Sarah is a Teaching Fellow in Pharmacology at the University of Surrey and Honorary Lecturer at University College London (UCL). She teaches across the Biosciences programmes at both Undergraduate and Postgraduate levels.
Prior to joining Surrey, Sarah's research centred around Pulmonary Hypertension and Connective Tissue Diseases. Her research considered the remodelling aspects of these conditions using both in vitro and pre-clinical models.
University roles and responsibilities
- Programme Director BSc Biochemistry
- Programme Director MSc Pharmaceutical Medicine/Clinical Pharmacology
Affiliations and memberships
Courses I teach on
Methods: We investigated the dose-response of JQ1 on SSc and healthy control (HC) donor (ne3) dermal fibroblasts. We assessed the effects on collagen deposition and processing using the Scar-in-a-Jar in vitrofibrosis assay, by western blot and immuno-florescence microscopy for collagen type I (n=4). To determine the effect of JQ1 in a pre-clinical model of skin fibrosis, female C57BL/6 mice were given three weekly subcutaneous injections of 100µl sterile saline (ne6) or 0.1U/ml bleomycin (ne6) for 14 days and treated with 12mg/kg/day JQ1 (ne6) or vehicle (ne6). After 14 days histological analysis for fibrogenic proteins and ECM was performed on skin, and pro-inflammatory chemokines in sera assessed by ELISA.
Results: IL-6 and MCP-1 secretion by SSc and HC donor fibroblasts was significantly (P
Conclusion: We have assessed the functional effects of the Brd inhibitor, JQ1, on SSc dermal fibroblasts and the development of dermal fibrosis in a pre-clinical model of dermal fibrosis. We demonstrate that JQ1 markedly attenuated
including pulmonary arterial hypertension (PAH-SSc), result from endothelial
damage and loss of barrier function. The causes of endothelial dysfunction are
unclear, but the integrity of the endothelium is likely to be significantly diminished
in SSc. Endothelial progenitor cells (EPCs) derived from peripheral blood
mononuclear cells (PBMCs) express endothelial and haematopoietic markers. It
is thought they home to sites of vascular injury and differentiate into endothelial
cells and restore the barrier. In SSc patients circulating levels of EPCs are
reduced. This study aimed to: (i) develop a robust method to isolate and grow
healthy control (HC) and SSc EPCs from PBMCs. (ii) Compare the cellular functions
of EPCs to mature endothelial cells.
Methods. Peripheral blood was taken from HC (n=10) and SSc donors (n=10).
EPCs were cultured from PBMCs, and EPC colonies grown to passage 4. EPCs
and human pulmonary artery endothelial cells (hPAECs) were seeded into transwell
inserts and grown to confluence. Cells were incubated with TNF-± (10ng/
3rd Systemic Sclerosis World Congress Poster Tours - Clinical
ml), and their capacity to form biological barriers and support immune cell influx
was assessed using FITC-albumin (0.5mg/ml) and neutrophil transmigration.
We further assessed the responses of EPCs to TNF-± stimulation by ELISA to
quantify pro-inflammatory cytokine release.
Results. We demonstrate that EPCs form biological barriers with similar capabilities
as mature hPAECs in vitro. TNF-± significantly enhanced permeability
of EPCs (p similar cellular activities as mature endothelial cells, TNFa stimulated
neutrophil transmigration in monolayers of EPCs (p and enhanced the secretion of IL-8 in both EPCs (p We sought to determine the frequency of EPC colony formation from PBMCs
and found no significant difference in the capacity to form EPC colonies in HC
and SSc patient PBMCs.
Discussion. We have developed a robust method for isolating EPCs from PBMCs.
We have demonstrated that endothelial progenitors can maintain an endothelial
barrier consistent with that observed by mature hPAECs in vitro. We
have established that EPCs respond to TNF-± in a similar manner to mature
PAECs, secreting pro-inflammatory cytokines such as IL-8 and supporting neutrophil
characterized by elevated deposition of extracellular matrix (ECM)
proteins, in particular collagen type I. The disease is heterogeneous;
organs commonly affected by fibrosis are the skin, kidney, lung and
heart. Vascular complications include pulmonary arterial hypertension
(PAH), which occurs in 12?40% of patients. CD14þmonocytes are afunctionally heterogeneous cell type. They have been noted to
differentiate into a number of cell phenotypes including macrophages
and collagen-producing fibrocytes. In culture fibrocytes adopt a
spindle shape, co-express haematopoietic CD45RO and 25F9, along
with mesenchymal markers including aSMA and collagen type I.
