ARTICLE https://doi.org/10.1038/s41467-020-17901-2 OPEN Vulnerability of progeroid smooth muscle cells to biomechanical forces is mediated by MMP13 Patricia R. Pitrez 1,2, Luís Estronca1,2, Luís Miguel Monteiro 1, Guillem Colell3, Helena Vazão1, Deolinda Santinha 1,2, Karim Harhouri 4, Daniel Thornton5, Claire Navarro 4,6, Anne-Laure Egesipe7, Tânia Carvalho 8, Rodrigo L. Dos Santos9, Nicolas Lévy4,10, James C. Smith 11, João Pedro de Magalhães1,5, Alessandro Ori 12, Andreia Bernardo 11, Annachiara De Sandre-Giovannoli 4,10,13, Xavier Nissan 7, Anna Rosell3 & Lino Ferreira 1,2 ✉ 1234567890():,; Hutchinson-Gilford Progeria Syndrome (HGPS) is a premature aging disease in children that leads to early death. Smooth muscle cells (SMCs) are the most affected cells in HGPS individuals, although the reason for such vulnerability remains poorly understood. In this work, we develop a microfluidic chip formed by HGPS-SMCs generated from induced plur- ipotent stem cells (iPSCs), to study their vulnerability to flow shear stress. HGPS-iPSC SMCs cultured under arterial flow conditions detach from the chip after a few days of culture; this process is mediated by the upregulation of metalloprotease 13 (MMP13). Importantly, double-mutant LmnaG609G/G609GMmp13−/− mice or LmnaG609G/G609GMmp13+/+ mice trea- ted with a MMP inhibitor show lower SMC loss in the aortic arch than controls. MMP13 upregulation appears to be mediated, at least in part, by the upregulation of glycocalyx. Our HGPS-SMCs chip represents a platform for developing treatments for HGPS individuals that may complement previous pre-clinical and clinical treatments. 1 Center for Neuroscience and Cell Biology, University of Coimbra, Coimbra, Portugal. 2 Faculty of Medicine, University of Coimbra, Coimbra, Portugal. 3 Neurovascular Research Laboratory, Vall d’Hebron Research Institute, Universitat Autònoma de Barcelona, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain. 4 Aix Marseille Univ, INSERM, MMG, Marseille, France. 5 Integrative Genomics of Ageing Group, Institute of Ageing and Chronic Disease, University of Liverpool, Liverpool L7 8TX, UK. 6 Progelife, Marseille, France. 7 CECS, I-STEM, AFM, Institute for Stem Cell Therapy and Exploration of Monogenic Diseases, Evry Cedex, France. 8 IMM, Instituto de Medicina Molecular, Universidade de Lisboa, Lisbon, Portugal. 9 25 Cambridge Science Park, Mogrify Ltd, Milton Road, Cambridge CB4 0FW, UK. 10 Molecular Genetics Laboratory, Department of Medical Genetics, La Timone Children’s Hospital, Marseille, France. 11 Developmental Biology Laboratory, Francis Crick Institute, London NW1 1AT, UK. 12 Leibniz Institute on Aging - Fritz Lipmann Institute, 07745 Jena, Germany. 13 CRB Assistance Publique des Hôpitaux de Marseille (CRB AP-HM, TAC), Marseille, France. ✉email: lino@uc-biotech.pt NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 1 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 H utchinson–Gilford Progeria Syndrome (HGPS) is caused induction media (Supplementary Fig. 3) followed by SMC by a single mutation in the lamin A/C gene (LMNA), maturation media (Supplementary Fig. 4) for an additional four resulting in the generation of an abnormal lamin A pre- passages. Matured SMCs are referred to as HGPS-iPSC SMCs or cursor named progerin1,2. One of the key reasons of premature N-iPSC SMCs based on their phenotype, genotype, and func- death is the loss of smooth muscle cells (SMCs) in the medial tional properties (see below). Both HGPS-iPSC SMCs and N- layer of large arteries, followed by the appearance of collagen and iPSC SMCs have similar or higher expression of SMC mRNA extracellular matrix (ECM) and the development of a severe transcripts than somatic hVSMCs (Supplementary Fig. 4a). arteriosclerotic process that leads to increased arterial stiffness3–5. Greater than 95% of both differentiated cells express α-SMA, The reasons of SMC loss remain to be determined. It has been smooth muscle myosin heavy chain (SMMHC), and calponin suggested that this may happen due to pathophysiological proteins (Fig. 1b). Moreover, HGPS-iPSC SMCs express progerin changes inherent to prelamin A/progerin accumulation, such as mRNA transcripts (Fig. 1c) and progerin protein (Supplementary the acceleration of vascular calcification via the activation of the Fig. 4b, c). Similar results were obtained for SMCs derived from DNA damage response and senescence-associated secretory HGPS-iPSCs generated from a second Progeria individual; how- phenotypes in vascular SMCs6 or the downregulation of PARP17. ever, the differentiated cells showed higher progerin protein levels It has also been shown that the combined effect of progerin than the first Progeria individual (Supplementary Fig. 5). accumulation and mechanical stress in mouse SMCs over- Importantly, HGPS-iPSC SMCs and N-iPSC SMCs are functional expressing progerin promoted cell detachment and death, while as they respond to vasoactive agents such as histamine and the disruption of the linker between nucleoskeleton and cytos- angiotensin (Supplementary Fig. 4d) and they contract after keleton complex ameliorated the toxic effects of progerin8. Nei- exposure to carbachol (Supplementary Fig. 4e). ther of these studies have fully addressed the reasons behind SMC SMCs derived from HGPS-iPSCs share similar features to detachment and thus which therapeutic approach could be progerin-expressing cells. Cell lines forced to express progerin effective to prevent SMC loss. show the activation of several NOTCH signaling pathway Induced pluripotent stem cells (iPSCs) offer an unlimited effectors15. Indeed, our results showed that HGPS-iPSC CD34+ source of SMCs to study HGPS. Recent studies have generated cells had higher expression of NOTCH signaling pathway mRNA iPSCs from fibroblasts obtained from individuals with HGPS transcripts than N-iPSC CD34+ cells (Supplementary Fig. 6). (hereafter referred to as HGPS-iPSCs)9–11. Strikingly, HGPS- Mature HGPS-iPSC SMCs also expressed higher levels of NOTCH iPSCs show low lamin A/C and progerin protein expression in the ligand and receptors than N-iPSC SMCs (Supplementary Fig. 6a). pluripotent state. However, the expression of progerin is reacti- In addition, HGPS-iPSC SMCs responded to farnesyltransferase vated after HGPS-iPSC differentiation into SMCs7,9. The differ- inhibitors, as has been shown in other Progeria cell models16–18. entiated cells show nuclear dysmorphology, cell growth In the current work, HGPS-iPSC SMCs treated with lonafarnib for retardation, susceptibility to apoptosis, proliferation reduction, 48 h accumulated nuclear prelamin A and showed a decrease in and DNA-repair defects; however, SMC performance under flow nuclear shape abnormalities and nuclear blebbing (Supplementary conditions has not been evaluated. Fig. 7a–c). Taken together, the cells differentiated from HGPS- In this work, we develop an in vitro cell system comprising iPSCs-expressed SMC and progeroid markers, are functional and SMCs derived from HGPS-iPSCs cultured under flow conditions exhibit physiological responses. in a microfluidic device. We identify MMP13 as a mediator of SMC detachment using chemical and genetic assays. The gener- ated double-mutant LmnaG609G/G609GMmp13−/− mice show an HGPS-iPSC SMCs are vulnerable to arterial shear stress. SMCs increase in SMCs in the aortic arch and a decrease in progerin- differentiated from N-iPSCs or HGPS-iPSCs were seeded in a positive cells. In addition, the inhibition of MMP13 in microfluidics system and cultured under flow conditions for up to LmnaG609G/G609G mice by Batimastat, a drug that has been pre- 7 days (Fig. 1d). Because SMCs from large arteries are the most viously tested in clinical trials in cancer patients, reduces SMC affected in blood vessels in HGPS, we used a flow of 20 dyne/cm2, loss. The results present here open perspectives for HGPS which is typically found in arterial blood vessels19. N-iPSC SMCs treatment. (Fig. 1g), hVSMCs, or HGPS fibroblasts (80% of which express progerin) (Fig. 1e, g) can be cultured in the microfluidics system for at least 7 days without a visible loss in cell number. In con- Results trast, HGPS-iPSC SMCs cultured under flow conditions formed SMCs derived from HGPS-iPSCs are functional and share cell clumps overtime (Fig. 1f), and most of the cells detached from similar features to progerin-expressing cells. iPSCs were gen- the substrate at day 4 as confirmed by cell number (Fig. 1g) and erated from HGPS skin fibroblasts and characterized as pre- metabolic analyses (Fig. 1h). During this time period, the per- viously described10. iPSCs generated from non-disease cells (N- centage of cells expressing progerin and displaying nuclear iPSCs), HGPS skin fibroblasts, and non-disease somatic human abnormalities increased significantly until day 4 (Supplementary vascular smooth muscle cells (hVSMCs) were used as controls. Fig. 8). Our results indicate that SMC detachment is mediated by The mutation in the LMNA gene, both in HGPS skin fibroblasts progerin accumulation, as the inhibition of progerin by antisense and HGPS-iPSCs, was confirmed by Sanger sequencing (Sup- morpholinos20 significantly decreased HGPS-iPSC SMC detach- plementary Fig. 1). As expected, undifferentiated HGPS-iPSCs ment (Supplementary Fig. 9). In addition, we showed that HGPS- expressed low levels of HGPS markers, such as progerin, as well as iPSC SMCs with high progerin expression (30% of the cells low levels of SMC markers, such as α-SMA and SMα-2212,13 express progerin at day 0) detached from the surface of the (Supplementary Fig. 2a). To induce the differentiation of HGPS- microfluidics system in a short time (<12 h) (Supplementary iPSCs or N-iPSCs into SMCs, CD34+ cells were isolated by Fig. 5g). To confirm that progerin accumulation is responsible for magnetic-activated cell sorting from embryoid bodies (EBs) cul- SMC loss, a frameshift mutant stem cell line was generated tured for 10 days in suspension (Fig. 1a)14. At this stage, HGPS- (HGPSΔ2-iPSCs) to