Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 May 18;11(1):2470.
doi: 10.1038/s41467-020-16065-3.

Cas9-AAV6-engineered human mesenchymal stromal cells improved cutaneous wound healing in diabetic mice

Affiliations

Cas9-AAV6-engineered human mesenchymal stromal cells improved cutaneous wound healing in diabetic mice

Waracharee Srifa et al. Nat Commun. .

Abstract

Human mesenchymal stromal cells (hMSCs) are a promising source for engineered cell-based therapies in which genetic engineering could enhance therapeutic efficacy and install novel cellular functions. Here, we describe an optimized Cas9-AAV6-based genome editing tool platform for site-specific mutagenesis and integration of up to more than 3 kilobases of exogenous DNA in the genome of hMSCs derived from the bone marrow, adipose tissue, and umbilical cord blood without altering their ex vivo characteristics. We generate safe harbor-integrated lines of engineered hMSCs and show that engineered luciferase-expressing hMSCs are transiently active in vivo in wound beds of db/db mice. Moreover, we generate PDGF-BB- and VEGFA-hypersecreting hMSC lines as short-term, local wound healing agents with superior therapeutic efficacy over wildtype hMSCs in the diabetic mouse model without replacing resident cells long-term. This study establishes a precise genetic engineering platform for genetic studies of hMSCs and development of engineered hMSC-based therapies.

PubMed Disclaimer

Conflict of interest statement

M.P. serves on the SAB and has equity in CRISPR Tx and Allogene Tx. Neither company had input into the work described in the manuscript. All other authors declare no competing interests.