Fibrocytes amplify the inflammatory/immune response through distinct
mechanisms, including antigen presentation, cytokine and chemokine
secretion, and the production of MMPs. We and others have shown
fibrocyte differentiation is enhanced by fibrogenic cytokines including
PDGF. Here we seek to understand the mechanism by which SSc
fibrocytes influence the local microenvironment of the tissue.
Methods: CD14þ peripheral blood mononuclear cells (PBMCs) were
isolated from SSc patient and healthy control blood. PBMCs were
cultured in the presence of macrophage colony stimulating factor
(MCSF; n > 10) and/or endothelin-1 (ET-1; n > 10); after 14 days of
culture number of fibrocytes was assessed. The effect of pharmacological
inhibitors including ETRA and ETRB antagonism on fibrocyte
differentiation (n ¼ 6 SSc and control) was investigated. Secreted
factors in culture media from SSc and control fibrocytes were
assessed by ELISA (n ¼ 6), and the effects of conditioned media
explored in 3D-collagen gel.
Results: Both MCSF and ET-1 significantly induced fibrocyte
differentiation, in combination differentiation was significantly augmented
(P more readily differentiated from CD14þ PBMCs than healthy control
donors in response to MCSF (P MCSF with ET-1 in combination (P BQ123 and BQ788 (respectively), and Bosentan (a dual ETR
antagonist) inhibited MCSF induced fibrocyte differentiation in a
concentration dependant manner. Furthermore SSc fibrocytes
secreted significantly more connective tissue growth factor (CTGF)
than control fibrocytes (P fibrocytes acting in a paracrine manner, conditioned me
The development of pulmonary arterial hypertension in scleroderma remains an important contributor to mortality in this condition, despite substantial improvements in outcomes due to modern therapeutic strategies. No animal models of scleroderma develop this important complication. We describe the constitutive vascular phenotype of a mouse model of scleroderma and show that pulmonary endothelial injury replicates the pathological changes of pulmonary arterial hypertension seen in human disease.
The T²RIIk-fib mouse strain expresses a kinase-deficient type II transforming growth factor ² (TGF²) receptor driven by a fibroblast-specific promoter leading to ligand-dependent upregulation of TGF² signalling; this mouse strain replicates key fibrotic features of scleroderma. We did structural, biochemical, and functional assessments of pulmonary and systemic vessels, including in-vivo haemodynamic studies, before and after vascular endothelial growth factor receptor (VEGFR) inhibition with SU5416, which induced pulmonary endothelial cell apoptosis. These assessments included biochemical analysis of the TGF², endothelin, and VEGF signalling axes in vivo; tissue sections; and explanted pulmonary arterial smooth muscle cells.
In the T²RIIk-fib mouse strain, a constitutive pulmonary vasculopathy with medial thickening, a perivascular proliferating chronic inflammatory cell infiltrate, and mildly raised pulmonary artery pressures resemble the well-described chronic hypoxia model of pulmonary hypertension. After administration of SU5416, the pulmonary vascular phenotype was more florid, with pulmonary arteriolar luminal obliteration by apoptosis-resistant proliferating endothelial cells; the result was right ventricular hypertrophy confirming haemodynamically significant pulmonary arterial hypertension. Altered TGF², endothelin, and ligand and receptor expression of VEGF were consistent with a scleroderma phenotype.
Interpretation This study replicates key features of scleroderma-associated pulmonary arterial hypertension in a mouse model. Our results suggest that pulmonary endothelial cell injury in a genetically susceptible mouse strain triggers this complication and support functional interplay between TGF², endothelin, and VEGF that provides insight into pathogenesis.
Abstract Vascular complications associated with systemic sclerosis (SSc) including pulmonary arterial hypertension (PAH-SSc), result from endothelial damage and loss of barrier function. Endothelial progenitor cells (EPCs) express endothelial (VEGFR2+, CD31+ ) and haematopoietic (CD133+ ) markers. They home to sites of vascular injury and differentiate into endothelial cells restoring the endothelium. In SSc patients circulating levels of EPCs are reduced. This study aimed to develop a robust method to grow EPCs from peripheral blood mononuclear cells (PBMCs) and to compare cellular functions to mature endothelial cells.
Methods EPCs and human pulmonary artery endothelial cells (hPAECs) were seeded into transwell inserts and grown to confluence. Cells were incubated with TNFa (50ng/ml), and their capacity to form biological barriers assessed using FITC-albumin (5mg/ml). FITC-albumin ?leak? was quantified by fluorescent absorbance over time. We further assessed the responses of EPCs to TNFa stimulation by ELISA to quantify pro-inflammatory cytokine release.