knockout the HGPS mutant allele and gen- CD34+ cells already express higher levels of progerin mRNA erated a disease cell line, as previously described in the mouse21 transcripts relative to N-iPSCs but relatively low levels of SMC (Fig. 2a and Supplementary Fig. 10). Specifically, a two-base pair mRNA transcripts compared with somatic hVSMCs (Supple- deletion on exon 11, upstream of the HGPS point mutation mentary Fig. 2b). HGPS-CD34+ cells were then cultured in SMC (1814C>T), was generated. Notably, HGPSΔ2-iPSCs expressed 2 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE a b c N-iPSC-SMCs N-iPSC-SMCs Differentiation HGPS-iPSC-SMCs HGPS-iPSC-SMCs 120 HGPS fibroblasts in suspension 101 * 100 100 ** *** Relative gene expression Undifferentiated **** **** cell colony CD34+ cells 80 10–1 **** % of cells 60 10–2 HGPS-iPSC SMCs 40 10–3 N-iPSC SMCs 4 Passages (≈15 d) 4 Passages (≈15 d) 20 10–4 Maturation media Inductive media 0 10–5 lp HC A A B1 in Ca SM N er N M LM og LM α- SM Pr d e f HGPS Fibroblasts hVSMCs Cell clumps Area of cell clumps Day 1 Day 7 Day 8 40,000 15 Cell clumps (number/μm2)(x106) Area of cell clumps (μm2) 30,000 10 HGPS-iPSC SMCs 20,000 0 2 4 6 Days Day 1 Day 4 Day 6 5 Cell counting 10,000 Cell metabolism Cell proliferation 0 0 Cell apoptosis 2 4 6 Time (days) g N-iPSCs-SMCs h N-iPSCs-SMCs i N-iPSCs-SMCs j N-iPSC SMCs HGPS-iPSCs-SMCs HGPS-iPSCs-SMCs HGPS-iPSCs-SMCs hVSMCs HGPS-iPSC SMCs hVSMCs HGPS fibroblasts HGPS fibroblasts HGPS fibroblasts HGPS fibroblasts 80 2.0 0.6 Normalized caspase 9 acticity p < 0.0001 p = 0.0084 Normalized cell number ** % Ki67 positive cells p = 0.0123 60 **** 100 Metabolic activity 1.5 0.4 p = 0.0305 * p < 0.0001 * p < 0.0001 1.0 40 **** **** 50 0.2 *** 0.5 20 *** 0.0 0.0 0 0 0 2 4 6 0 2 4 6 0 4 0 4 0 4 0 4 0 4 0 4 Time in flow shear Time in flow shear Time (days) Time (days) stress (days) stress (days) little or no progerin upon differentiation into SMCs as demon- artery ECs (HUAECs, on top of the SMCs) under flow conditions. strated at the transcript and protein levels and did not detach Initially, we screened different culture conditions and we found under flow culture conditions (Fig. 2). that endothelial growth media-2 (EGM-2) medium was a suitable HGPS-iPSC SMC detachment does not seem to be mediated by medium to support both cells (Supplementary Fig. 14). Then, we cell apoptosis. Before cell detachment, HGPS-iPSC SMCs showed: co-cultured HUAECs and HGPS-iPSC SMCs at different ratios (i) poor proliferation (as monitored by Ki67 staining) confirming (1.6, 1, and 0.6) under flow conditions. In all the ratios tested, we their contractile phenotype (Fig. 1i), (ii) similar levels of apoptosis had a monolayer of HUAECs (Supplementary Fig. 15a) and as N-iPSC SMCs as confirmed by caspase-9 activity (Fig. 1j), (iii) HGPS-iPSC SMCs at time zero. After 6 days in flow conditions, a an osteogenic differentiation program (Supplementary Fig. 11a, significant percentage (>40%) of HGPS-iPSC SMCs was lost b), (iv) increased DNA damage6 (Supplementary Fig. 12), and (v) (Supplementary Fig. 15b). For the highest ratio tested (1.6), the downregulation of NOTCH15,22 (Supplementary Fig. 13) signal- loss of HGPS-iPSC SMCs occurred without visible loss of ECs. ing pathways. Because the in vivo shear stress from blood flow is Yet, for EC:SMC ratios below 1, part of ECs also detached from not directly sensed by SMCs but by endothelial cells (ECs), we co- the microfluidic chamber indicating that, a low ECs density, cultured SMCs differentiated from HGPS-iPSCs (directly may turn ECs vulnerable to flow conditions. Importantly, cell attached to the microfluidics substrate) with human umbilical vulnerability to flow conditions was only observed in co-cultures NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 Fig. 1 Vulnerability of HGPS-iPSC SMCs to arterial flow conditions. a Schematic representation of the methodology used to differentiate iPSCs into SMCs. b Expression of SMC markers on iPSC-derived SMCs. Percentage of positive cells expressing SMC markers as evaluated by immunofluorescence (at least 100 cells were counted per each marker). Results are mean ± SEM (n = 3 independent experiments). c Expression of progeria markers on iPSC- derived SMCs. Gene expression by qRT-PCR (gene expression was normalized by the housekeeping gene GAPDH). HGPS fibroblasts were used as control. Results are mean ± SEM (n = 4 technical replicates from a pool of three independent experiments). *, **, ***, **** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. d Schematic representation of the protocol used. Cells were cultured for 6–8 days in arterial flow conditions (20 dyne/cm2). e Light microscopy images of HGPS fibroblasts, hVSMCs, or HGPS-iPSC SMCs (10% of the cells accumulate progerin protein) at different culture days. Only HGPS-iPSC SMCs detached from the microfluidic system at day 4. Scale bar is 50 μm. f Number and area of cell clumps in HGPS-iPSC SMCs at different times (at least two images (×10) have been quantified per time). For area of cell clumps n > 2 images examined over three independent experiments; for cell clumps, n = 3 independent experiments. g Number of cells per surface area (mm2) during cell culture under arterial flow (at least three images (×10) have been quantified per time; n = 3–7 independent experiments). Cell number was normalized by the number of cells present at day 0. h Cell metabolism evaluated by the Presto Blue assay. Absorbance at 570 nm was measured and normalized to the 600-nm values for the experimental wells. n = 3 independent experiments. i Expression of nuclear proliferation marker, Ki67 (at least three images (×10) have been quantified per time). The percentage of Ki67 positive cells was evaluated by immunofluorescence. n > 3 images examined over three independent experiments. j Cell apoptosis evaluated by caspase-9 activity. Results were normalized by cell number. n = 3 independent experiments. From c to g, results are mean ± SEM. *, **, ***, **** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). Statistical analyses were performed by a two-tailed unpaired Student’s t test i and j. of HGPS-iPSC SMCs but not N-iPSC SMCs (Supplementary confirmed by qRT-PCR (Fig. 4a), and the molecular network of Fig. 15c). genes that were differentially expressed between days 0 and 4 in It has been shown that a knock-in mouse line carrying a the HGPS-SMCs was examined by Ingenuity Pathway Analysis homozygous Lmna c.1827C>T;p.Gly609Gly mutated allele (Supplementary Fig. 17). Interestingly, pathway analysis sug- (LmnaG609G/G609G) recapitulates most of the described alterations gested that MMP13 is either a direct or indirect target of multiple associated with HGPS, including the loss of SMCs20. Thus, to genes upregulated at day 4. Moreover, MMP13 transcript levels validate the results obtained for the HGPS-iPSC SMCs, we are elevated in HGPS-iPSC SMCs when compared with SMCs isolated SMCs from wild-type (WT mSMC) and homozygous generated from the attenuated disease version of this line LmnaG609G/G609G (HOZ mSMC) mice. Both cells expressed (HGPSΔ2-iPSC SMCs), specially post shear stress (Fig. 2g). calponin and α-SMA, while HOZ mSMCs, but not WT SMCs, To further explore the gene array results, we evaluated whether showed dysmorphic nuclei and nuclear blebbing (Fig. 3a, b). WT the presence of ECM secreted by hVSMCs could prevent the mSMCs were cultured under flow conditions (120 dyne/cm2 to detachment of HGPS-iPSC SMCs under arterial flow conditions. mimic mice arterial flow shear stress23,24) for up to 26 days Thus, we cultured HGPS-iPSC SMCs on decellularized ECM without visible loss of cells (Fig. 3c). In contrast, HOZ mSMCs deposited by hVSMCs or directly on top of mitotically inactivated detached from the substrate after 8–9 days. These results confirm hVSMCs (Supplementary Fig. 18). Both conditions were unable that HOZ mSMCs are vulnerable to flow shear stress similar to to prevent HGPS-iPSC SMC detachment. Next, we tested whether HGPS-iPSC SMCs. Overall, our results indicate that HGPS-iPSC conditioned media collected from HGPS-iPSC SMCs in flow SMCs are vulnerable to flow shear stress, as in the case of SMCs conditions for 4 days could induce the detachment of flow shear isolated from mice carrying a HGPS-like mutation in the Lmna stress-insensitive hVSMCs (Fig. 4b). Surprisingly, hVSMCs gene. detach after perfusion with HGPS-iPSC SMC-conditioned media but not with N-iPSC SMC-conditioned media (Fig. 4c). Following these results and given that MMP13 appears to be the HGPS-iPSC SMCs have significant changes in extracellular downstream effector for the genes misregulated at day 4 matrix (ECM) secretion and MMP expression. To gain insights (Supplementary Fig. 17b) we decided to quantify the concentra- into the mechanism behind SMC detachment, we performed tion of MMP13 in HGPS-iPSC SMC and N-iPSC SMC culture microarray analyses on HGPS-iPSC SMCs and N-iPSC SMCs at media after flow shear stress. Remarkably, MMP13 levels days 0 and 4 (before cell detachment). At day 0, 2084 genes were increased 30-fold in the HGPS-iPSC SMC culture media, but differentially expressed (Log2FC ≥ 1; p < 0.05) in HPGS-iPSC not in the control cell culture media (Fig. 4d). Similarly, higher SMCs vs. N-iPSC SMCs. Of these genes, 51 genes were associated MMP13 levels were observed in media collected from HOZ with cell senescence, as determined by the intersection of all the mSMCs under flow shear stress, when compared with media from differentially expressed genes with the CellAge database25 (279 WT mSMCs (Fig. 3d). Because MMP13 is produced by cells as an genes) (Supplementary Data 1). At the protein levels, HGPS-iPSC inactive form (proMMP13), which is then activated by cell SMCs expressed higher levels of p21 and SA-β-galactosidase than membrane MMPs, namely MMP14 (also called MT1-MMP) and N-iPSCs-SMCs and the level of senescence markers increased MMP2 (also called gelatinase A)26, the catalytic activity of after culture of HGPS-iPSC SMCs in flow conditions (Supple- MMP13 secreted by HGPS-iPSC SMCs was analyzed (Supple- mentary Fig. 16a and Supplementary Data 5). We next performed mentary Fig. 19). The concentration of proMMP13 and active pathway analysis on the differentially expressed genes from MMP13 increased approximately eight- and five-fold, respec- HGPS-iPSC SMCs at day 0 vs. day 4 (Supplementary Fig. 17 and tively, in culture media of HGPS-iPSC SMCs cultured in flow Supplementary Data 2, 3). In general, ECM activation, secretion, conditions from day 0 to day 4. Moreover, the concentration of and cell adhesion pathways were upregulated, whereas cell cycle proMMP13 and active MMP13 in cell culture media collected and DNA replication pathways were downregulated under from N-iPSC SMCs cultured in flow conditions for 4 days was arterial flow conditions at day 4. Among the fifty-seven genes that more than fourfold lower than the one observed with HGPS-iPSC were at least threefold down- or upregulated compared with day 0 SMCs. Altogether, our results indicate that HGPS-iPSC SMCs (p < 0.001) (Fig. 4a), five were related to ECM secretion (COL6A3, cultured under flow conditions showed increased cell senescence, IBSP, BGN, SGCG, and EPPK1) and one to metalloproteases ECM activation, secretion, and cell adhesion pathways upregula- (MMP13). The expression of these genes, as well as others, was tion and dysregulation in the expression of MMP13. 