Figures

Fig. 1
Fig. 1. The Cas9-AAV6 platform is an effective and versatile tool for genome editing in human MSCs.
a Schematics outline procedures for genome editing using the Cas9-AAV6 platform in human MSCs derived from the bone marrow (BM), adipose tissue (AD), and umbilical cord blood (UCB) for site-specific transgene integration. MS-modified guide RNAs and Cas9 nuclease are electroporated as All-RNA cocktail or RNP complex. Then, homologous repair templates containing GFP reporter cassette insert are delivered by electroporation-aided transduction (EAT) of AAV6 vectors for 15 min during recovery period. Integration of GFP-overexpression cassette can be detected ~7–10 days later. HA: homology arm b Representative single-channel and overlay images of a GFP+ hBM-MSC colony originated from low-density seeding (220 cells/cm2) one passage in culture after nuclease electroporation and EAT. Scale bars represent a distance of 400 µm. GFP+ colony formation is observed in targeting from all 12 biological hMSC donors described in this study. c Dots and lines represent average frequencies of GFP+ cells in hBM-MSCs (biological replicates: n = 6) at different timepoints after HBB locus targeting using All-RNA nuclease and AAV6 EAT or AAV EAT only. Pi signifies the end of the ith passage post targeting. Error bars represent standard error of mean. d GFP+ cell frequencies at one passage post targeting of hBM-MSCs (biological replicates: n = 7), hAD-MSCs (n = 3), and hUCB-MSCs (n = 3), which were targeted with All-RNA cocktails or RNP complexes followed by AAV6 EAT are represented in dot plots for indicated gene loci (HBB, CCR5, and RANKL). Connected dots represent hMSCs from the same human donors. Mean indel frequencies were compared between two modes of targeting and two-tailed p-values from paired t-test are shown for significant differences of mean. Source data are available in the Source data file.
Fig. 2
Fig. 2. Genome-editing procedures and gene integration at HBB preserved ex vivo hMSC characteristics.
a HBB-targeted GFP+ population of hMSCs were enriched by FACS for genotypic and phenotypic characterization. b Genomic DNA of the GFP+ population was subjected to digital droplet PCR-based quantification of transgene-integrated HBB alleles. Absolute quantification of integrated allele was normalized to that of endogenous CCRL2 locus. Plot represents fraction of integrated alleles in sorted GFPhigh hMSC population (biological replicates: n = 12) targeted with either All-RNA or RNP nucleases and AAV donors. c (Top left) Representative image shows crystal violet staining of colony-forming units (CFU-F). Boxes enclose single colonies. (Bottom left) Representative agarose gel image shows size-separation of integrated and non-integrated HBB amplicons. Genotyping is repeated in parallel PCR reactions for each colony derived from three different hMSC donors, which yielded similar amplicon distribution as a result of gel electrophoresis. (Right) Plot represents frequency of colonies with mono-allelic and bi-allelic integration genotypes in three different biological hMSC donors. Numbers of colonies screened for each donor is listed underneath the plot as n(colonies). d Staggered histograms show representative distribution of CD105, CD73, and CD90 expression level and the lack of hematopoietic markers expression on the surface of sorted GFP+ hMSCs compared to wild-type cells (WT) and isotype controls (ISO). e Representative image panel represents morphology (brightfield), GFP expression (green), and characteristic staining of differentiated sorted GFP+ hMSCs. Accumulated lipid droplets in adipogenic differentiation were stained with C12-BODIPY or LipidTOX® stains (magenta). Calcium deposition in osteogenic differentiation was confirmed with Alizarin Red staining. Chondrogenesis was confirmed with pellet formation and Alcian Blue staining of pellet sections. Distances are indicated by scale bars. Tri-lineage differentiation potential is confirmed in all hMSCs derived from 12 different donors. Source data are available in the Source data file.
Fig. 3
Fig. 3. Both immunogenicity and poor survival contribute to clearing of subcutaneously injected hBM-MSCs in mouse xenotransplantation models.
a Cas9-AAV6-engineered bi-cistronic firefly luciferase+GFP+ hBM-MSCs (Fluc MSCs; 2.5 × 105 cells/injection) were injected into 6-mm cutaneous wound beds of db/db mice on the day of wounding (wounds: n = 6), and subcutaneously in unwounded db/db mice (injections: n = 6) or in NSG mice (n = 10). In vivo luciferase activities were measured over the course of up to 24 days with d-luciferin intraperitoneal injection and IVIS imaging systems. b Representative overlay bioluminescence and still animal images show source and intensities of emitted photons. Representative time series show decreasing luciferase activities of Fluc MSCs in injected mice overtime. c Kaplan–Meier’s plot shows time to complete clearance of luciferase activities from individual injection sites in each subject groups. d Dots and trendlines represent average net flux measured from each injection sites over time normalized to those measured on the day of Fluc MSC injection. Net flux values above baseline are shown. Dots and error bars represent mean value and standard deviation. Trendlines represent best plateau and one-phase decay curve fit of normalized flux. R2, x0, and T1/2 represent goodness of fit, plateau duration, and half-life, respectively. Complete clearance is determined by baseline photon flux measured from un-injected control animals. Luminescence flux from wounds or skin injected with Fluc MSCs was measured over time. Source data are available in the Source data file.
Fig. 4
Fig. 4. Engineered therapeutic factor-secreting hBM-MSCs improved kinetics of wound healing in acute diabetic wounds.
a Excisional stented wounds created on the dorsal skin of hyperglycemic (>500 mg/dl blood glucose) db/db mice were treated with 2.5 × 105 engineered therapeutic factor-secreting hBM-MSCs and imaged with digital photography every other day until complete closure. Relative wound size were determined by measuring the wound area relative to constant silicone ring area, and normalized to original wound size. Areas under the curve (AUC) were calculated for each individual wound. Time to complete closure was determined in 2-day intervals and unclosed wounds were excluded at the end of observation periods. b Bars represent average (n = 2 technical cell-plating replicates for all groups except PDGFB MSCs (n = 3 technical cell-plating replicates)) total level of therapeutic proteins secreted by 2.5 × 105 engineered hMSCs, wild-type, and GFP-integrated controls over 24 h, as quantified by ELISA assay. c Dots and lines show changes in average relative wound size over time of engineered hBM-MSC treatments against PBS vehicle-treated and baseline MSC controls (biological replicate: n = 10 wounds per treatment group). Error bars represent standard error of mean. Daggers represent same control datasets. d Box and whiskers plots show AUC of individual wounds (n = 10 per treatment group), which were normalized against mean AUC of PBS controls (n = 10 PBS control wounds) within the same experimental runs. Minima and maxima are bounded by whiskers. Box lower bounds, centers, and upper bounds represent the 25th, 50th, and 75th percentiles, respectively. Mean AUCs of all treatment groups were different according to ordinary one-way ANOVA test (p < 0.0001). Mean AUCs of engineered hBM-MSCs treated wounds were compared to that of control hBM-MSCs-treated wounds using Dunnett’s test. Reported p-values were corrected for multiple comparison. Asterisks indicate level of significance (**p < 0.01; ****p < 0.0001). e Kaplan–Meier’s plots represent time to complete wound closure of engineered hBM-MSCs treated wounds, as compared to control hBM-MSC treatment. Chi-square p-values represent significant difference in time to closure between two groups according to Log-rank test. Daggers represent same control datasets. Source data are available in the Source data file.
Fig. 5
Fig. 5. Growth factor activities of PDGFB and VEGFA MSCs influenced local changes in newly formed skin tissue.
a Harvested tissues from 1-month-old wounds were cross-sectioned and stained with Masson’s trichrome dyes before light microscopy imaging. Amounts of granulation tissue were quantified as cross-section areas from high-resolution digital images. Representative images show impaired granulation tissue formation, which can be found in wounds from controls treatment groups, and mature granulation tissue in wounds from PDGFB MSC and VEGFA MSC treatment groups. Epi epithelium, GT granulation tissue, UW unwounded skin. b Box and whiskers plots show distributions of granulation tissue areas measured from cross-sections of individual wounds from PBS and hMSC controls, PDGFB MSC, and VEGFA MSC treatment groups (n = 10, 8, 8, and 10 wounds, respectively). Minima and maxima are bounded by whiskers, lower bounds, centers, and upper bound of boxes represent the 25th, 50th, and 75th percentiles, respectively. Differences of means were tested by ordinary one-way ANOVA and p-values were computed using Tukey’s multiple comparisons test. Asterisks indicate p-values (*p < 0.05; **p < 0.01, n.s.: p > 0.05). c (Left panel) Harvested tissues from one-month-old wounds were cross-sectioned (x-section) and stained with anti-mouse CD31 antibody. Stained blood vessels were detected from digital staining images using ImageJ software. Blood vessel density was calculated from whole granulation tissues. (Right panel) Representative blood vessel images from controls, PDGFB MSC, and VEGFA MSC treatment groups are shown. d Box and whiskers plots show distributions of blood vessel densities measured from cross-sections of individual wounds from PBS and hMSC controls, PDGFB MSC, and VEGFA MSC treatment groups (n = 6, 6, 7, and 8 wounds, respectively). Minima and maxima are bounded by whiskers, lower bounds, centers, and upper bound of boxes represent the 25th, 50th, and 75th percentiles, respectively. Differences of means were indicated by ordinary one-way ANOVA and p-values were computed using Tukey’s multiple comparisons test. Asterisks indicate p-values (**p < 0.01; ***p < 0.001, n.s.: p > 0.05). Source data are available in the Source data file.
Fig. 6
Fig. 6. Engineered hMSCs are compatible with hydrogel scaffold treatment for wound healing.
a VEGFA MSCs were delivered by injection in the wound bed or by HyStem®-HP hydrogel embedding on top of the wound bed on the day of wounding. b Box and whiskers plot shows AUC of individual wounds, which were normalized against mean AUCs of PBS controls (biological replicates: n = 10 per treatment group). Minima and maxima are bounded by whiskers, lower bounds, centers, and upper bound of boxes represent the 25th, 50th, and 75th percentiles, respectively. Mean AUCs of all treatment and control groups were different according to ordinary one-way ANOVA test (p < 0.0001). Plotted treatment groups have no significant differences in AUCs. c Wound images show appearances of wounds treated with HyStem®-HP hydrogel alone (label: “Hydrogel”) and embedded VEGFA MSCs (label: “Hydrogel:VEGFA MSCs”) over the course of observation period. Masson’s trichrome stained cross-section images show structure of granulation tissues at the end of observation period. Cyan boxes indicate location of high-resolution images. Cyan arrowheads indicate residual Hystem®-HP hydrogel. Source data are available in the Source data file.

