCs) were cultured in α-minimal essential media (α-MEM) supplemented with 16% fetal bovine serum (FBS), 1% L-glutamine (2 mM) and 1% penicillin/streptomycin (100U/ml) (all from Thermo Fisher Scientific, Scoresby, Victoria, Australia) at 37 °C, and were seeded at a density of 60 cells/cm2 in tissue culture flasks (BD Falcon, North Ryde, New South Wales, Australia) with media replaced every 3–4-days until 70–80% confluency was reached. Passages 3–4 (P3-P4) cells were used for transduction.

Lentiviral production

To expand pLV-EF1A (eukaryotic translation elongation factor-1α)-hRLN2-EGFP lentiviral vector DNA (Fig. 1A), ~ 1 mL (75 ng) of vector DNA was mixed with ~ 10uL of one-shot stb13 competent E.coli (C737303; Thermo Fisher Scientific, Scoresby, Victoria, Australia), and incubated for 30 min, before subjected to heat shock in a water bath for 30 s at 42 °C, and incubated on ice immediately following that. Bacteria were then transferred to 500µL SOC medium (15,544,034; Thermo Fisher Scientific) and shaken at 37 °C for 1 h at 225 rpm, before 250µL was spread onto a pre-warmed ampicillin (50µL/mL) LB-amp plate and incubated overnight at 37 °C. On the following day, one colony was picked into 5-7 mL LB-Amp broth to expand in a shaker (for 8 h at 225 rpm, at 37 °C), before being diluted at 1:1000 with 100 mL LB-Amp broth and incubated in 37 °C in a shaker at 250 rpm for 16 h. The next day, plasmid DNA was extracted and purified using a QIAGEN Plasmid Plus Midi Kit (#12,943 and #12,945) following the manufacturer’s protocol. The DNA concentration of the pLV-EF1A-hRLN2-EGFP and pLV-EF1A-EGFP plasmids were 1483 ng/µL and 553 ng/µL, respectively.

Fig. 1

Design, transduction and characterisation of BM-MSCs-eRLX + GFP. A Shows a schematic illustration of the pLV-EF1A-hRLN2-EGFP vector that was designed and transduced into BM-MSCs. The vector comprised bicistronically expressed human relaxin-2 cDNA and EGFP genes linked by internal ribosome entry site (IRES), under an EF1A promoter. B Lentivirus was produced with a three-plasmid expression system with transfer (which contained the RLN-2-EGFP genes or EGFP gene of interest), packaging and envelope plasmids in HEK293T cell lines, before being transduced into passage (P)3–4 BM-MSCs at a MOI of 2. After flow sorting, the successfully transduced cells (GFP+) were expanded and passaged until P6, when they were subjected to flow cytometry analysis of GFP+ cells to test the long-term stability of lentiviral transduction, which represented averages of ~ 77.5% and ~ 86.7% of BM-MSCs transduced with virus containing pLV-EF1A-hRLN2-EGFP or pLV-EF1A-EGFP, respectively. Representative fluorescent microscopy images also confirmed the GFP expression in the P6 BM-MSCs-eRLX + GFP and BM-MSCs-eGFP, respectively. Scale bar = 100 µm

Lentiviral expansion and titre determination

Lentiviral particles containing pLV-EF1A-hRLN2-EGFP or pLV-EF1A-EGFP were produced in HEK293T cells by transient co-transfection using the three-plasmid expression system. Briefly, 17.5 million cells were plated into a T175 flask a day before transfection and 24 h later, the culture medium was replaced by supplemented advanced DMEM containing 2% FBS (Invitrogen, Mount Waverley, Victoria, Australia), 1% non-essential amino acids (NEAA; Sigma-Aldrich, Castle Hill, New South Wales, Australia), 1% GlutaMax (2 mM; Invitrogen, Australia) and 1% penicillin/streptomycin (100U/ml) prior to the addition of transfection mix containing transfer vector DNA (with human relaxin-2 gene), packaging plasmid DNA (psPAX2; Addgene Plasmid #12,260) and envelope plasmid DNA (pMD2.G; Addgene plasmid #12,259) at a 2.5:1.6:1 ratio in 1.25 ml of sterile H20 per T175 flask, which contained 4.5 µg PEI/µg total DNA. The medium was replaced after 16–24 h (on day-2) and again after 48 h (on day-3) post-transfection. On day-3, the conditioned medium containing the lentiviral particles was collected into sterile 50 mL tubes. To increase the viral titer, the cell culture supernatant containing lentiviral particles, at 72 h post-transfection, was filtered through a 0.45 µm stericup-HV sterile vacuum filtration system (SCHVU01RE; Millipore Merck, Darmstadt, Germany) and transferred to Millipore concentrators with a 100 kDa cut-off (Amicon®Ultra-15, Merck Millipore, Tullagreen, Cork, Ireland) and centrifuged repeatedly for 20–30 min at 4000 rpm at 4 °C until the sample of each supernatant was concentrated to ~ 200μL. The concentrated samples containing lentiviral particles were finally transferred to 0.2 ml Eppendorf tubes and stored at -80 oC until required.