4 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE a Chromosome 1 HGPS Point Mutation Site 5’ 3’ LMNA gene locus Exon 11 3’ 5’ HGPS allele PAM sgRNA N-iPSC 5’ 3’ WT allele 5’ 3’ WT allele HGPS-iPSC 5’ 3’ WT allele 5’ 3’ HGPS allele HGPSΔ2-iPSC 5’ 3’ WT allele 5’ 3’ b Morphology Nuclei Lamin A Nuclei Progerin Nuclei SMMHC Nuclei Calponin f g HGPS-iPSC SMCs HGPS-iPSC SMCs HGPSΔ2-iPSC SMCs c 100 d 120 e HGPSΔ2-iPSC SMCs 120 2.0 100 LMNA G608G expression p = 0.0016 Normalized cell number p < 0.0001 ** 10–1 MMP13 expression 90 90 1.5 p = 0.0002 10–1 **** *** % of cells % of cells (by day 0) p = 0.0035 –2 10 60 60 1.0 10–2 ** 10–3 30 30 0.5 10–3 10–4 0 0 0.0 **** 10–4 0 2 4 6 -iP S ts SM s s SM alp og A .N n ng C 0 4 eb i Bl ucle SC MC C D eri Δ2 SC las H Pr in bi C M Time (days) m Time (days) PS -iP ob La G S r H GP fib H PS G H MMP13 mediates HGPS-iPSC SMC loss under flow conditions. confirm these results, HGPS-iPSC SMCs were subjected to siRNA Next, we tested whether the chemical inhibition of MMPs could knockdown of MMP13 and cultured under arterial flow condi- prevent HGPS-iPSC SMC detachment. For this purpose, we used tions for 10 days (Fig. 4f, g). Our results show that the knock- Batimastat (BB-94)27, a broad spectrum matrix metalloprotease down of MMP13 in SMCs increased the stability of HGPS-iPSC inhibitor (IC50 = 33 nM for MMP1328), and a specific MMP13 SMCs in flow culture conditions compared with non-treated inhibitor pyrimidine-4,6-dicarboxylic acid, bis-(4-fluoro-3- cells. We also analyzed the effects of MMP13 and BB94 inhibition methyl-benzylamide) (IC50 = 8 nM)29. Remarkably, both inhi- in HOZ mSMCs (Fig. 3e). Similar to what was observed with bitors significantly decreased the detachment of HGPS-iPSC HGPS-iPSC SMCs, the detachment was significantly delayed SMCs cultured under arterial flow conditions (at least until day when one of the inhibitors was used. To further demonstrate the 12) (Fig. 4e), and this effect was much superior to that of lona- importance of MMP13 in HGPS-iPSC SMC detachment, we farnib (Supplementary Fig. 7d) or inhibition through the pyr- enforced the expression of MMP13 in somatic SMCs (hVSMCs) ophosphate calcification process30 (Supplementary Fig. 11c). To and cultured the modified cells in flow culture conditions NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 Fig. 2 Expression of progeria and SMC markers in HGPSΔ2-iPSC SMCs. a gRNA directs Cas9 nuclease against mutated exon 11 of LMNA gene, upstream the HGPS mutation, disrupting progerin, without altering lamin A and lamin C. Sanger sequencing for LMNA (NM_170707.4 transcript) exon 11 was performed for: N-iPSCs, HGPS-iPSCs and HGPSΔ2-iPSCs, confirming the deletion of two-base pairs in the HGPSΔ2-iPSCs. b Expression of lamin A, progerin, and SMC proteins monitored by immunofluorescence. Scale bar is 100 μm. n = 6 independent experiments. c Expression of progerin (LMNA G608G gene) in HGPS and HGPSΔ2 cell lines. Results are mean ± SEM (n = 4 technical replicates from a pool of three independent experiments). Statistical analyses were performed by a two-tailed unpaired Student’s t test. d Quantification of lamin A, progerin, dysmorphic nuclei, and nuclei blebbing. Results are mean ± SEM (n = 6 independent experiments). **** denotes statistical significance (p < 0.0001). e Percentage of cells that have been differentiated from HGPSΔ2-iPSCs that express SMC markers at protein level. Results are mean ± SEM (n = 5–6 independent experiments). f Number of cells per surface (mm2) as quantified by high-content microscopy (at least three images (×10) have been quantified per time). The number of cells was evaluated after 6 days under arterial flow and was normalized by the number of cells present at day 0. n > 3 images examined over three independent experiments. g MMP13 mRNA transcripts quantified by qRT-PCR analyses in HGPS-iPSC SMCs or HGPSΔ2-iPSC SMCs cultured under flow conditions. MMP13 mRNA transcripts were normalized by GAPDH. n = 4 technical replicates from a pool of three independent experiments. **, *** denote statistical significance (p < 0.01, p < 0.001). Statistical analyses were performed by a two-tailed unpaired Student’s t test. a WT mSMCs HOZ mSMCs b 100 100 % of cells with organized fibers 80 80 α-SMA 60 60 % of cells 40 40 20 20 Lamin A 0 0 α- in A ng eb ei SM n Bl ucl bi po .N al C D c d e HOZ WT mSMC WT mSMCs HOZ MMP13i HOZ mSMC HOZ mSMCs HOZ BB94 120 0.03 120 **** ** Concentration of MMP13 (pg/cell) 100 100 p = 0.0022 % of adhered cells % of adhered cells 80 0.02 80 60 60 40 0.01 40 20 20 0 0.00 0 26 0 3 9 0 8 0 8 0 5 10 15 Time (days) Time (days) Time (days) Fig. 3 Characterization and impact of flow shear stress in SMCs isolated from wild-type (WT) and homozygous (HOZ) LmnaG609G/G609G mice. a Mouse SMCs were cultured for 9–26 days in arterial flow conditions (120 dyne/cm2). Immunofluorescence analyses performed on mouse SMCs (6-week- old WT and HOZ Lmna G609G/G609G mice) at passage 4 for α-SMA and Lamin A. Nuclei were stained with DAPI. Scale bar is 20 µm. n = 3–4 images examined over three independent experiments. b Percentage of dysmorphic nuclei, nuclei blebbing, and SMC organized fibers in mSMCs (assessed in static conditions). n = 3-4 images examined over three independent experiments. c Percentage of adhered cells over time. Cells were cultured under flow conditions. n = 3–4 independent experiments. Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. d Quantification of MMP13 in HOZ mSMCs and WT mSMCs. Cells were analyzed at day 0 and day 8 under flow. Fluorescence signal was normalized by cell number. n = 3–4 independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. e Percentage of adhered cells over time. Cells were cultured under flow conditions. n = 5–6 independent experiments. In graphs b–e, results are mean ± SEM. *,**,***,**** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). 6 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE a HGPS-iPSC SMCs 10–7 N-iPSC SMCs 10–6 14 12 10–5 *** **** 10 Fold change –4 8 p -value 10 6 * p = 0.0023 10–3 ** ** 4 *** 2 *** ** ** 10–2 0 * 10–1 –2 **** ** *** **** **** –4 100 A SE 1 EP 4 G N 8 3 3 C S3 A3 G 6 2 SP PH TN TN N LF K –6 –4 –2 0 2 4 6 PK L S 4A P1 AN IP SN L2 C BG PX PT EP A L6 IB AM LI SG FA M R M H Log2 (fold change) O M O b Effect of media in the detachment of hVSMCs 4 days Concentration of MMP13 Cell detachment MMP13 inhibition Progerin accumulation 4 – 12 days Alkaline phosphatase activity c d HGPS-iPSC SMCs e N-iPSC SMCs HGPS-iPSC SMC CM hVSMCs N-iPSC SMC CM HGPS fibroblasts d7 d12 150 8 2.0 p < 0.0001 **** **** **** Normalized cell number MMP13 (x10–4, pg/cell) % of adhered hVSMCs 125 6 1.5 **** Concentration of **** 100 to day 0 75 4 1.0 p = 0.0156 50 * 2 p = 0.0417 0.5 25 * 0 0 0.0 0 12 24 36 48 04 04 04 04 M 4 M 4 3i B d 3i M 9 M B9 te P1 P1 BB Time (h) Time (days) ea -tr on N f No transfection g h i siRNA control HGPS-iPSC SMCs siRNA MMP13 HGPS fibroblasts 0.04 40 150 5 *** (transfected with control siRNA) Alkaline phosphatase activity normalized by cell number Relative gene expression *** % of cells with progerin 4 0.03 30 Fold of control 100 3 0.02 20 p < 0.0001 **** 2 50 0.01 10 1 0.00 0 0 0 7 10 – + – + – + Time (days) MMP13 inhibition MMP13 inhibition (Supplementary Fig. 19). Notably, the number of cells observed at inhibition of MMP13 in HGPS-iPSC SMCs cultured for 7 days in day 7 is lower than the one observed in WT cells indicating that flow conditions reduced the percentage of progerin-positive cells some of the modified cells were lost during the flow culture (Fig. 4h); however, it did not decrease progerin expression in cells conditions. with high levels of progerin, such as HGPS fibroblasts. In We then asked whether the modulation of MMP13 activity addition, the chemical inhibition of MMP13 did not reduce the could affect progerin expression associated with the vulnerability activity of alkaline phosphatase in HGPS-iPSC SMCs cultured for of HGPS-iPSC SMCs to flow shear stress. Interestingly, chemical 7 days in flow conditions (Fig. 4i). Overall, the results obtained NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 Fig. 4 MMP13 activity in HGPS-iPSC SMCs cultured under flow shear stress. a Volcano plot representing differentially expressed genes in HGPS-iPSC- SMC cultured under flow conditions at day 0 and 4. Each point represents one of 53,617 genes. 