Results EPCs form a biological exclusion barrier with similar capabilities as mature hPAECs. TNFa significantly enhanced the permeability of EPCs (P
Discussion We developed a robust method for isolating EPCs from PBMCs. We have demonstrated that EPCs can maintain an endothelial barrier consistent with that observed by mature hPAECs in vitro. We have established that EPCs respond to TNFa in a similar manner to mature PAECs. We have shown no significant difference in the capacity of PBMCs from SSc patients to form EPC colonies compared to HCs.
complication of SSc that occurs in around 10% of cases. We have
previously shown that a TGFb dependent transgenic mouse strain
(TbRIIk-fib) is susceptible to organ based pathology relevant to SSc
including development of pulmonary hypertension (PH) after pharmacological pulmonary endothelial injury by SU5416, a VEGF receptor inhibitor. In this study, we have prevented the development of PH in this mouse strain using macitentan, a novel dual ETA/B receptor
antagonist recently licensed to treat PAH in CTD based upon a
significant effect on morbidity and mortality in PAH.
Methods: SU5416, a VEGF receptor inhibitor, was administered to all TbRIIk-fib transgenic (TG) mice and wild-type (WT) littermate
controls to induce endothelial injury with subsequent endothelial
proliferation and PH in transgenic mice only. Mice were treated with
either 50 mg/kg macitentan daily by oral gavage or vehicle alone (n
¼ 8 for each group) either before SU5416 injection or on day 8 following
injection. The development of PH in each group was assessed by
histology and immunohistochemistry of vessel architecture, in vivo
haemodynamic studies and RV mass index measurements.
Biochemical responses to TGFb, endothelin and VEGF stimulation
before and after macitentan were examined in cultured T
bRII k-fib lung fibroblasts.
Results: Compared with WT littermates, after SU5416, all TG mice
developed a prominent perivascular chronic inflammatory infiltrate and
smooth muscle layer hypertrophy, as previously described. RV mass
index was elevated in TG animals receiving vehicle compared with
other groups (TG vehicle 0.29 0.007, TG macitentan 0.24
0.007, P 28.8 mmHg 0.72, TG macitentan 22.0 1.62, P any significant change in systemic arterial blood pressure in any
group. Treatment with macitentan after day 8 was sufficient to
normalize haemodynamic consequences of SU5416 administration.
There was obliterative pulmonary arteriolar occlusion in 21% of
vessels in TG mice treated with vehicle. In contrast, no vessels in
WT mice or TG mice treated with macitentan developed this
histological change. Explanted TG lung fibroblasts showed an increase
in proliferation and migration with upregulation of VEGF and TGFb
signalling and downregulation of endothelin receptor A compared with
of systemic sclerosis (SSc) that occurs in around 10% of cases. We have previously shown that an imbalance
between TGFbeta and BMPRII signalling
in the transgenic mouse strain T²RIIk-fib
contributes to the development of PH following pulmonary endothelial injury. In
this study, we have both prevented and treated PH in this mouse strain using macitentan,
an endothelin receptor antagonist licensed to treat PAH in connective tissue
Methods: SU5416 was administered to all T²RIIk-fib transgenic (TG) mice and littermate wildtype (WT) animals to induce endothelial injury with
subsequent endoluminal proliferation and PH in
transgenic mice only. Mice were treated daily with either macitentan or vehicle
alone (n=8 each group) from either 2 days before or 8 days following SU5416
administration to assess prevention or treatment respectively. The development
of PH in each group was assessed by histology and immunohistochemistry of vessel
architecture, in vivo haemodynamic studies and RV mass index measurements
following 3 weeks of treatment. Microarray analysis of right ventricular tissue
: All TG mice developed a perivascular inflammatory
infiltrate and smooth muscle layer hypertrophy after SU5416 administration. RV mass index was elevated in TG
animals receiving vehicle compared to other groups. In particular
co-administration of macitentan to TG animals treated with SU5416 resulted in
normal RV mass (TG vehicle 0.29±0.007, TG macitentan at day -2 0.24±0.007,
p SU5416 was abrogated by macitentan (TG vehicle 28.8±3.2, TG+macitentan
at day -2 22.0±2.9, TG+macitentan at day +8, 24.4±1.8,
p Explanted TG lung fibroblasts showed an increase in TGFbeta signalling and downregulation of BMPRII compared with
WT littermates following macitentan treatment. Pulmonary
arteriolar occlusion occurred in 21% of vessels in TG mice treated with vehicle
with no occlusion in any other vessels. Gene expression analysis of whole right
ventricle showed alterations in key genes known to be associated with cardiac
muscle remodelling and failure. Figure 1 shows the
cluster analysis of TG mice treated with SU5416 compared with those also
treated with macitentan.