To determine the virus titre (the number of host cells per unit virus can infect), cryopreserved BM-MSCs were seeded in 12-well plates at a density of 5 × 104 cells per well. After incubation at 37 °C for 24 h, the culture medium was replaced with 0.5 mL of virus particles diluted by 1000-, 10,000- and 100,000-fold in α-MEM supplemented with 2% FBS, 1% L-glutamine (2 mM) and 1% penicillin/streptomycin (100U/ml), in the presence of polybrene transfection reagent (1:1700; TR-1003-G; EMD Millipore, Merck KGaA, Darmstadt, Germany) in duplicate. Notably, cells that were transduced with pLV-EF1A-hRLN2-EGFP and pLV-EF1A-EGFP were termed BM-MSCs-eRLX + GFP and BM-MSCs-eGFP (as a control engineered cell type), respectively. The culture medium was replaced with 1 mL of MSC medium every 24 h. At 72 h post-transduction, transduction efficiency was measured by detecting % GFP-positive (GFP+) cells on a LSR Fortessa flow cytometer at a concentration of 2.5 × 105 cells/mL in FACS buffer (0.5% BSA, 0.5% 0.5 M EDTA in PBS) with propidium iodidie (PI; 1:500; P4864; Sigma-Aldrich). The transduction efficiency (F) was then used to calculate the virus titer (infectious units per µl of virus), by multiplying the number of targets MSCs (Cn), and dividing that by the volume of input virus (V).

BM-MSC transduction

Cryopreserved human BM-MSCs were plated at a density of 5 × 104 cells per well in a 12-well plate. According to the virus titre calculated using the formula described above (1.58 × 107TU/ml and 1.5 × 108TU/ml for Lentiviral particles containing pLV-EF1A-hRLN2-EGFP or pLV-EF1A-EGFP, respectively), transduction was performed with a multiplicity of infection (MOI; the ratio of infectious virions to cells) of 2, for both cell lines, at a final volume of 0.5 ml per well using the same method described above. Cells were sorted on the BD influx system flow cytometer at 72 h post-transduction. The GFP+ cells from the cultured BM-MSCs-eRLX + GFP and BM-MSCs-eGFP, respectively were collected and cultured until 80% confluency was reached (after ~ 7–10-days), before they were used for assessment of cell proliferation, cell differentiation, measurement of RLX levels using the human relaxin-2 ELISA or passaged for expansion. For long-term storage, cells were trypsinized, pelleted and resuspended in a freezing medium containing 30% FBS, 1% penicillin/streptomycin, and 5% dimethylsulfoxide (DMSO; Sigma-Aldrich, St Louis, MO, USA), frozen at -1 °C/min until reaching -80 °C, before being stored in liquid nitrogen. To ensure the stable transduction of both lentiviral vectors in long-term culture, flow cytometry was performed on a LSR Fortessa flow cytometer after 2 passages of flow sorting (P6) to quantify % GFP+ cells within BM-MSCs-eRLX + GFP and BM-MSCs-eGFP populations again, which represented ~ 77.5% and ~ 86.7% of each cell population, respectively (Fig. 1B), confirming the long-term incorporation of the vector construct in BM-MSCs using the transduction method used.

RXFP1 staining of BM-MSCs-eRLX + GFP

RXFP1 expression on BM-MSCs-eGFP (as control transduced cells) and BM-MSCs-eRLX + GFP was determined by immunofluorescence staining. Briefly, 4 × 104 cells, respectively, were cultured on a 12 mm glass coverslip (pre-coated with 0.1% gelatin) in a 24-well plate. Following overnight incubation at 37 °C, cells were fixed in 4% PFA for 10 min at room temperature, washed in PBS for 5 min (3 times) and blocked for non-specific protein binding in 2% Bovine serum albumin (BSA) for 1 h at room temperature. The fixed cells were then washed again in PBS for 5 min (3 times) and incubated with a rabbit polyclonal RXFP1 antibody (A9227 [35]; 1:2000 dilution; Immunodiagnostik AG, Bensheim, Germany) in 2% BSA overnight at 4 °C, whilst the negative control was incubated only with 2% BSA. On the second day, the primary antibody was washed with PBS, before cells were incubated with goat anti-rabbit Alexa-fluor 555 secondary antibody (A21428; 1:500; Invitrogen, Scoresby, Victoria, Australia) in 2% BSA for 1 h at room temperature, before washed 3 times with PBS. Nuclear counterstaining was performed by adding ~ 10ul of VECTASHEILD@ Antifade Mounting Medium (With DAPI; H-1200, Vector Laboratories) prior to coverslipping. Immunoreactive RXFP1 (red staining) was visualised at × 200 magnification using the Leica AF6000LX microscope, in association with the GFP+ cells (green) and DAPI (blue)-stained cells, and merged using ImageJ software.

Proliferation assay

To assess whether viral transduction of BM-MSCs affected cell viability and proliferative capacity, BM-MSCs were seeded into 96-well plates at 5 × 103 cells per well containing 100 µl of culture media, or media supplemented with RLX at 1 ng/ml or 10 ng/ml (in triplicate), and cultured at 37 °C, for 3- or 7-days. Separate sets of triplicate wells were used to culture BM-MSCs-eRLX + GFP or BM-MSCs-eGFP under the same condition. After 3- and 7- days of culture, cell proliferation was determined using the CellTiter 96 AQueous One Solution Proliferation Assay (Promega, Fitchburg, WI, USA) according to the manufacturer’s instructions.

Differentiation assay

To assess whether RLX-expressing BM-MSCs had the multipotency of naive BM-MSCs, the adipogenic, osteogenic and chondrogenic potential of BM-MSCs-eRLX + GFP were assessed using the Human Mesenchymal Stem Cell Functional Identification Kit (SC006; R&D Systems, Minneapolis, MN, USA) at passage 6 (P6), as per the manufacturer's instructions. Adipocytes, osteocytes and chondrocytes were identified via positive histological staining with Oil-red-O, Alizarin red and alcian blue, respectively; or via immunohistochemical staining for fatty acid binding protein 4 (FABP4), osteocalcin and aggrecan, respectively. Adipocytes and osteocytes that were subjected to histological and immunofluorescent staining were imaged at × 200 magnification using an Olympus fluorescent microscope and CellSens Software (version 1.15; Olympus, Bartlett, TN, USA), whereas the chondrocytes pellets were imaged with a fluorescent microscope (Provis AX70; Olympus).

Migration assay

The migration capacity of naïve BM-MSCs, in the absence of presence of RLX (10 ng/ml), as well as BM-MSCs-eRLX + GFP or BM-MSCs-eGFP was determined over a 24 h (hr) period. To perform the assay, 2 × 104 cells were suspended in 70µL growth media and seeded into 2 well-silicone in 35 mm dishes for wound healing assay (Ibidi Inc., Fitchburg, WI, USA) and incubated at 37 °C for 24 h to form a confluent cell layer. Artificial wounds within the dishes were then created at a gap distance of 500 ± 100 µm [35]. The dishes were replenished with 2 mL of MSC growth media supplemented with various treatments of interest. The number of cells that migrated to close the wound was assessed by using light microscopy. Photomicrographs of the migrated cells were captured at 0-, 2-, 4-, 6-, 12- and 24-h post-treatment. The percentage of gap closure area was determined using ImageJ software (NIH, Bethesda, MD, USA). The slope of the linear phase was used to characterise the average wound closure rate and was expressed as a % of that measured from the migration capacity of naïve BM-MSC controls.

Secretion profile of BM-MSCs-eRLX + GFP compared to naïve BM-MSCs

To characterise the secretion profile BM-MSCs-eRLX + GFP, a human cytokine array was performed using a Proteome Profiler Human XL Cytokine Array Kit (ARY022B; R&D Systems), from the 7-day cell culture supernatant of BM-MSCs-eRLX + GFP and naïve BM-MSCs (from n = 4 separate supernatant samples per cell type). Briefly, the nitrocellulose membranes (containing capture and control antibodies spotted in duplicate) were blocked for 1 h, before incubating with 400µL of cell culture supernates (from 0.5 × 106 cells of each type) overnight at 4 °C on a rocking platform shaker. The membrane was then incubated for 1 h with the detection antibody cocktail, washed and incubated with streptavidin-HRP. Finally, the Chemi Reagent mix was spread on the membrane to obtain an autoradiograph, and the duplicate dots were quantified by densitometry using a ChemiDoc MP imaging System (Bio-Rad; South Granville, New South Wales, Australia). The OD of each duplicate dot (representing one cytokine/chemokine; see mean ± SEM OD of the cytokine/chemokine intensity detected in STable 1 [Supplementary File]) from BM-MSCs-eRLX + GFP was corrected for that of the reference levels for each membrane and expressed relative to the corresponding values from naïve BM-MSCs (% change compared to that secreted by naïve BM-MSCs alone), which was expressed as 0 in each case. Only the % change from the detectable cytokines are presented in Fig. 2F.

Fig. 2

Proliferation, migration and cytokine-secretion capacity of BM-MSCs-eRLX + GFP. A RXFP1 (red staining) was detected on the cell surface of BM-MSCs-eRLX + GFP; which was colocalised with DAPI-nuclear (blue) staining. A negative control image is also provided in which the cells shown were not stained with the primary RXFP1 antibody used. B Representative images show the trilineage differentiation ability of BM-MSCs-eRLX + GFP (as per their naïve/unmodified counterparts) to differentiate in vitro into adipocytes, osteocytes and chondrocytes, as confirmed with histological and immunohistochemical staining (for adipocytes using oil-red O and fatty acid binding protein 4 (FABP4) staining; osteocytes using alizarin red and osteocalcin staining; and chondrocytes using alcian blue and aggrecan staining) colocalized with DAPI-staining. Scale bar = 2 mm or 100 µm as indicated. C The mean ± SEM 3-day and 7-day proliferation capacity of BM-MSCs alone or treated with 1 ng/ml or 10 ng/ml RLX vs BM-MSCs-eRLX + GFP or BM-MSCs-eGFP, and D 24-h migration capacity of these cells (expressed relative to that of the untreated naïve BM-MSCs). *p < 0.05 vs the (unstimulated) BM-MSCs group; as determined using a one-way ANOVA and Tukey’s post-hoc test. E The ELISA-determined levels of H2 RLX that were produced by naïve BM-MSCs (n = 4) vs BM-MSCs-eRLX + GFP (n = 4) that were cultured for 3- and 7-days in vitro are also shown. F Additionally, 30 cytokines that were measurably altered (up- or down-regulated; as determined by a human cytokine array with antibodies to 105 cytokines) in the cell supernatant of BM-MSCs-eRLX + GFP that were cultured for a 7-day period are shown, which were expressed as the % change relative to their naïve BM-MSCs counterparts. Data are presented from n = 4 separate experiments, and cytokines that play a role in immunomodulation, angiogenesis (tissue repair) and ECM remodeling/cell proliferation are grouped and highlighted. Data points between 200 and 15,000 are not shown due to the break in the y-axis. Abbreviations: MCP, monocyte chemoattract protein; IL, interleukin; VCAM-1, vascular adhesion molecule-1; MIF, macrophage migration inhibitory factor; VEGF, vascular endothelial growth factor; IGFBP, insulin growth factor binding protein; uPAR, urokinase-type plasminogen activator receptor; PDGF-AA, platelet-derived growth factor containing two A subunits; BDNF, brain-derived neurotrophic factor; FGF, fibroblast growth factor; SDF, stromal cell-derived factor; GDF, growth differentiation factor; DKK1, Dickkop Wnt signaling pathway inhibitor 1

Exosome extraction from BM-MSCs-eRLX + GFP

P6 BM-MSCs-eRLX + GFP were grown to approximately 80% confluency and starved in DMEM supplemented with 1% FBS and 1% penicillin/streptomycin for 72 h. Media was collected and clarified at 2000 g for 30 min, followed by 0.45 μm filtration. Clarified media was ultracentrifuged (Sorvall WX 100 + Ultracentrifuge; Beckman Coulter Type 70.1 Ti fixed-angle rotor) at 100,000 g for 1 h at 4 °C (Beckman Coulter Australia, Mount Waverley, Victoria, Australia). The top 9 ml of each tube was discarded, the bottom 1 mL was pooled and spun again using the same parameters, and the rest (conditioned medium) was stored at -80 °C. Media was aspirated, and the sides of the tubes were washed thoroughly with 500μL of PBS. Particle size and yields were determined using the Zetaview® PMX-120 Nanoparticle Tracking Analyzer (NTA) (Laser = 488 nm, Sensitivity = 80 AU, Shutter Speed = 100 s−1, Frame Rate = 30/s). Analysis was conducted on a 1:100 dilution of isolated exosome stock (Max Area = 1000px, Min Area = 10px, Max Brightness = 30). A flowchart depicting the process of exosome extraction and detailing the size of the exosomes extracted is shown in Figure S1 [Supplementary File]. The extracted exosomes were lysed with RIPA Buffer (#9806; 10x) containing proteinase inhibitor (#5872; 100x) and PMSF (#8553; 200x; all from Cell Signaling Technology; Danvers, MA, USA) for measurement of H2 RLX levels.

Animals

All animal care and experimental procedures performed adhered to the NIH guidelines for the Care and Use of Laboratory Animals for Scientific Purposes, and were approved by the Animal Ethics Committee of Monash University. The work has been reported in line with the ARRIVE guidelines 2.0. The sex of animals was considered in the study design: given that male mice are more prone to undergo a more rapid rate of kidney disease progression and hypertensive renal injury, in line with hypertension and CKD being more common in men compared to women [36], male mice were used since they provided a larger therapeutic window in which the efficacy of cell-based treatments evaluated could be measured. Initially, groups of equal size were designed using randomization and blinded analysis; although confounding variables were not controlled. 6-week-old male C57BL/6 J mice were obtained from Monash Animal Services (Monash University, Clayton, Victoria, Australia) to compare the renoprotective effects and biodistribution of BM-MSCs-eGFP versus BM-MSCs-eRLX + GFP in a model ischemia reperfusion injury (IRI; under animal ethics number: MARP/2022/29417). Mice subjected to IRI were anaesthetised with inhaled isoflurane (2–3% in oxygen; Baxter Healthcare) and provided analgesia (carprofen; s.c) just prior to surgery, and 24 h and 48 h post-surgery. 11–12-week-old male Balb/c mice were obtained from Monash Animal Services (under animal ethics number: MARP/2021/30019) to compare the therapeutic effects, biodistribution and safety of BM-MSCs-eGFP versus BM-MSCs-eRLX + GFP in a model of high salt (HS)-induced hypertensive nephropathy; which did not require any anaesthesia or analgesia. In the latter model, the therapeutic effects of BM-MSCs-eRLX + GFP were also compared to, or combined with, the current standard of care medication, perindopril. All mice were housed (in groups of four per cage) under a controlled environment, given a 5- to 6-day acclimatisation period, and maintained on a 12-h light/12-h dark cycle with free access to normal rodent lab chow (Barastock Stockfeeds, Pakenham, Victoria, Australia) and water.

In each case, power calculations were performed to ensure that adequate group sizes were used for the studies conducted; where it was determined that with a 20–25% standard deviation, we would be 80% powered to detect a 23–28% effect with n = 8 animals per group. In either (the IRI or HS) model established, the effects of RLX alone was not assessed, since our previous work had already determined that the combined effects of RLX and BM-MSCs provide broader renoprotection over that of RLX alone in normotensive [29] and hypertensive [28] models of kidney disease.