Similar articles

Cited by

References

    1. Panés J, et al. Expanded allogeneic adipose-derived mesenchymal stem cells (Cx601) for complex perianal fistulas in Crohn’s disease: a phase 3 randomised, double-blind controlled trial. Lancet. 2016;388:1281–1290. doi: 10.1016/S0140-6736(16)31203-X. - DOI - PubMed
    1. Galipeau J, Sensé Bé L. Mesenchymal stromal cells: clinical challenges and therapeutic opportunities. Stem Cell. 2018;22:824–833. - PMC - PubMed
    1. Le Blanc K, et al. Treatment of severe acute graft-versus-host disease with third party haploidentical mesenchymal stem cells. Lancet. 2004;363:1439–1441. doi: 10.1016/S0140-6736(04)16104-7. - DOI - PubMed
    1. Lalu MM, et al. Safety of cell therapy with mesenchymal stromal cells (SafeCell): a systematic review and meta-analysis of clinical trials. PLoS ONE. 2012;7:e47559. doi: 10.1371/journal.pone.0047559. - DOI - PMC - PubMed
    1. Chen JS, Wong VW, Gurtner GC. Therapeutic potential of bone marrow-derived mesenchymal stem cells for cutaneous wound healing. Front. Immunol. 2012;3:192. doi: 10.3389/fimmu.2012.00192. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances