Serotonin transporter, sex, and hypoxia: microarray analysis in the pulmonary arteries of mice identifies genes with relevance to human PAH

Abstract

Pulmonary arterial hypertension (PAH) is up to threefold more prevalent in women than men. Female mice overexpressing the serotonin transporter (SERT; SERT+ mice) exhibit PAH and exaggerated hypoxia-induced PAH, whereas male SERT+ mice remain unaffected. To further investigate these sex differences, microarray analysis was performed in the pulmonary arteries of normoxic and chronically hypoxic female and male SERT+ mice. Quantitative RT-PCR analysis was employed for validation of the microarray data. In relevant groups, immunoblotting was performed for genes of interest (CEBPβ, CYP1B1, and FOS). To translate clinical relevance to our findings, CEBPβ, CYP1B1, and FOS mRNA and protein expression was assessed in pulmonary artery smooth muscle cells (PASMCs) derived from idiopathic PAH (IPAH) patients and controls. In female SERT+ mice, multiple pathways with relevance to PAH were altered. This was also observed in chronically hypoxic female SERT+ mice. We selected 10 genes of interest for qRT-PCR analysis (FOS, CEBPβ, CYP1B1, MYL3, HAMP2, LTF, PLN, NPPA, UCP1, and C1S), and 100% concordance was reported. Protein expression of three selected genes, CEBPβ, CYP1B1, FOS, was also upregulated in female SERT+ mice. Serotonin and 17β-estradiol increased CEBPβ, CYP1B1, and FOS protein expression in PASMCs. In addition, CEBPβ, CYP1B1, and FOS mRNA and protein expression was also increased in PASMCs derived from IPAH patients. Here, we have identified a number of genes that may predispose female SERT+ mice to PAH, and these findings may also be relevant to human PAH.

pulmonary arterial hypertension (PAH) is characterized by both remodeling and vasoconstriction of the pulmonary vasculature. Mutations in the gene encoding for the bone morphogenetic protein receptor type-2 (BMPR-II) are the predominant genetic cause of heritable PAH (HPAH) (27). BMPR-II mutations have been described in up to 80% of HPAH and 20% of idiopathic PAH (IPAH) patients. Despite this, penetrance is relatively low in BMPR-II mutation carriers as <20% of those actually develop PAH (26). Therefore, it is recognized that “second hit” risk factors contribute to disease pathogenesis.

A sex bias exists for both HPAH and IPAH, with females up to threefold more likely to present with disease (16, 28, 43). Despite this, the underlying reasons for these sex differences remain obscure. Estrogens are one possible risk factor in PAH. The use of oral contraceptives has been associated with the development of PAH (23). Genotyping studies have also revealed alterations in estrogen signaling in PAH. For example, female PAH patients exhibit increased expression levels of estrogen receptor-1, which is the gene encoding for estrogen receptor alpha, compared with unaffected females (30). Decreased expression of the estrogen-metabolizing enzyme cytochrome P450 1B1 (CYP1B1), leading to impaired estrogen metabolism, has also been described in female BMPR-II mutated PAH patients compared with unaffected female BMPR-II carriers (2).

Multiple studies have implicated serotonin in the pathobiology of PAH. In mice, peripheral serotonin synthesis is required for the development of both hypoxia-induced PAH (25) and dexfenfluramine-induced PAH (5), whereas the exogenous administration of serotonin uncovers a PAH phenotype in BMPR-II mutant mice (20). Mice overexpressing the SERT (SERT+ mice) also develop PAH and exaggerated hypoxia-induced PAH (21). Consistent with this, mice with targeted SERT overexpression in the PASMCs under the guidance of its own SM22 promoter also develop PAH and severe hypoxia-induced PAH (8). SERT expression is increased in human pulmonary artery smooth muscle cells (PASMCs) derived from IPAH patients, and this increased expression mediates enhanced serotonin-induced proliferation in these cells (6). Taken together, this evidence highlights the critical role of smooth muscle-SERT in mediating serotonin effects in experimental and human PAH.