Induction and treatment of a murine model of ischemia–reperfusion injury (IRI)

To assess the therapeutic effects of BM-MSCs-eRLX + GFP as a treatment for AKI, a unilateral IRI-induced model of AKI was established in 6-week-old male C57BL/6 J mice (n = 32), weighing approximately 25-30 g. Mice were initially anaesthetised with 4% isoflurane in an induction chamber, and then maintained with 1–2% isoflurane via a nasal cone for the remainder of the surgery. Using a vascular clamp (0.4–1.0 mm; Fine Science Tools, Heidelberg, Germany), the renal pedicle (which included the renal artery and vein) was identified and clamped for 40 min. After 40 min, the clamp was removed and during this reperfusion stage, subgroups of mice were either i) left untreated (IRI group; n = 8) or given an intra-renal injection of ii) 1 × 106 BM-MSCs-eGFP (n = 8) or iii) 1 × 106 BM-MSCs-eRLX + GFP cells (n = 8) (all suspended in 200 µl PBS). iv) To compare the renoprotective effects of BM-MSCs-eRLX + GFP cells to that of BM-MSCs and RLX administered separately, a separate group of mice (n = 8) had a small dorsal incision made during the ischemic period, in order to implant 7-day osmotic mini pumps (Model 1007D; with a release rate of 0.5 µλ/hour; ALZET, Cupertino, California, USA) containing RLX (Pump-RLX; 0.5 mg/kg/day). This dose of RLX had previously been found to produce circulating RLX levels of ~ 18-20 ng/ml in mice after 5-days [37]. During the reperfusion stage, these mice were injected with 1 × 106 BM-MSCs (suspended in 200 µl PBS). v) A sham-operated control group (n = 8) was also included with mice being anaesthetised and left flank incision made to expose the kidney, but the kidney remained untouched. The four groups of mice were maintained until 7-days post-sham or IRI surgery, before being killed for blood and kidney tissue collection and analysis.

Additionally, the in vivo tracking of transplanted BM-MSCs-eGFP and BM-MSCs-eRLX + GFP was conducted on a separate cohort of IRI mice (n = 24), which were either subjected to IRI alone (n = 2 per time point) or were subjected to IRI and received a single i.v-injection of BM-MSCs-eGFP (n = 3 per time point) or BM-MSCs-eRLX + GFP (n = 3 per time point). As our previous work had determined that i.v-administered BM-MSCs-eGFP immediately migrated to the lungs of IRI mice, but then homed to the injured kidney after 24 h and remained in the kidney after 72 h [34], IRI mice treated with BM-MSCs-eGFP or BM-MSCs-eRLX + GFP were then killed at 5-, 7- and 14-days post-transplantation for the analysis of the biodistribution of cells, and quantification of GFP+ cells within the kidney. At the appropriate time-points, mice were killed with an overdose of inhaled isoflurane (5% in oxygen) followed by cardiac puncture and exsanguination, so that blood and appropriate organs could be isolated for analysis.

Induction and treatment of a murine model of high salt (HS)-induced hypertensive kidney disease

To assess the therapeutic effects of BM-MSCs-eRLX + GFP as a treatment for hypertensive CKD, a HS (2% NaCl in drinking water)-induced model of hypertension was established in 11–12-week-old male Balb/C mice (n = 40; weighing 20-25 g) for 8-weeks; which was adapted from a previously published study [38]. Balb/c mice exhibited hypertension, kidney inflammation and fibrosis after 6-weeks of 2% NaCl drinking [38]. We therefore employed an 8-week model of HS loading, from which sub-groups of HS-fed mice were treated during the final two-weeks to assess the therapeutic impact of various treatments in mitigating the established salt-sensitive hypertension and related renal pathologies. The i) control group (n = 8) was maintained on normal drinking water (NDW) for the 8-week period. At week-7 post-HS loading, subgroups of the n = 48 HS-fed mice were randomised to either receive ii) no treatment (HS group; n = 8); or an intravenous (i.v; tail vein) injection of iii) 1 × 106 BM-MSCs-eGFP (and then again on week-8 (× 2); n = 8); iv) BM-MSCs-eRLX + GFP (× 1; n = 8); v) BM-MSC-eRLX + GFP (× 2; then again on week-8; n = 8); vi) Pump-RLX delivered via an s.c. implanted 14-day osmotic minipump (0.5 mg/kg/day; Model 2002; with a release rate of 0.5 µl/hour; ALZET) and i.v. injection of BM-MSCs (1 × 106/mouse; and then again on week-8) (Pump-RLX + BM-MSCs; n = 8); or vii) the angiotensin-converting enzyme inhibitor, perindopril (2 mg/kg/day via drinking water; a dose that induced anti-hypertensive effects without inducing weight loss [28, 39]; n = 8) via drinking water.