26 and 31 genes were upregulated (red; fold change ≥ 3; p < 0.001) and downregulated (yellow; fold change ≤ 3; p < 0.001), respectively. Graph shows qRT-PCR validation for 16 genes with fold changes >3. Fold change was between days 0 and 4. Gene expression was normalized by the housekeeping gene GAPDH. Results are mean ± SEM, n = 4 technical replicates from a pool of three independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. b Schematic representation of the experimental protocol used. c Effect of HGPS-iPSC SMC or N-iPSC SMCs conditioned media (in both cases obtained after 4 days under flow conditions) on hVSMCs cultured under flow conditions. n = 1–5 images examined over three independent experiments. d Quantification of MMP13 activity (cell culture media) by ELISA. Cells were analyzed at days 0 and 4 under flow. Fluorescence signal was normalized by cell number. n = 3 independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. e Effect of MMP13 or BB94 inhibition in HGPS-iPSC SMC detachment. The number of cells was evaluated after 7 and 12 days under arterial flow and was normalized by the number of cells present at day 0. n = 3–5 images examined over three independent experiments. Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. f MMP13 knockdown by siRNA in HGPS-iPSC SMCs. MMP13 mRNA transcripts were quantified by qRT-PCR and normalized by GAPDH. Mean ± SEM (n = 4 technical replicates from a pool of three independent experiments). Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. g Number of cells per microfluidic area during culture under flow shear conditions normalized by the number of cells in control experimental groups (i.e., cells transfected with control siRNA). n = 7 independent experiments for day 7 and n = 6 independent experiments for day 10. h Percentage of progerin-positive cells after 7 days under flow conditions with SmGM-2 media supplemented or not with MMP13 inhibitor. n = 1–5 images examined over three independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. i Activity of alkaline phosphatase in HGPS- iPSCs-SMC normalized by cell number per mm2, in cells cultured 4 days under flow conditions. Cells were treated or not with MMP13 inhibitor. n = 3 independent experiments. In graphs a–h, results are mean ± SEM. *, **, ***, **** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). after chemical and genetic inhibition, the increase of MMP13 LmnaG609G/G609G Mmp13+/+ mice at week 5 were intraperitoneal after flow shear stress and the effect of HGPS-iPSC SMC- (IP) injected five times a week (Fig. 7a). At week 10, Batimastat- conditioned media on cell detachment, indicate that MMP13 treated LmnaG609G/G609G Mmp13+/+ mice had similar heart rates mediates SMC loss. to non-treated animals (Fig. 7c); however, they showed higher SMCs in the aortic arch than non-treated mice, as confirmed by Inhibition of MMP13 in LmnaG609G/G609G mice significantly cell nuclei counts and verified by the increase levels of SMC increased the number of SMCs in aortic arch. To confirm the markers determined by qRT-PCR analyses (Fig. 7b, d, e). No importance of MMP13 dysregulation in progeroid animal models, differences were observed between non-treated and Batimastat- we quantified MMP13 in the plasma of LmnaG609G/G609G and WT treated mice regarding progerin accumulation in the aortic arch mice (Fig. 5a). The results showed that the levels of MMP13 were (Supplementary Figs. 20c). Overall, our data shows that the higher in mutant mice (Fig. 5b). Then, we asked whether the in vivo inhibition of MMP13 by genetic or chemical interventions inhibition of MMP13 in LmnaG609G/G609G mice could decrease yielded mice having significantly higher numbers of SMCs in the SMC loss. For this purpose, we generated double-mutant lines, aortic arch. LmnaG609G/G609GMmp13−/− and LmnaG609G/G609GMmp13+/− (Supplementary Fig. 20), and evaluated the heart rate and SMC loss Activation of MMP13 is mediated by the activation of the in the aortic arch20 of these mice at week 10 (Fig. 5a). Heart rate glycocalyx. The glycocalyx is a surface layer of proteoglycans and was chosen as a measure of the overall health status of the HGPS glycosaminoglycans that are immobilized in the cell membrane. model and the derived double-mutant lines, given that bradycardia Glycocalyx components have been shown to be involved in flow was a clinical abnormality evidenced in both LmnaG609G/G609G shear stress sensing by SMCs35,36. To identify the mechanism mouse as well as Zmpste 24−/− progeria mouse models20,31. underlying the upregulation of MMP13 in HGPS-iPSC SMCs Both double-mutant mice showed higher heart rates (Fig. 5d) cultured under arterial flow, we analyzed glycocalyx gene mRNA and numbers of SMCs (Fig. 5c, e) in the aortic arch than transcripts (Fig. 8b). Interestingly, glycocalyx transcripts were LmnaG609G/G609G Mmp13+/+ mice. Interestingly, LmnaG609G/G609G upregulated in HGPS-iPSC SMCs cultured under flow conditions Mmp13−/− and LmnaG609G/G609GMmp13+/− mice showed a lower for 4 days (Fig. 8b). From these upregulated genes, syndecan 2 number of progerin-positive cells in the aortic arch than non- gene (SDC2), which encodes the transmembrane (type I) heparan mutated mice (Fig. 5c, f). In addition, LmnaG609G/G609GMmp13+/− sulfate proteoglycan, was also upregulated in hVSMCs or N-iPSC mice (but not LmnaG609G/G609GMmp13−/− mice) showed an SMCs cultured for 4 days in flow conditions (Supplementary increase of the aortic media thickness being similar to the non- Fig. 21). Because not all the glycocalyx mRNA transcripts were mutated mice (Fig. 6a), as confirmed by orcein staining. We upregulated in hVSMCs and N-iPSC SMCs, the results suggest performed proteomic analyses of aortic arches from mutated and that the composition of glycocalyx is likely different in these cells non-mutated mice (n ≥ 5 mice per strain) using data independent when compared with HGPS-iPSC SMCs. Next, we analyzed the acquisition mass spectrometry32,33. Principal component analysis expression of heparan sulfate at the protein level. In contrast to based on 2260 proteins detected showed that the proteome profiles control cells, the expression of heparan sulfate increased when of aortic arches from LmnaG609G/G609GMmp13+/− mice were more HGPS-iPSC SMCs were cultured under flow conditions (Fig. 8a). closely related to the profile of WT mice to that of LmnaG609G/ Importantly, the enzymatic cleavage of heparan sulfate by G609G Mmp13+/+ mice (Fig. 6c). From the 161 proteins differen- heparinase III (Supplementary Fig. 22) decreased MMP13 con- tially expressed between the mutant and WT mice aortic arches (q centration in the cell culture media (Fig. 8c) and significantly < 0.05 and abs(log2 fold change) > 0.58), ~25% of the proteins had decreased the detachment of HGPS-iPSC SMCs cultured under similar expression in LmnaG609G/G609GMmp13+/− mice and WT flow conditions (Fig. 8d). Moreover, the enzymatic cleavage of mice (Fig. 6c and Supplementary Data 4). heparan sulfate slightly decreased alkaline phosphatase activity Motivated by these results, we then tested a therapeutic (Fig. 8e). approach to reduce SMC loss in LmnaG609G/G609G Mmp13+/+ To further investigate a potential ECM target of MMP13 in mice. For this purpose, we used Batimastat because human SMCs, we monitored the expression of ECM components in safety has been previously demonstrated in clinical trials34. hVSMCs, HUAECs, N-iPSC SMCs, and HGPS-iPSC SMCs. Our 8 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE a b WtWt KiWt 50 Concentration of MMP13 p = 0.0001 Age: 10 weeks 40 normalize by weight Heart rate *** Immunohistochemistry 30 Lmna+/+ = WtWt LmnaG609G/G609G Mmp13 +/+ = KiWt 20 LmnaG609G/G609G Mmp13 +/– = KiHt LmnaG609G/G609G Mmp13 –/– = KiKo 10 0 c WtWt KiWt KiHt KiKO α-SMA Nuclei Progerin Nuclei HS Nuclei 555 WtWt KiWt KiHt d KiKO e f g 2.5 10 2.5 p = 0.0142 ** ** * ** * α-SMA-positive cell nuclei/mm2 Intensity HS (a.u.)/cell number p = 0.003 **** 600 2.0 8 2.0 % Progerin positive cells ** (normalized by WtWt) (normalized by WtWt) (normalized by WtWt) * * p = 0.0343 Heart rate (bpm) * 1.5 6 1.5 400 1.0 4 1.0 200 0.5 2 0.5 0 0.0 0 0.0 Fig. 5 MMP13 inhibition significantly increases SMC number in aortic arch of LmnaG609G/G609G mice. a Schematic representation of the animal protocol. WtWt, KiWt, KiHt, and KiKo mice (age: 10 weeks) were evaluated. b Quantification of MMP13 activity (plasma from WtWt, n = 9, and KiWt, n = 6, mice) by ELISA. Fluorescence signal was normalized by mice weight. Statistical analyses were performed by a two-tailed unpaired Student’s t test. c Immunofluorescence analyses in the aortic arch for α-SMA, progerin, and heparan sulfate (HS). Cell nuclei were stained with DAPI. Scale bar is 100 µm for α-SMA staining and 50 µm for progerin and heparan sulfate staining. For α-SMA staining, n = 5 animals, except for KiHt (four animals). For progerin staining, n = 5 animals, except for KiHt (three animals). For heparan sulfate n = 6 WtWt, n = 6 KiWt, n = 4 KiHt, and n = 5 for KiKo. d Heart rates in mice (n = 8 WtWt, n = 6 KiWt, n = 7 KiHt, and n = 5 KiKo). Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. e Number of SMC nuclei in aortic arch per tissue area (mm2) (n = 2–3 slides examined over five animals, except for KiHt (four animals)). Statistical analyses were performed by a two-tailed unpaired Student’s t test. f Percentage of progerin-positive cells in SMCs. n = 5 animals, except for KiHt (three animals). Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. g Expression of heparan sulfate as evaluated by immunofluorescence. Intensity of heparan sulfate was calculated in each picture (at least 16 pictures per condition) and normalized by cell number mice (n = 6 WtWt, n = 6 KiWt, n = 4, KiHt and n = 5 KiKo). In b, d–g, results are mean ± SEM. *, **, ***, **** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 a Lmna+/+ = WtWt WtWt KiWt LmnaG609G/G609G Mmp13 +/+ = KiWt KiHt LmnaG609G/G609G Mmp13 +/– = KiHt KiKO LmnaG609G/G609G Mmp13 –/– = KiKo Aortic media thickness (μm) 60 * * WtWt KiWt KiHt KiKo 40 * * 20 0 b c KiHt KiKO KiWt WtWt 40 PC2 0 –40 –100 –50 0 50 100 PC1 Fbxo2 Actn2 Myh1 Des Ckmt2 Myh7 Myh6 Myl2 Cryab Tnnt2 Slc25a35 Abhd14b Mccc2 Psme1 Bphl Gstk1 Thop1 Dhrs7 Palm Fah Akap12 Eml1 Add1 Serpinc1 Bpgm Hbb–b2 Sept5 Rrbp1 Ppp1r9b 2 Sel1l3 Ank1 1 Apoa1 Tf 0 Alb –1 Hbb–b1 Slc4a1 –2 Ca2 Serpina1b KO Ht t W KO t t Ki H W Ki Ki Ki Ki Ki results indicate that hVSMCs express higher levels of mRNA that results indicate that activation of MMP13 is mediated, at least in encode collagen 1A1, collagen 3A1, collagen 4A2, and collagen part, by glycocalyx activation. 6A3 than HUAECs (Supplementary Fig. 21c). It has been shown that MMP13 degrades very efficiently the native helix of all fibrillary collagens, including collagen type I37. Our proteomic Discussion results indicate that indeed collagen 1A1 is upregulated in HGPS- In this study, we developed a microfluidic chip formed by a iPSC SMCs exposed to flow conditions (Supplementary Fig. 16b) monoculture or a co-culture of HGPS-SMCs (generated from and thus it may be a potential target for MMP13. Overall, our iPSCs) with ECs to study the reason underlying HGPS-SMC 10 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE Fig. 6 Proteins differentially expressed in the aortic arch at week 10 on wild-type and mutant (LmnaG609G/G609GMmp13−/− and LmnaG609G/ G609GMmp13+/−) mice. a Orcein-stained ascending aorta (elastic fibers stain in dark brown/black). Black arrow defines the internal elastic lamina while the white arrow defines the adventitial border. Images illustrate morphological changes rather than aortic media thickness differences. KiWT mice show less compact elastic lamellae and higher irregular profiles of the elastic lamellae (labeled with *) than the other mice. Scale bar is 50 µm. In graph, aortic media thickness was measured from the internal elastic lamina to the adventitial border. Black arrow defines the internal elastic lamina while the white arrow defines the adventitial border. Results are mean ± SEM, n = 3 animals, except for KiHt (four animals). * denotes statistical significance (p < 0.05). Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. b Principal component analysis (PCA) of proteome profiles obtained from aortic arches of wild-type (WtWt) and mutant (KiWt, KiHt, KiKo) mice. c Heatmap based on 161 protein groups differentially expressed between KiWt and WtWt mice, in aortic arch, at week 10 (q < 0.05 and abs(log2 fold change) > 0.58). Progerin is a mutated protein and thus not identified by the mass spectrometry. MMP13 is a secreted protein and the levels in cells were not detectable by mass spectrometry. For comparison purposes, the protein fold changes of WtWt vs. KiHt and WtWt vs. KIKo were included in the heatmap. Blue color indicates proteins downregulated in KiWt, KiHt, or KiKo as compared with WtWt, whereas red color corresponds to proteins upregulated in KiWt, KiHt, or KiKo as compared with WtWt. n = 6 for KiWt and n = 5 for WtWt, KiHt, and KiKo; age: 10 weeks. a b BB94 Placebo Batimastat (30 mg/Kg) or vehicle α-SMA Nuclei Heart rate Immunohistochemistry 1 week x5 qRT-PCR LmnaG609G/G609G Age:5 weeks c Wild type d e Placebo Placebo Placebo BB94 BB94 BB94 600 30,000 102 101 101 Relative gene expression SMMHC Relative gene expression α-SMA Relative gene expression Calp p < 0.0001 p = 0.0047 **** 101 ** 100 100 Cell number/mm2 p = 0.007 400 20,000 p = 0.0029 Heart rate 100 ** ** –1 –1 10 10 10–1 200 10,000 10–2 10–2 10–2 0 0 10–3 10–3 10–3 Fig. 7 MMP treatment using BB94 significantly increases SMC number in aortic arch of LmnaG609G/G609G mice. a Schematic representation of the animal protocol. LmnaG609G/G609G mice (n = 8 for treatment group and control group; age: 5 weeks) were IP injected five times a week (30 mg/kg/day; 3 mg/mL in PBS). b Immunofluorescence analyses performed on mouse SMC for α-SMA showing higher number of SMCs in treated aortic arch. Cell nuclei were stained with DAPI. SMCs were stained for α-SMA. Scale bar is 100 µm. For BB94 treatment n = 5 animals. For placebo treatment n = 7 animals. c Heart rates in mice. Wild-type mice were not exposed to BB94. n = 3 for wild-type mice, n = 8 for placebo treatment group and n = 7 for BB94 treatment group. d Number of SMC nuclei in aortic arch per tissue area (mm2) in mice treated or not with BB94. For BB94 treatment, n > 6 images examined over five animals. For placebo treatment, n > 9 images examined over seven animals. e Expression of SMC genes in aortic arches of mice treated or not with BB94. Gene expression was normalized by the housekeeping gene GAPDH. n > 3 technical replicates over six animals. **, ***, **** denote statistical significance (p < 0.01, p < 0.001, p < 0.0001). Statistical analyses were performed by a two-tailed unpaired Student’s t test d and e. vulnerability to flow shear stress. To generate the chip, we (i) Multiple protocols have been described in the literature for the developed a protocol to differentiate HGPS-iPSCs into functional differentiation of iPSCs into SMCs, either via an intermediate HGPS-SMCs, (ii) demonstrated that HGPS-iPSC SMCs shared progenitor stage or directed differentiation14,42–44. These proto- similar properties with other known progerin-expressing cells, cols are highly variable in terms of SMC differentiation efficiency, (iii) confirmed that HGPS-iPSC SMCs were vulnerable to arterial timescale, and functionality (nondividing contractile phenotype flow shear stress, and (iv) validated the results in ex vivo SMCs vs. proliferative phenotype, secretory profile), likely due to the isolated from LmnaG609G/G609G mice. Using the chip, we have choice of precursor population to derive the SMC subtypes, the identified MMP13 upregulation as an important mediator of chemical composition of the differentiation medium, as well as HGPS-SMC vulnerability to flow shear stress and we confirmed the choice of inductive SMC factors (e.g., PDGF-BB, TGF-β1, MMP13’s role in vivo in LmnaG609G/G609G mice (Fig. 8f). MMP13 retinoic acid). Three previous studies have reported the differ- is upregulated in a number of pathological states including entiation of HGPS-iPSCs into SMCs7,9,45 by direct differentia- atherosclerosis and rheumatoid arthritis38. The upregulation of tion7 or by using an intermediate progenitor (i.e., mesenchymal MMP13 in HGPS-SMCs cultured under arterial flow conditions stem cells45 or CD34+ cells9). In some cases, SMCs were not is in line with examples in the literature showing that enzymatic terminally differentiated (as confirmed by the expression of ECM remodeling is significantly altered in HGPS cells39–41. SMMHC)7, in others the percentage of SMCs was relatively low NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 11 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 a HGPS-iPSC SMCs N-iPSC SMCs hVSMCs HGPS fibroblasts Day 0 Day 4 Fold change in heparan sulfate **** 3 **** intensity (day 4/day 0) **** HS Nuclei 2 1 0 b HGPS-iPSC SMCs c d e f 10 1.5 3 F SMCs Loss Fold change (day 4 vs. day 0) p < 0.0001 p = 0.0103 p < 0.0001 4 * Normalized cell number **** L (heparinase III/control) MMP13 concentration 8 cell number (x10–4) **** Absorbance (a.u.)/ o 1.0 (day 6 by day 2) 2 3 w MMP13 6 2 4 0.5 1 Progerin 2 1 HGPS-SMCs 0 0.0 0 0 SD 1 4 III in rol III as l SD 2 PC LC III in ol in ro C C C ar ntr e t e ar nt e SD PR ep Con G as as ep Co ep Co ar H H H Fig. 8 MMP13 expression in SMCs is triggered by an increase in heparan sulfate. a Cells were cultured under flow conditions for 4 days and the expression of heparan sulfate was evaluated by immunofluorescence. Intensity of heparan sulfate was calculated in each picture and normalized by cell number. The normalized fluorescence intensity at day 4 was divided with the one at day 0. Scale bar is 10 μm. n > 4 images examined over six independent experiments. Statistical analyses were performed by one-way ANOVA followed by Newman–Keuls’s post test. b Gene expression of glycocalyx markers (SDC1: syndecan 1, SDC2: syndecan 2, SDC4: syndecan 4, GPC: glypican, PLC: perlecan), as evaluated by qRT-PCR, in HGPS-iPSC SMCs cultured under flow conditions. Gene expression was normalized by the housekeeping gene GAPDH, and the normalized gene expression at day 4 divided by day 0. n = 3 technical replicates from a pool of three independent experiments. c HGPS-iPSCs-SMC cultured under flow condition were treated or not with heparinase III and the number of cells per microfluidic area during culture was calculated and normalized by the number of cells present at day 2. n = 3 independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. d Quantification of MMP13 activity (cell culture media) by ELISA. Cells were analyzed at day 4 under flow. Fluorescence signal was normalized by cell number and then by control experimental group. n > 9 images examined over six independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. e Expression of alkaline phosphatase in HGPS-iPSCs SMC, normalized by cell number per mm2, in cells cultured 4 days under flow conditions. Cells were treated or not with heparinase III. n = 2 technical replicates over three independent experiments. Statistical analyses were performed by a two-tailed unpaired Student’s t test. In a–e, results are mean ± SEM. *, **, ***, **** denote statistical significance (p < 0.05, p < 0.01, p < 0.001, p < 0.0001). f Summary of the results. (i.e., only 50–60% of the differentiated cells showed specific SMC implications behind this phenotypic difference remain to be markers including α-SMA, calponin, and SMMHC)45 and no determined. Nevertheless, most of the HGPS-iPSC SMCs indication of SMC functionality9 (e.g., contractility, intracellular expressed calponin at the protein level, both at the induction and accumulation of calcium after exposure to vasoactive agents) was maturation steps (Supplementary Figs. 3, 4). A previous study has reported. In the present study, we showed that the differentiation reported heterogeneous sized calponin 1-staining inclusion bodies of HGPS-iPSCs induces the activation of the NOTCH signaling in the cytoplasm of HGPS-SMCs9; however, such structures were pathway, a hallmark of progerin-expressing cells15. This is not observed in the current study. observed in the CD34+ progenitor cells and after their differ- It has been reported that in WT animals the aorta was one of entiation into SMCs. The CD34+ cells have been reported to the tissues with the highest expression of lamin A, while in express KDR and CD3143 and, thus, are likely of lateral plate progeroid animals the aorta was the first place where progerin mesoderm origin42,44. Importantly, the differentiated cells express was detected8. This explains the highest susceptibility of HGPS- high levels of all the SMC markers analyzed (α-SMA, calponin SMCs located in the aorta to biomechanical forces. It has been and SMMHC), are contractile in response to the muscarinic reported that mouse SMCs overexpressing progerin exposed to receptor agonist, carbachol, as observed in typical human aortic biomechanical forces detach from the culture vessel after sub- SMCs, and, when matured in culture for ~30 days, they express strate stretching and die8. Yet, the mechanism of SMC detach- progerin. Therefore, our differentiation protocol compares ment is still poorly understood. Our study indicates that MMP13 favorably to other protocols in term of SMC yield and func- mediates SMC detachment as chemical or genetic inhibition of tionality. Interestingly, HGPS-iPSC SMCs express lower levels of MMP13 reduces significantly SMC loss. In addition, we found calponin than in N-iPSC SMCs but the reason and possible that the accumulation of progerin is a mediator and not the cause 12 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE of SMC detachment because HGPS fibroblasts accumulate high Future studies should address the effect of SMC preservation levels of progerin and do not detach in flow conditions. Yet, both in large vessels in the lifespan of the animals. It is possible that inhibition of progerin by morpholinos and the knockout of the the prevention of SMC loss from the large arteries might be HGPS mutant allele in HGPS-SMCs decreased or prevented SMC insufficient to lead to a significant increase in animal lifespan. detachment in flow culture conditions. Evidence collected at week 128 (before the LmnaG609G/G609G died Although LmnaG609G/G609GMmp13+/− and LmnaG609G/G609G of progeria disease) in a therapy that ameliorated SMC loss Mmp13−/− mice showed similar amelioration of SMCs loss in the showed no significant alterations in terms of body weight (which aortic arch, our proteomic analyses in the same tissue showed that is correlated with lifespan20). Our study performed for 10 weeks LmnaG609G/G609GMmp13+/− mice had a closer protein profile to showed also no significant changes in body weight (Supple- WT than LmnaG609G/G609GMmp13−/− mice. This was consistent mentary Fig. 20d) between LmnaG609G/G609GMmp13+/− mice with the media thickness size, which was more similar between WT and LmnaG609G/G609GMmp13+/+ mice. Therefore, it is possible mice and LmnaG609G/G609G Mmp13+/− than to LmnaG609G/G609G that therapies which ameliorate SMC loss should be combined Mmp13+/+ mice. Previous studies have shown that Mmp13−/− with therapies that further reduce the level of progerin in cells of mice had defects in vascularization46 and thus the full deficiency of the major organs, in particular the heart, which seems to present MMP13 in the aortic arch might not be desirable to establish a electrical defects31. Another issue that deserves further investi- phenotype closer to the normality. gation is the relationship between MMP13 and progerin. Both The accumulation of proteoglycans in Progeria mouse mod- in vitro and in vivo results indicate that the silencing of MMP13 els47 as well as in atherosclerotic lesions in HGPS individuals5 has leads to a significant reduction of progerin in SMCs and the been demonstrated. According to our results, the upregulation of reason for this pattern is presently not known. Overall, our study MMP13 in HGPS-SMCs under flow conditions is mediated by demonstrates that the control of MMP13 expression decreases the upregulation of glycocalyx components, which have been the vulnerability of SMCs in large vessels and this strategy may previously implicated as flow shear stress sensors35. The inhibi- be of potential value to reduce the impact of the disease in tion of components of glycocalyx by enzymatic treatment Progeria individuals. decreases significantly the MMP13 levels, the osteogenic program of SMCs and SMCs detachment. Although the connection Methods between MMP13 and glycocalyx has been shown previously for iPSCs culture and differentiation. iPSCs were generated from HGPS skin fibro- non-disease SMCs, we show here that the accumulation of gly- blasts provided by Coriell Institute and characterized according to Nissan et al.10. cocalyx is responsible for the MMP13 expression under shear iPSCs were derived using the Yamanaka’s original method with OCT4, KLF4, SOX2, c-MYC, transferred using retroviral vectors. All HGPS cells were obtained stress conditions, which subsequently leads to the loss of HGPS- from Coriell Institute for Medical Research, which in turn were collected under SMCs. It is possible that the activation of MMP13 expression Institutional Review Board approval and individual informed consent (https:// triggered by an upregulation of glycocalyx is mediated by the www.coriell.org/0/Sections/Support/NIA/Model.aspx?PgId=351). HGPS-iPSCs phosphorylation of ERK and FAK and the activation of c-Jun clone 1 (passages 43-51); HGPS-iPSCs clone 2 (passages 35-42), and N-iPSCs (passages 30-35) were maintained on mitotically inactivated mouse embryonic signaling pathway35 or mediated via NOTCH signaling path- fibroblast (MEF) feeder layer, according to Ferreira et al.43. Culture medium for the way48. Our in vivo results indicated that the expression of present work consisted of 80% KO-DMEM (Life Technologies), 0.5% L-glutamine heparan sulfate proteoglycans in the aortic arches at week 10 on (Life Technologies), 0.2% β-mercaptoethanol (Sigma), 1% nonessential amino acids LmnaG609G/G609GMmp13+/+ mice was not statistically different (Invitrogen), and penicillin-streptomycin (50 U/mL:50 mg/mL) (Lonza), supple- from the expression profile found in WT mice. It is possible mented with 20% KnockOut™ Serum Replacement (Gibco®) and 10 ng/mL of b- FGF (Peprotech). Colonies were expanded by routine passage every 3/4 days with that further time is needed to see this upregulation as seen in 1-mg/ml collagenase type IV (Life Technologies). To induce EBs formation, the other progeroid animal models3,16 or in HGPS individuals5. Since iPSCs were treated with collagenase IV (1 mg/mL, Gibco) for 1 h and then trans- the upregulation of heparan sulfate was not observed in ferred (2:1) to low attachment plates (Corning) containing 10 mL of differentiation LmnaG609G/G609GMmp13+/+ mice, it is not surprising that we medium (80% KO-DMEM (Life Technologies), 20% fetal bovine serum (FBS, Invitrogen), 0.5% L-glutamine (Life Technologies), 0.2% β-mercaptoethanol could not observe a statistical decrease in heparan sulfate in (Sigma), 1% nonessential amino acids (Invitrogen), and penicillin-streptomycin LmnaG609G/G609GMmp13−/− mice. (50 U/mL:50 mg/mL) (Lonza)). EBs were cultured for 10 days at 37 °C, 5% CO2 in The in vivo treatment results presented here using the MMP a humidified atmosphere, with media changes every 2 days. CD34+ cells were inhibitor Batimastat open possibilities for the treatment of HGPS isolated from EBs at day 10 using MACS (Miltenyi Biotec). The percentage of CD34+ cells in EBs was between 0.4 and 1.5%. Isolated cells were grown on 24-well and vascular aging49,50. Batimastat acts as an inhibitor of plates (~3 × 104 cells/cm2) coated with 0.1% gelatin in the presence of EGM-2 metalloproteinase activity by binding the zinc ion in the active (Lonza) supplemented with PDGFBB (50 ng/mL, Prepotech). After four passages, site of MMPs. Batimastat has been used previously for the the medium was replaced by Smooth Muscle Growth Medium-2 (SmGM-2) treatment of human cancer (e.g., malignant ascites51 and (Lonza CC-3182) (maturation medium), for additional four passages. hVSMCs (Lonza) were used as controls for the differentiation studies. Cell cultures were malignant pleural effusions34) with demonstrated results and few maintained at 37 °C, 5% CO2 in a humidified atmosphere, with media changed side-effects in phase I/II clinical trials. Therefore, the current every 2 days. A step-by-step protocol can be found at Protocol Exchange52. study proposes Batimastat as a drug to be considered for future Progeria trials. It should be noted that most of the compounds Cell culture under arterial flow conditions. A suspension of HGPS-iPSC SMCs identified so far in preclinical tests to treat Progeria have been (clone 1), HGPS-iPSC SMCs (clone 2), N-iPSC SMCs, hVSMCs, or HGPS fibro- focused: (i) in the reduction of progerin quantities, by either blasts between 5 × 104 and 1.3 × 105 cells/cm2 was applied to the entry port of an reducing its production or increasing its degradation; (ii) in the IBIDI channel (µ-Slide I 0,4 Luer, or µ-Slide VI 0,4 Luer, IBIDI) and allowed to flow inside by capillary force. After 4 h, a confluent cell layer was formed, which was reduction of progerin toxicity by targeting its aberrant prenyla- then perfused with SmGM-2 medium or fibroblasts medium (DMEM supple- tion: or (iii) in the identification of compounds capable of mented with FBS (20%, v/v, Gibco), sodium pyruvate (Sigma, 1 mM) and restoring pathological phenotypes downstream of progerin penicillin-streptomycin (50 U/mL:50 mg/mL)) at physiological flow rate (20 dyne/ accumulation. Although these treatments showed encouraging cm2). Unless specified, all tests were performed at days 0 and 4 on flow culture conditions. Cell number and cell clumps were determined on slides stained with results in preclinical studies and, in some cases in clinical trials, DAPI (20×) and normalized by image area (0.3524 mm2). Cell clumps areas were they do not address SMC loss over time. The administration of a evaluated by ImageJ software. drug that prevents SMC loss in early stages of disease combined with drugs that further reduce accumulation of progerin and MMP activity. MMP activity was quantified on cell extracts by a fluorometric progerin toxicity could be of added value to extend the lives of red assay kit (Abcam). Cell extracts were obtained by incubating the cells with HGPS individuals. Triton X-100 (0.5%, v/v, in PBS, Sigma) for ~15 min, the cells were centrifuged NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 13 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 and the supernatant collected. Part of cell extract (25 µL) was added to 4- Statistical analysis. Statistical analyses were performed with GraphPad Prism aminophenylmercuric acetate (25 µL, 2 mM) and incubated for 40 min at 37 °C. software. Statistical significance was analyzed using two-tailed unpaired Student’s t Then, a MMP red substrate (50 µL) was added to the mixture and the fluores- test between two different groups. For multiple comparisons, a one-way ANOVA cence intensity measured in a fluorimeter (Ex/Em = 540/590 nm) after 1 h, at analysis followed by Newman–Keuls post test was performed. Results were con- room temperature. An ELISA kit was used to quantify the expression of MMP13 sidered significant when p < 0.05. Data are shown as mean ± SEM unless other protein. Cell culture media collected from different experiments and plasma from specification. WtWt and KiWt mice was used for MMP13 quantification (MMP13 human ELISA kit from Abcam and Mmp13 mouse ELISA kit from USCN) according to Reporting summary. Further information on research design is available in the Nature manufacture recommendation. Briefly, standard or sample (100 µL) was added to Research Reporting Summary linked to this article. each well and incubated for 1 h at 37 °C. Then, solutions were aspirated and detection reagent A (100 µL) was added and incubated for 1 h at 37 °C. After washing three times, detection reagent B (100 µL) was added, incubated 30 min at Data availability 37 °C and washed five times. Substrate solution (90 µL) was then added and left to The microarray datasets generated during and/or analyzed during the current study are incubate for 10–20 min at 37 °C. Finally, stop solution was added to the wells (50 available in the GEO/NCBI (GEO accession: GSE108368). The mass spectrometry µL) and the absorbance of the solution monitored at 450 nm. proteomics data that support the findings of this study have been deposited in the ProteomeXchange Consortium via the PRIDE54 partner repository with the dataset identifier PXD011652. The mass spectrometry proteomics (Tandem Mass Tags—TMT) Glycocalyx analyses. To quantify the intensity of heparan sulfate, cells were data have been deposited to the ProteomeXchange Consortium via the PRIDE54 partner stained with heparan sulfate (1:50 for staining, 10E4 Epitope, USBiological) as repository with the dataset identifier PXD019316. Databases used: Uniprot database described in supplementary information. ImageJ software was used to quantify the (Swissprot entry only, release 2016_01, 16,747 entries); CellAge database (http:// overall intensity of each image, which was then normalized for cell number. Heparinase III from Flavobacterium heparinum (Sigma), was used for the enzy- genomics.senescence.info/cells/). The authors declare that the data supporting the matical degradation of heparan sulfate. Briefly, HGPS-iPSC-SMCs cultured under findings of this study are available within the paper and its supplementary information flow condition during 4 days were subjected to heparinase III treatment (0.5 U/ml files. All the figures have associated source data. No restriction is applied to the data for 30 min at 37 °C), and the number of cells per microfluidic area during culture presented. Source data are provided with this paper. was calculated. Received: 21 November 2018; Accepted: 14 July 2020; Treatment of LmnaG609G/G609G mice with Batimastat. Sixteen LmnaG609G/G609G mice (male and female) were used. After sex and body weight randomization, animals were allocated in different groups and treated with vehicle (eight LmnaG609G/G609G control mice) or BB94 inhibitor (eight LmnaG609G/G609G mice treated with Batimastat in vehicle solution). IP injections were used to administrate 30 mg/kg/day of BB94 at 3 mg/mL in PBS containing 0.01% Tween 80. The References treatment was administered five times per week during 6 weeks (from week 5 to 1. Hennekam R. C. Hutchinson-Gilford progeria syndrome: review of the week 10). The treatment duration was reduced from 10 to 6 weeks due to intra- phenotype. Am. J. Med. Genet. A 140, 2603–2624 (2006). abdominal accumulation of BB94 (precipitate). At the end of week 10 the mice 2. Eriksson M. et al. Recurrent de novo point mutations in lamin A cause were sacrificed, and the selected parameters were evaluated. Hutchinson-Gilford progeria syndrome. Nature 423, 293–298 (2003). 3. Varga R. et al. Progressive vascular smooth muscle cell defects in a mouse model of Hutchinson-Gilford progeria syndrome. Proc. Natl Acad. Sci. USA Double mutant generation and heart rate monitoring. LmnaG609G/G609G mice 103, 3250–3255 (2006). present infertility as described by Osorio and colleagues, therefore the LmnaG609G/ G609G Mmp13−/− mice (KiKO) were generated from LmnaG609G/+ and Mmp13−/+ 4. Hamczyk M. R. et al. Vascular smooth muscle-specific progerin expression accelerates atherosclerosis and death in a mouse model of Hutchinson-Gilford heterozygous (in a C57BL/6 background) as our colony founders (F0). The off- progeria syndrome. Circulation 138, 266–282 (2018). spring presenting the LmnaG609G/+ Mmp13−/+ (F1) were used for further back- 5. Olive M. et al. Cardiovascular pathology in Hutchinson-Gilford progeria: crossing to generate the Progeria double mutants (KiKO) and Progeria control correlation with the vascular pathology of aging. Arterioscler Thromb. Vasc. (KiWT) genotypes used in the present study. All mice were bred in-house in Biol. 30, 2301–2309 (2010). ventilated cages in a temperature and humidity-controlled room with a 12-h light/ 6. Liu, Y., Drozdov, I., Shroff, R., Beltran, L. E. & Shanahan, C. M. Prelamin A dark cycle. The founder LmnaG609G/+ mice were a kind gift from Dr Lopez-Otin20. Genotyping analyses were performed to select those mice carrying the accelerates vascular calcification via activation of the DNA damage response LmnaG609G mutation in homozygosis and the MMP13 deficiency or WT genes. and senescence-associated secretory phenotype in vascular smooth muscle Briefly, DNA was obtained from tails using the PureLink® Genomic DNA Mini Kit cells. Circ. Res. 112, e99–e109 (2013). (Invitrogen) and DNA yields used for the PCR reaction using the Platinum®Taq 7. Zhang, H., Xiong, Z. M. & Cao, K. Mechanisms controlling the smooth muscle DNA Polymerase (Invitrogen) and a combination of custom-designed cell death in progeria via down-regulation of poly(ADP-ribose) polymerase 1. oligonucleotides for the amplification of the Lmna and Mmp13 genes. PCR Proc. Natl Acad. Sci. USA 111, E2261–E2270 (2014). products were run in agarose gels with RedSafe Nucleic Acid Staining Solution 8. Kim P.H. et al. Disrupting the LINC complex in smooth muscle cells reduces (Labotaq) for detecting the amplified Lmna DNA fragments (G609G allele at 240 aortic disease in a mouse model of Hutchinson-Gilford progeria syndrome. bp and WT at 100 bp) and Mmp13 fragments (KO at 1485 bp and WT at 1300 bp). Sci. Transl. Med. 10 1–12 (2018). For heart rate monitoring mice were anesthetized with isoflurane (5% induction 9. Liu G. H. et al. Recapitulation of premature ageing with iPSCs from and 2% maintenance in oxygen) and a mouse paw pulse sensor (Kent Scientific Hutchinson-Gilford progeria syndrome. Nature 472, 221–225 (2011). Corporation) placed in the hindlimb paws until stable heart beats were detected 10. Nissan X. et al. Unique preservation of neural cells in Hutchinson-Gilford and recorded by the PhysioSuiteTM noninvasive monitoring system (Kent progeria syndrome is due to the expression of the neural-specific miR-9 Scientific Corporation). During the procedure and until mice recovered from microRNA. Cell Rep. 2, 1–9 (2012). anesthesia body temperature was controlled with a heating pad. 11. Chen Z. et al. Reprogramming progeria fibroblasts re-establishes a normal All procedures were approved by the Ethics Committee of Animal epigenetic landscape. Aging Cell 16, 870–887 (2017). Experimentation (CCEA 57/16) of the Vall d'Hebron Research Institute and were 12. Frid, M. G., Shekhonin, B. V., Koteliansky, V. E. & Glukhova, M. A. conducted in compliance with Spanish legislation and in accordance with the Phenotypic changes of human smooth muscle cells during development: Directives of the European Union. late expression of heavy caldesmon and calponin. Dev. Biol. 153, 185–193 (1992). 13. Duband, J. L., Gimona, M., Scatena, M., Sartore, S. & Small, J. V. Calponin and Proteomic analysis of aortic arches from WT and mutant mice. Formalin-fixed SM 22 as differentiation markers of smooth muscle: spatiotemporal and paraffin-embedded slices of aortic arch (4 um) were processed for mass distribution during avian embryonic development. Differentiation 55, 1–11 spectrometry analysis as described in the Supplementary Material Information and (1993). according to Heinze et al.33. The obtained peptides were analyzed using Data 14. Vazao, H., das Neves, R. P., Graos, M. & Ferreira, L. Towards the maturation Independent Acquisition53 on an Orbitrap Fusion Lumos mass spectrometer and characterization of smooth muscle cells derived from human embryonic (Thermo Fisher) connected online with a Waters nanoAcquity UPLC system stem cells. PLoS ONE 6, e17771 (2011). (details regarding instrument settings and data acquisition parameters can be found in the Supplementary Information). Spectral library generation, data pro- 15. Scaffidi, P. & Misteli, T. Lamin A-dependent misregulation of adult stem cells cessing, and differential expression analysis were performed in Spectronaut 11 associated with accelerated ageing. Nat. Cell Biol. 10, 452–459 (2008). (Biognosys AG) using default settings. PCA analysis based on the protein report 16. Capell B. C. et al. Inhibiting farnesylation of progerin prevents the table exported from Spectronaut was performed using R version 3.5.0. Data are characteristic nuclear blebbing of Hutchinson-Gilford progeria syndrome. available via ProteomeXchange with identifier PXD011652. Proc. Natl Acad. Sci. USA 102, 12879–12884 (2005). 