Macitentan prevents and treats the development of
Oncostatin M (OSM) is a pleiotropic member of the gp130/ IL-6 cytokine family,
produced by a variety of immune cells, including macrophages, neutrophils and
activated T cells. OSM signals through two receptors; gp130/ LIF low affinity
receptor (LIFR) and gp130/ OSM high affinity receptor (OSMR); activating
JAK/STAT, ERK1/2 and p38 MAPK pathways. The rationale for the role of OSM in
Systemic Sclerosis (SSc) lies in that key disease components, i.e.
inflammation, vascular dysregulation and fibrosis, complement the biological
activities of the target.
Methods: Serum OSM levels
were measured by custom OSM ELISA in 63 diffuse cutaneous (Dc) SSc and 18
age-matched healthy donors (HD). Transcriptomics data derived from published
DcSSc skin biopsies were mined for OSM and known OSM-regulated genes. The presence of OSM, OSMR, and phospho(p)STAT3 in skin
biopsies from HD (n=12), involved limited cutaneous (Lc)SSc (n=12), and
involved and uninvolved DcSSc (n=13) patients was assessed by
immunohistochemistry (IHC). Primary dermal fibroblasts isolated from HD or
DcSSc skin were incubated with OSM at 2, 20 or 200ng/ml and cell associated
collagen type I (Col-1), and connective tissue growth factor (CTGF) assessed by
Western blot analysis at 24hrs and 72hrs.
Results: Serum OSM levels
were significantly (P=0.004) increased in DcSSc patients compared with HDs (Table
1). OSM and OSM-regulated genes, including S100A9 and VCAM-1, were upregulated
in the DcSSc skin. IHC analysis confirmed OSM, OSMR and pSTAT3 expression in
skin biopsies from SSc patients and HDs, with a similar expression pattern in
both. SSc skin biopsies exhibited significantly higher inflammatory cell
infiltrates compared to HDs, and were strongly positive for OSM, OSMR and
pSTAT3. Uninvolved DcSSc skin had significantly more OSM positive immune cell
infiltrates and increased pSTAT3 expression than HDs, but not compared to
involved DcSSc skin. Involved DcSSc skin exhibited significantly more pSTAT3
positive fibroblasts compared to HDs (p positive immune cell infiltrates was further significantly enhanced in involved
compared to uninvolved tissue. Increased numbers of positive OSM and pSTAT3
fibroblasts were present in involved versus uninvolved skin. OSM induced CTGF
after 24hrs in 4/5 DcSSc dermal fibroblast cell lines tested. No induction
above baseline was observed in the two HD dermal fibroblast cell lines.
Following 72hrs stimulation,
Methods CD14+ PBMCs were isolated from SSc and healthy donor blood. Fibrocyte differentiation in the presence of MCSF and/or ET-1 was assessed after 14 days. The effect of endothelin receptor (ETR) antagonists (selective/dual) on fibrocyte differentiation (n = 6) was investigated. SSc and control fibrocyte secretomes were assessed by ELISA (n = 6), and the effects on fibroblast-mediated gel contraction determined.
Results MCSF and ET-1 alone and in combination induced fibrocyte differentiation (P
Discussion CD14+ SSc PBMCs readily differentiate into fibrocytes in response to ET-1 and MCSF via ETRA and ETRB. Our data suggests fibrocytes contribute to the development of PAH in SSc via a paracrine mechanism modulating the functional activities of resident tissue fibroblasts.
Vascular complications in Scleroderma (SSc) patients are associated with high mortality, particularly in patients who develop pulmonary arterial hypertension (SSc-PAH). Vascular complications, thought to arise from initial activation and dysfunction of the endothelium can lead to: elevated vascular leak, inflammation, mesenchymal hypertrophy by activation of resident smooth muscle cells and fibroblasts, and neointima formation. Recent studies suggest that as well as resident mesenchymal cells, endothelial cells can undergo endothelial-mesenchymal transition (EndoMT), and acquire a mesenchymal phenotype which may contribute to the expansion of the mesenchymal cell population. Here we sought to determine the prevalence of EndoMT in SSc-PAH patients and pre-clinical models of PAH, and assess the cellular effects on pulmonary artery endothelial cells (PAECs) functions.