In this study, we investigated genotypic differences in the development of PAH in SERT+ mice. Female SERT+ mice develop PAH and exaggerated hypoxia-induced PAH, whereas male SERT+ mice remain unaffected compared with their respective wild-type (WT) controls. This was only apparent at 5 mo of age. This experimental model of PAH is the first to exhibit female susceptibility and may provide insight into the female bias observed in human PAH. To investigate genotypic changes associated with the development and progression of PAH, microarray analysis was performed in the pulmonary arteries of 2 mo old SERT+ mice. Genes of interest were further assessed via quantitative RT-PCR and immunoblotting. With relevance to human PAH, this was also investigated in human PASMCs.

METHODS

Ethical information.

All animal procedures conform with the United Kingdom Animal Procedures Act (1986) and with the “Guide for the Care and Use of Laboratory Animals” published by the US National Institutes of Health (NIH publication no. 85-23, revised 1996). Animal approval was granted by the University Committee Board. Experimental procedures using human PASMCs conform to the principles outlined in the Declaration of Helsinki.

SERT+ mice.

Female and male SERT+ mice (2 and 5 mo of age) were generated as previously described (21). Where appropriate, mice were exposed to 14 days of hypobaric hypoxia (equivalent to 10% O₂). Age-matched C57B/6J×CBA littermate mice were studied as controls.

Hemodynamic measurements.

Heart rate, right ventricular pressure, and systemic arterial pressure were measured and analyzed as previously described (17). Briefly, right ventricular pressure was measured via transdiaphragmatic right heart catheterization and systemic arterial pressure was measured via cannulation of the left common carotid artery. Hemodynamic measurements from six to eight mice for each group were assessed.

Lung histology.

Sagittal sections of lung were elastica-Van Gieson stained and microscopically assessed for the muscularization of pulmonary arteries (<80 μm external diameter) in a blinded fashion as previously described (17). Remodeled arteries were confirmed by the presence of a double elastic laminae. Lung sections from five mice for each group were studied. Approximately 150 arteries from each lung section (∼750 arteries in total for each group) were assessed.

Right ventricular hypertrophy.

Right ventricular hypertrophy (RVH) was assessed by weight measurement of the right ventricular free wall (RV) and left ventricle plus septum (LV+S). The ratio expressed is RV/LV+S. RVH measurements from six to eight mice for each group were assessed.

Human PASMCs.

Human PASMCs were provided by Prof N. W. Morrell (University of Cambridge, Cambridge, UK). Briefly, PASMCs were derived from the pulmonary arteries (1–3 mm internal diameter) of three nonfamilial IPAH patients. PASMCs derived from macroscopically normal lung biopsies (pulmonary arteries, 1–3 mm internal diameter) excised from non-PAH donors were studied as control. The homogeneity of PASMCs was confirmed via cell morphology and positive staining for α-smooth muscle actin. PASMCs (passage 3–5) were seeded in six-well plates at a density of 20,000 per/well and grown to 80% confluence in Dulbecco's modified eagle medium (GIBCO) supplemented with 10% fetal bovine serum (Sera Laboratories International) prior to quiescence for 24 h. Where appropriate, PASMCs were stimulated with 1 μmol/l serotonin or 1 nmol/l 17β-estradiol for 24 h. Cell lysates were prepared for immunoblotting as previously described (5). All experiments were performed in triplicate.

Microarray analysis.

Microarray analysis was performed in the pulmonary arteries of female and male WT and SERT+ mice at 2 mo of age. To investigate the development and progression of PAH in SERT+ mice we assessed genotypic differences at 2 mo of age, where no PAH phenotype was reported. One advantage to this is the exclusion of possible compensatory gene expression changes associated with end-stage PAH, which may be apparent in SERT+ mice at 5 mo of age where a PAH phenotype is reported. This was also repeated in mice following exposure to chronic hypoxia. For the purpose of biological replicates, pulmonary arteries from individual mice were subject to microarray analysis (n = 4 each group, total 32 microarrays, Accession number E-MTAB-455). Total RNA was extracted from the main, left, and right pulmonary arteries using the TissueLyser and RNeasy Fibrous Mini Kit (Qiagen). RNA was subjected to an additional DNase purification step to eliminate genomic DNA contamination (RNase-free DNase, Qiagen). RNA integrity and quantification were assessed using the NanoDrop ND-1000 Spectrophotometer (Nano-Drop Technologies, Wilmington, DE) and Agilent 2100 Bioanalyzer system (Agilent Technologies). Absorbance of the RNA samples was quantified at 260 and 280 nm, and the 260/280 ratio was calculated. All samples showed a 260/280 ratio ≥ 1.9 and RNA integrity number ≥ 8.0, which was indicative of RNA purity. Complementary RNA (cRNA) was synthesized from RNA using the Illumina TotalPrep RNA Amplification Kit (Applied Biosystems). Briefly, cDNA was synthesized from 200 ng RNA with a T7 oligo(dT) RNA polymerase promoter. After second-strand synthesis, in vitro transcription was performed resulting in the synthesis of biotin-labeled antisense cRNA. For microarray analysis, 750 ng cRNA was hybridized to the Illumina MouseRef-8 v1.1 Expression BeadChip using the Whole-Genome Expression Direct Hybridization Kit and scanning performed with a BeadStation 500GX (Illumina). Of the total probe-sets (∼25,600) in each microarray, at least 24,613 were detected in each of the 32 arrays performed. Data were analyzed with BeadStudio software (Illumina). Hybridization signal strength was normalized to the median array and expression levels determined using the Average Normalization Beadstudio algorithm. For identifying differentially expressed genes, the following parameters recommended by Illumina were used: P value <0.05, diff. score >15, average signal >100.

Quantitative real-time PCR.

Quantitative real-time PCR (qRT-PCR) was employed for validation of the microarray data. Gene-specific primers corresponding to the PCR targets were designed based on published sequences in GenBank using the primer 3 program and synthesized by IDT (Integrated DNA Technologies; supplementary table).¹ We confirmed the absence of nonspecific amplification by examining PCR products by agarose gel electrophoresis, ensuring amplification of single discrete bands with no primer-dimers. Real-time PCR was carried out in a DNA Engine OPTICON2 (MJ Research). Each reaction was performed according to the Brilliant II SYBRGreen PCR Master Mix (Agilent Technologies) protocol using 10 ng of RNA. Three replicates were performed for each sample plus template-free samples as negative controls. Cycling parameters consisted of an initial reverse transcription step for 30 min at 50°C, followed by a 10 min incubation at 95°C to fully activate the DNA polymerase and 40 amplification cycles at 95°C for 30 s, 56°C/58°C for 30 or 40 s, and 72°C for 30 s. Fluorescence measurements were assessed at the end of the annealing phase at 78, 82, and 86°C. The CT values were determined using the Opticon2 software, and the total amount of RNA was normalized against β-actin. Data are expressed as fold-change.

Western blotting.

Immediately following death, the pulmonary arteries were snap-frozen in liquid N₂ and stored at −80°C until use. To obtain a sufficient concentration of protein for analysis, arteries from four mice were suspended in 250 μl lysis buffer (50 mmol/l Tris pH 7.4, 1 mmol/l DTT, 1× complete-protease inhibitor tablet; Roche Diagnostics, West Sussex, UK) and homogenized using a microrotary blade.

Human PASMCs (passage 3–5) were derived from both IPAH patients and control subjects and solubilized in RIPA buffer. Immunoblotting was performed as previously described (5). Experiments were repeated in triplicate, and α-tubulin was used for protein loading control. Densitometrical analysis was performed using TotalLab TL100 software. Data are expressed as the ratio of protein density to α-tubulin density.

Statistical analysis.

Data were analyzed by a two-way ANOVA followed by Bonferroni's post hoc test, one-way ANOVA followed by Dunnett's post hoc test or unpaired t-test as appropriate. Data are expressed as means ± SE.

RESULTS

Assessment of PAH.

In normoxia, right ventricular systolic pressure (RVSP), pulmonary vascular remodeling (PVR, % remodeled vessels), and right ventricular hypertrophy (RVH) were similar in 2 mo old male and female WT and SERT+ mice. However, at 5 mo of age female SERT+ mice exhibited PAH, as assessed by increased RVSP and RVP (Fig. 1), whereas male SERT+ mice remained unaffected (Fig. 2). Following exposure to chronic hypoxia, all groups developed hypoxia-induced PAH as assessed by significant increases in RVSP, PVR, and RVH. However, at 2 mo of age, chronically hypoxic female SERT+ mice exhibited increased RVH compared with WT mice. Similarly, at 5 mo of age these mice exhibited increased RVSP, PVR and RVH compared with 5 mo WT mice. Exaggerated hypoxia-induced PAH was not apparent in male SERT+ mice at 2 or 5 mo of age. There were no systemic effects reported in both female and male normoxic and chronically hypoxic SERT+ mice compared with WT mice, as assessed by no changes in systemic arterial pressure or heart rate (data not shown).

Fig. 1.

Right ventricular systolic pressure (RVSP, n = 6–8; A), pulmonary vascular remodeling (PVR, n = 5; B), and right ventricular hypertrophy (RVH, n = 6–8; C) measurements in female wild-type (WT) and SERT+ mice at 2 (2M) and 5 mo (5M) of age, in both normoxic and chronic hypoxia. In normoxia, no pulmonary arterial hypertension (PAH) phenotype was observed in 2 mo female SERT+ mice, however, was apparent at 5 mo of age, as assessed by increased RVSP and PVR. Hypoxia-induced elevations of RVSP, PVR and RVH were observed in all groups; 5 mo old SERT+ mice exhibit increased RVSP, PVR, and RVH. Data are expressed as means ± SE and analyzed by 2-way ANOVA followed by Bonferroni's post hoc test. *P < 0.05, **P < 0.01, cf. normoxic mice; §P < 0.05, §§P < 0.01, cf. WT mice.

Fig. 2.

RVSP (n = 6–8, A), PVR(n = 5, B), and RVH (n = 6–8, C) measurements in male WT and SERT+ mice at 2 (2M) and 5 mo (5M) of age, in both normoxic and chronic hypoxia. In both normoxia and chronic hypoxia, 2 and 5 mo old male SERT+ mice exhibit similar RVSP, PVR, and RVH compared with their respective WT controls. Data are expressed as means ± SE and analyzed by 2-way ANOVA followed by Bonferroni's post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001 cf. normoxic mice.

Genotypic differences in SERT+ mice.

We were interested in exploring the genotypic differences associated with the development and progression of PAH in female SERT+ mice. In total, we identified a total of 155 genes that were significantly (P < 0.05) differentially expressed in female SERT+ mice compared with their WT controls; 71 genes show increased expression (Table 1), while the remaining 84 genes show reduced expression (Table 2). To determine their biological relevance, we functionally categorized these genes by biological processes. A considerable number of these genes (>40%) were assigned to one or more biological processes, of which 15 categories were present in total (Fig. 3). Specifically, a large number of these genes were assigned to biological functions with relevance to PAH. These included oxidation-reduction, cell differentiation, regulation of transcription, apoptosis, muscle contraction, cellular calcium ion homeostasis, and glycolysis.

Table 1.

List of genes upregulated in the pulmonary arteries of 2 mo old female SERT+ mice compared with 2 mo old female wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Oxidation Reduction
CYP2S1	cytochrome P450, family 2, subfamily s, polypeptide 1	NM_028775.2	1.52	0.046
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	2.51	0.009
FASN	fatty acid synthase	scl014104.1_1	1.70	0.044
ALDH1A7	aldehyde dehydrogenase family 1, subfamily A7	scl52665.13.1_14	2.20	0.018
GPD1	glycerol-3-phosphate dehydrogenase 1	NM_010271.2	2.74	0.010
CYP1B1	cytochrome P450, family 1, subfamily b, polypeptide 1	scl49594.5.189_22	1.54	0.037
Cell Differentiation
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.50	0.049
LGALS3	lectin, galactose binding, soluble 3	NM_010705.1	1.76	0.022
DMKN	dermokine	scl32804.19.1_0	1.92	0.017
Regulation of Transcription
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.50	0.049
FOS	FBJ osteosarcoma oncogene	NM_010234.2	2.79	0.017
Xbp1	X-box binding protein 1	NM_013842.2	1.56	0.029
HOXA4	homeo box A4	NM_008265.2	2.02	0.018
HOXB5	homeo box B5	NM_008268.1	1.90	0.015
AXUD1	AXIN1 upregulated 1	scl35215.8_496	1.87	0.017
Immune Response
CFD	complement factor D	NM_013459.1	1.86	0.045
SPON2	spondin 2, extracellular matrix protein	NM_133903.2	1.76	0.017
Apoptosis
CIDEC	cell death-inducing DFFA-like effector c	NM_178373.2	2.91	0.014
SRGN	Serglycin	scl019073.1_109	1.62	0.021
AXUD1	AXIN1 upregulated 1	scl35215.8_496	1.87	0.017
Metabolic Process
UGT1A10	UDP glycosyltransferase 1 family, polypeptide A10	scl0394435.7_126	1.66	0.046
ACLY	ATP citrate lyase	NM_134037.2	1.68	0.044
FASN	fatty acid synthase	scl014104.1_1	1.70	0.044
ALDH1A7	aldehyde dehydrogenase family 1, subfamily A7	scl52665.13.1_14	2.20	0.018
AACS	acetoacetyl-CoA synthetase	NM_030210.1	1.94	0.026
GPD1	glycerol-3-phosphate dehydrogenase 1	NM_010271.2	2.74	0.010
UAP1	UDP-N-acetylglucosamine pyrophosphorylase 1	NM_133806.2	1.74	0.026
Lipid Metabolic Process
SCD1	stearoyl-Coenzyme A desaturase 1	NM_011182.2	2.51	0.009
AACS	acetoacetyl-CoA synthetase	NM_030210.1	1.94	0.026
Lipid Biosynthetic Process
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	2.51	0.009
ACLY	ATP citrate lyase	NM_134037.2	1.68	0.044
FASN	fatty acid synthase	scl014104.1_1	1.70	0.044
ELOVL6	ELOVL family member 6	scl00170439.1_29	2.16	0.014
Brown Fat Cell Differentiation
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	2.51	0.009
ADIPOQ	adiponectin, C1Q and collagen domain containing	scl49310.3_131	2.66	0.017
UCP1	uncoupling protein 1	NM_009463.2	15.14	0.000
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.50	0.049
BC054059	cDNA sequence BC054059	scl19994.5.1_11	2.54	0.017
Glycolysis
ENO2	enolase 2, gamma neuronal	NM_013509.2	1.61	0.045

Table 2.

List of genes downregulated in the pulmonary arteries of 2 mo old female SERT+ mice compared with 2 mo old female wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Oxidation Reduction
SC4MOL	sterol-C4-methyl oxidase-like	NM_025436.1	2.09	0.034
PRDX2	peroxiredoxin 2	NM_011563.2	2.89	0.017
Cell Differentiation
TRIM54	tripartite motif-containing 54	NM_021447.1	2.58	0.026
OBSCN	obscurin, cytoskeletal calmodulin and titin-interacting RhoGEF	scl40175.7.1_79	2.13	0.042
CSRP3	cysteine and glycine-rich protein 3	NM_013808.3	5.14	0.003
Regulation of Transcription
TBX20	T-box 20	NM_194263.1	2.06	0.041
Immune Response
PRG4	proteoglycan 4	scl000882.1_25	2.57	0.019
Apoptosis
ACTC1	actin, alpha, cardiac	NM_009608.1	2.17	0.037
COMP	cartilage oligomeric matrix protein	scl33728.21.1_0	4.73	0.011
Metabolic Process
PGAM2	phosphoglycerate mutase 2	scl40555.3.1_120	8.59	0.003
Lipid Metabolic Process
CPT1B	carnitine palmitoyltransferase 1b, muscle	NM_009948.1	2.81	0.022
LPL	lipoprotein lipase	scl0016956.1_234	1.92	0.047
Heart Development
MB	myoglobin	NM_013593.2	22.71	0.000
TNNI3	troponin I, cardiac	NM_009406.2	6.86	0.003
MYL2	myosin, light polypeptide 2, regulatory, cardiac, slow	scl27267.9.1_12	70.40	0.000
TNNT2	troponin T2, cardiac	NM_011619.1	3.18	0.015
Muscle Contraction
MYBPC3	myosin binding protein C, cardiac	NM_008653.1	3.34	0.014
ACTN2	actinin alpha 2	NM_016798.2	9.71	0.003
TBX20	T-box 20	NM_194263.1	2.06	0.041
MYOM2	myomesin 2	scl34033.37.1_91	2.40	0.028
TTN	titin	scl19104.8.1_3	6.48	0.003
TNNT2	troponin T2, cardiac	NM_011619.1	3.18	0.015
Lipid Biosynthetic Process
SC4MOL	sterol-C4-methyl oxidase-like	NM_025436.1	2.09	0.034
Cellular Calcium Ion Homeostasis
PLN	phospholamban	scl38924.3_494	6.93	0.003
TNNI3	troponin I, cardiac	NM_009406.2	6.86	0.003
CSRP3	cysteine and glycine-rich protein 3	NM_013808.3	5.14	0.003
Brown Fat Cell Differentiation
MB	myoglobin	NM_013593.2	22.71	0.000
Glycolysis
ENO3	enolase 3, beta muscle	NM_007933.2	3.82	0.010
PGAM2	phosphoglycerate mutase 2	scl40555.3.1_120	8.59	0.003
Sarcomere Organization
MYBPC3	myosin binding protein C, cardiac	NM_008653.1	3.34	0.014
MYH6	myosin, heavy polypeptide 6, cardiac muscle, alpha	scl46291.1.1_325	2.60	0.018
TTN	Titin	scl19104.8.1_3	6.48	0.003
TNNT2	troponin T2, cardiac	NM_011619.1	3.18	0.015
Regulation of Heart Contraction
MYBPC3	myosin binding protein C, cardiac	NM_008653.1	3.34	0.014
HRC	histidine rich calcium binding protein	NM_010473.1	3.71	0.011
MYH6	myosin, heavy polypeptide 6, cardiac muscle, alpha	scl46291.1.1_325	2.60	0.018
TNNT2	troponin T2, cardiac	NM_011619.1	3.18	0.015