14 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 ARTICLE 17. Toth J. I. et al. Blocking protein farnesyltransferase improves nuclear shape in 43. Ferreira L. S. et al. Vascular progenitor cells isolated from human embryonic fibroblasts from humans with progeroid syndromes. Proc. Natl Acad. Sci. USA stem cells give rise to endothelial and smooth muscle like cells and form 102, 12873–12878 (2005). vascular networks in vivo. Circ. Res. 101, 286–294 (2007). 18. Mallampalli, M. P., Huyer, G., Bendale, P., Gelb, M. H. & Michaelis, S. 44. Maguire, E. M., Xiao, Q. & Xu, Q. Differentiation and application of induced Inhibiting farnesylation reverses the nuclear morphology defect in a HeLa cell pluripotent stem cell-derived vascular smooth muscle cells. Arterioscler. model for Hutchinson-Gilford progeria syndrome. Proc. Natl Acad. Sci. USA Thromb. Vasc. Biol. 37, 2026–2037 (2017). 102, 14416–14421 (2005). 45. Zhang J. et al. A human iPSC model of Hutchinson Gilford Progeria reveals 19. Chiu, J. J. & Chien, S. Effects of disturbed flow on vascular endothelium: vascular smooth muscle and mesenchymal stem cell defects. Cell Stem Cell 8, pathophysiological basis and clinical perspectives. Physiol. Rev. 91, 327–387 31–45 (2011). (2011). 46. Inada M. et al. Critical roles for collagenase-3 (Mmp13) in development of 20. Osorio F. G. et al. Splicing-directed therapy in a new mouse model of human growth plate cartilage and in endochondral ossification. Proc. Natl Acad. Sci. accelerated aging. Sci. Transl. Med. 3, 106ra107 (2011). USA 101, 17192–17197 (2004). 21. Santiago-Fernandez O. et al. Development of a CRISPR/Cas9-based 47. Capell B. C. et al. A farnesyltransferase inhibitor prevents both the onset and therapy for Hutchinson-Gilford progeria syndrome. Nat. Med. 25, 423–426 late progression of cardiovascular disease in a progeria mouse model. Proc. (2019). Natl Acad. Sci. USA 105, 15902–15907 (2008). 22. Shimizu T. et al. Notch signaling induces osteogenic differentiation and 48. Xiao D. et al. Notch signaling regulates MMP-13 expression via Runx2 in mineralization of vascular smooth muscle cells: role of Msx2 gene induction chondrocytes. Sci. Rep. 9, 15596 (2019). via Notch-RBP-Jk signaling. Arterioscler. Thromb. Vasc. Biol. 29, 1104–1111 49. Cathcart, J. M. & Cao, J. MMP inhibitors: past, present and future. Front (2009). Biosci. (Landmark Ed.) 20, 1164–1178 (2015). 23. Cheng C. et al. Large variations in absolute wall shear stress levels within one 50. Nosoudi N. et al. Prevention of abdominal aortic aneurysm progression by species and between species. Atherosclerosis 195, 225–235 (2007). targeted inhibition of matrix metalloproteinase activity with batimastat-loaded 24. Suo J. et al. Hemodynamic shear stresses in mouse aortas: implications for nanoparticles. Circ. Res. 117, e80–e89 (2015). atherogenesis. Arterioscler Thromb. Vasc. Biol. 27, 346–351 (2007). 51. Parsons, S. L., Watson, S. A. & Steele, R. J. Phase I/II trial of batimastat, a 25. Tacutu R. et al. Human ageing genomic resources: integrated databases and matrix metalloproteinase inhibitor, in patients with malignant ascites. Eur. J. tools for the biology and genetics of ageing. Nucleic Acids Res. 41, Surg. Oncol. 23, 526–531 (1997). D1027–D1033 (2013). 52. Pitrez P. R. et al. Differentiation of human induced-pluripotent stem cells into 26. Knauper V. et al. Cellular mechanisms for human procollagenase-3 (MMP- smooth muscle cells. Protoc. Exch. https://doi.org/10.1038/10.21203/rs.3.pex- 13) activation. Evidence that MT1-MMP (MMP-14) and gelatinase a 964/v1 (2020). (MMP-2) are able to generate active enzyme. J. Biol. Chem. 271, 17124–17131 53. Bruderer R. et al. Extending the limits of quantitative proteome profiling with (1996). data-independent acquisition and application to acetaminophen-treated 27. Wojtowicz-Praga S. et al. Phase I trial of a novel matrix metalloproteinase three-dimensional liver microtissues. Mol. Cell Proteom. 14, 1400–1410 inhibitor batimastat (BB-94) in patients with advanced cancer. Invest. New (2015). Drugs 14, 193–202 (1996). 54. Vizcaino J. A. et al. 2016 update of the PRIDE database and its related tools. 28. Quillard T. et al. Selective inhibition of matrix metalloproteinase-13 increases Nucleic Acids Res. 44, 11033 (2016). collagen content of established mouse atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 31, 2464–2472 (2011). 29. Engel C. K. et al. Structural basis for the highly selective inhibition of MMP- Acknowledgements 13. Chem. Biol. 12, 181–189 (2005). This work was funded by FEDER through the Program COMPETE and by Portuguese 30. Villa-Bellosta R. et al. Defective extracellular pyrophosphate metabolism fund through FCT in context of the projects EXPL/BIM-MED/2267/2013 and POCI- promotes vascular calcification in a mouse model of Hutchinson-Gilford 01-0145-FEDER-029229, as well as the European project ERAatUC (ref. 669088). progeria syndrome that is ameliorated on pyrophosphate treatment. PRP wishes to thank FCT for a BD fellowship (SFRH/BD/71042/2010). AR is sup- Circulation 127, 2442–2451 (2013). ported by the Miguel Servet research contract CPII15/00003 from Instituto de Salud 31. Rivera-Torres J. et al. Cardiac electrical defects in progeroid mice and Carlos III, Spain. The FLI is a member of the Leibniz Association and is financially Hutchinson-Gilford progeria syndrome patients with nuclear lamina supported by the Federal Government of Germany and the State of Thuringia. alterations. Proc. Natl Acad. Sci. USA 113, E7250–E7259 (2016). The authors gratefully acknowledge support from the FLI proteomics core 32. Buczak K. et al. Spatial tissue proteomics quantifies inter- and intratumor facility. The authors would like to thank Dr Carlos Lopez-Otín for providing the heterogeneity in hepatocellular carcinoma (HCC). Mol. Cell Proteom. 17, LmnaG609G/+ mice. 810–825 (2018). 33. Heinze I. et al. Species comparison of liver proteomes reveals links to naked Author contributions mole-rat longevity and human aging. BMC Biol. 16, 82 (2018). P.R.P. and L.F. designed the study, did the literature search and wrote the manu- 34. Macaulay V. M. et al. Phase I study of intrapleural batimastat (BB-94), a script. P.R.P. conducted the study. P.R.P., L.E., and H.V. collected the in vitro data. matrix metalloproteinase inhibitor, in the treatment of malignant pleural P.R.P., G.C., A.R., A.S.-G., K.H., C.N., and N.L. conducted and analyzed in vivo data. effusions. Clin. Cancer Res. 5, 513–520 (1999). D.T. and J.P.M. processed and analyzed raw genomic data. A.-L.E. and X.N. gen- 35. Shi, Z. D., Wang, H. & Tarbell, J. M. Heparan sulfate proteoglycans mediate erated iPSCs from Progeria fibroblasts, provided expertise in the Progeria biology interstitial flow mechanotransduction regulating MMP-13 expression and cell and in the interpretation of the results. L.M., A.B., R.L.S., P.R.P., L.F. and J.C.S. motility via FAK-ERK in 3D collagen. PLoS ONE 6, e15956 (2011). generated the isogenic cell line. T.C. performed pathological evaluation of the tis- 36. Ainslie, K. M., Garanich, J. S., Dull, R. O. & Tarbell, J. M. Vascular smooth sues. D.S. performed proteomics experiments and analyzed the data with the support muscle cell glycocalyx influences shear stress-mediated contractile response. J. of A.O. Appl Physiol. (1985) 98, 242–249 (2005). 37. Knauper V. et al. The role of the C-terminal domain of human collagenase-3 (MMP-13) in the activation of procollagenase-3, substrate specificity, and Competing interests tissue inhibitor of metalloproteinase interaction. J. Biol. Chem. 272, The authors declare no competing interests. 7608–7616 (1997). 38. Austin, K. M., Covic, L. & Kuliopulos, A. Matrix metalloproteases and PAR1 activation. Blood 121, 431–439 (2013). Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 39. Gordon L. B. et al. Clinical trial of a farnesyltransferase inhibitor in children 020-17901-2. with Hutchinson-Gilford progeria syndrome. Proc. Natl Acad. Sci. USA 109, 16666–16671 (2012). Correspondence and requests for materials should be addressed to L.F. 40. Varela I. et al. Combined treatment with statins and aminobisphosphonates extends longevity in a mouse model of human premature aging. Nat. Med. 14, Peer review information Nature Communications thanks the anonymous reviewers for 767–772 (2008). their contribution to the peer review of this work. Peer reviewer reports are available. 41. Song, M., San, H., Anderson, S. A., Cannon, R. O. III & Orlic, D. Shear stress- induced mechanotransduction protein deregulation and vasculopathy in a Reprints and permission information is available at http://www.nature.com/reprints mouse model of progeria. Stem Cell Res. Ther. 5, 41 (2014). 42. Cheung, C., Bernardo, A. S., Pedersen, R. A. & Sinha, S. Directed Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in differentiation of embryonic origin-specific vascular smooth muscle subtypes published maps and institutional affiliations. from human pluripotent stem cells. Nat. Protoc. 9, 929–938 (2014). NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications 15 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17901-2 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. © The Author(s) 2020 16 NATURE COMMUNICATIONS | (2020)11:4110 | https://doi.org/10.1038/s41467-020-17901-2 | www.nature.com/naturecommunications