Using lung tissue from SSc-PAH patients (n=3), healthy control (HC) donors (n=3), and from the hypoxia/SU5416 pre-clinical murine model of PAH (n=5), EndoMT was determined by immunofluorescence based on co-expression of vWF and ±SMA. EndoMT was induced in human PAECs (n=3) in vitro by TNF± [5ng/ml], IL-1² [0.1ng/m;] and TGF² [5ng/ml] in combination. Morphological changes were assessed by light microscopy and phalloidin staining. Western blotting and immunofluorescence was used to quantify: CD31, vWF, occludin, VE-cadherin, ±SMA, calponin and collagen type 1 expression. Conditioned media was collected from PAECs, PAECs following treatment to initiate EndoMT and SSc-PAH and HC fibroblasts; levels of inflammatory secretion was quantified by MSD arrays. The capacity of homogenous EndoMT monolayers (n=6) and mixed cultures of 1:10 EndoMT:PAECs (n=6) cells to form exclusion barriers was assessed using trans-well permeability FITC-albumin assays.
Co-localisation of vWF and ±SMA was observed in d5% of pulmonary arteries from SSc-PAH patients and hypoxia/SU5416 mice. PAECs treated with TNF±, IL-1² and TGF² exhibited significant changes in morphology, loss of endothelial markers and elevated expression of mesenchymal markers by day 6. There was a significant (P0.01) 5-fold increase in permeability compared to PA
Vascular complications are a key pathological feature of systemic sclerosis (SSc) affecting the microcirculation and arterioles. Under normal circumstances the endothelium acts as a biological barrier supporting controlled permeability, immune surveillance and cellular trafficking. In SSc endothelial damage, contributes to barrier dysfunction and elevated immune cell infiltration and inflammation. The cause(s) of the initial endothelial dysfunction in SSc is unclear. Blood outgrowth endothelial cells (BOECs) are thought to home to sites of vascular injury and differentiate into endothelial cells, aiding the repair and restoration of normal endothelial functions. Circulating levels of BOECs have been shown to be reduced in SSc patients, and recent studies suggest that BOECs may be dysfunctional in vascular diseases. Here we sought to assess SSc-BOECs and their contribution to vascular dysfunction in SSc.
BOECs were established from peripheral blood PBMCs from healthy donors (HD) and SSc patients. BOECs from HD (n=5) and SSc patients (n=6) were profiled using Illumina HT12 gene arrays and secreted inflammatory cytokines profiled by meso discovery scale (MSD) arrays. The capacity of SSc-BOECs (n=4) and HD-BOECs (n=4) to: 1) Establish biological barriers alone or in co-culture with mature endothelial cells, was assessed using electric cell-substrate impedance sensing (ECIS); 2) Support PBMC (n=4) trans-endothelial migration, by forming monolayers in 24-well transwell inserts stimulated with or without TNF± (10ng/ml).
Results: Using gene set enrichment analysis, it was determined that SSc-BOECs exhibit a significantly altered gene expression profile including inflammatory chemokines and cytokines. In addition SSc-BOECs exhibited significantly elevated pro-inflammatory cytokine secretion compared to HD-BOECs, including IL-6 (P
We have demonstrated that BOECs isolated from SSc patients exhibit a significantly altered gene profile compared to that
Methods: SU5416, a VEGF receptor inhibitor, was administered to all T²RIIk-fib transgenic (TG) mice and littermate wildtype (WT) animals to induce endothelial injury with subsequent endoluminal proliferation and PH in transgenic mice only. Mice were treated with either 50mg/kg macitentan daily by oral gavage or vehicle alone (n=8 each group). The development of PH in each group was assessed by histology and immunohistochemistry of vessel architecture, in vivo haemodynamic studies and RV mass index measurements.
Results: Compared with WT littermates, after SU5416, all TG mice developed a prominent perivascular chronic inflammatory infiltrate and smooth muscle layer hypertrophy, as previously described. RV mass index was elevated in TG animals receiving vehicle compared to other groups (TG vehicle 0.29±0.007, TG macitentan 0.24±0.007, p
Conclusion: Macitentan prevents the development of histological and haemodynamic PH in this mouse model of SSc. These findings support a pivotal role for perturbed endothelin activity in a model that is induced by altered TGFbeta signalling and triggered by experimental VEGF inhibition. It underpins the value of this model as a platform for expe