In a separate study that assessed the renoprotective effects of BM-MSCs-eRLX + GFP (× 2) in the presence of perindopril treatment, control groups of mice that were maintained on NDW (n = 8) or HS (n = 24; n = 8 subjected to HS alone) for 8-weeks were again established. Additional sub-groups of HS-fed mice were then subjected to perindopril alone (as above; n = 8); or perindopril + BM-MSCs-eRLX + GFP (× 2; n = 8). In both studies, all treatments were maintained for two-weeks, from weeks-7–8 post-HS, before all groups of mice were killed by an overdose of inhaled isoflurane (5% in oxygen) followed by cardiac puncture and exsanguination, for blood, heart and kidney tissue collection and analysis.

Safety study

To assess the long-term safety of BM-MSC-eRLX + GFP (× 2) treatment, a further set of HS-fed (n = 9) mice that were either untreated (n = 3) or treated with BM-MSC-eRLX + GFP (× 2; n = 6) were established and treated as outlined above; whilst NDW-fed mice (n = 3) were also included as controls. All mice were maintained on NDW or HS drinking for 8-weeks (as above), but were then maintained on NDW for the further 9-months (39-weeks) after treatment cessation (to eliminate the long-term effects of HS-drinking as a confounding variable). At the completion of this period, when mice were 59-weeks of age, mice were transported to an independent Pathology company (Cerberus Sciences, Scoresby, Victoria, Australia) for histopathological assessment of various organs. The kidneys, adrenal glands, heart, aorta, lungs, thyroid gland, liver, spleen, stomach, pancreas, mesenteric lymph node, gall bladder, brain and skin of all mice were histologically assessed for abnormalities. Interstitial kidney fibrosis was also measured from the four groups of mice that survived the 9-month treatment cessation period; by Masson’s trichrome staining of interstitial ECM deposition.

SBP measurements

Systolic blood pressure (SBP) measurements of all NDW and HS-fed mice were measured via tail cuff plethysmography, using the multichannel MC4000 Blood Pressure Analysis System (Hatteras Instrument Inc; Cary, NC, USA), at baseline and then fortnightly over the 8-week experimental period; and monthly over the 9-month treatment cessation period (for the safety study). At least 20 measurements per time point were pooled to obtain a mean value for each animal.

Confirmation of BM-MSC homing to the kidney

To determine if the i.v-injected engineered cells could be detected in the kidneys and other organs of IRI or HS-fed mice, the left kidney pole, heart, lungs, spleen and liver of IRI mice; and left kidney pole and heart of HS-fed mice that were treated with BM-MSC-eGFP or BM-MSC-eRLX + GFP were subjected to genomic DNA (gDNA) extraction using a DNeasy Blood&Tissue kit (Qiagen, Clayton, Victoria, Australia) following the manufacturer's protocol. The extracted gDNA was then used for the analysis of EGFP gene expression by polymerase chain reaction (PCR). gDNA from 1 × 106 BM-MSCs-eGFP that were not injected into mice was used as a positive control. On the other hand, the replacement of gDNA with distilled water was included as a negative control. The PCR mixture contained the following components, including 10µL GoTaq Green Master Mix (Ref: M712B; Promega; Madison, WI, USA), 1 µL forward primer: 5’-TAC GGC AAG CTG ACC CTG AAG TTC-3’and 1 µL reverse primer: 5’CGT CGT CCT TGA AGA AGA TGG TGC G-3’ primer pairs (Integrated DNA Technologies, Singapore), 1 µL DNA template and 7 µL of Nuclease-free water. PCR mixtures were run under the following conditions: 94 °C for 2 min, 35 cycles (of denaturing for 30 s at 94 °C, annealing for 30 s at 60 °C, and extending for 1 min at 72 °C), then 72 °C for 7 min, before samples were held at 10 °C before collection. PCR products were then separated by electrophoresis on a 2% (w/v) agarose gel, where the primer sequences used were expected to produce a 196 base-pair product. Notably, 100 bp or 50 bp DNA ladders were used for gels containing DNA samples extracted from IRI or HS-fed mice, respectively. Gels were imaged with a ChemiDoc MP imaging System (Bio-Rad; South Granville, New South Wales, Australia).