Fig. 3.

Hierarchical cluster analysis of the differentially expressed genes in female and male WT and SERT+ mice (A). Representation of the differentially expressed genes in female SERT+ mice (B) and male SERT+ mice (C), arranged by biological processes.

To further investigate the genotypic changes underlying these sex differences in SERT+ mice, we also performed microarray analysis in the pulmonary arteries of male SERT+ mice. We observed that a total of 148 genes were significantly differentially expressed in male SERT+ mice compared with male WT mice. Of these, 110 genes were increased (Table 3), whereas the remaining 38 genes were decreased (Table 4). When categorized by biological processes, only 25% of these genes were assigned to biological function, and nine categories were represented in total.

Table 3.

List of genes upregulated in the pulmonary arteries of 2 mo old male SERT+ mice compared with 2 mo old male wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Transport
SCNN1G	sodium channel, nonvoltage-gated 1 gamma	scl32105.12.878_39	1.90	0.029
CLIC6	chloride intracellular channel 6	NM_172469.1	2.28	0.009
SLC39A4	solute carrier family 39	NM_028064.2	2.04	0.021
RAB25	RAB25, member RAS oncogene family	scl21969.5.1_66	2.04	0.021
GABRP	gamma-aminobutyric acid	scl0001520.1_108	2.83	0.002
Ltf	lactotransferrin	NM_008522.2	11.43	0.000
Transport: Oxygen Transport
MB	myoglobin	NM_013593.2	2.74	0.000
HBB-B2	hemoglobin, beta adult minor chain	NM_016956.2	2.35	0.000
Signal Transduction
FCER1G	Fc receptor, IgE, high affinity I, gamma polypeptide	scl15940.5.1_15	2.68	0.002
GRB7	growth factor receptor bound protein 7	scl40936.14_9	1.86	0.021
Cell Adhesion
MUC4	mucin 4	scl0140474.33_123	1.82	0.031
WISP2	WNT1 inducible signaling pathway protein 2	NM_016873.1	1.92	0.010
SPON2	spondin 2, extracellular matrix protein	NM_133903.2	1.54	0.045
Oxidation Reduction
CYP2E1	cytochrome P450, family 2, subfamily e, polypeptide 1	NM_021282.1	1.98	0.008
CYP2F2	cytochrome P450, family 2, subfamily f, polypeptide 2	scl32906.13.1_13	12.08	0.000
CYP2A5	cytochrome P450, family 2, subfamily a, polypeptide 5	NM_009997.1	3.32	0.004
CYP4A12B	cytochrome P450, family 4, subfamily a, polypeptide 12B	scl013118.12_302	4.46	0.001
ABP1	amiloride binding protein 1	NM_029638.1	1.62	0.044
GPX2	glutathione peroxidase 2	NM_030677.1	2.20	0.004
ALDH3A1	aldehyde dehydrogenase family 3, subfamily A1	NM_007436.1	2.10	0.013
ALDH1A1	aldehyde dehydrogenase family 1, subfamily A1	scl011668.12_94	2.55	0.027
PRDX2	peroxiredoxin 2	NM_011563.2	2.25	0.012
Immune Response
PGLYRP1	peptidoglycan recognition protein 1	scl33021.4.288_87	2.08	0.035
CXCL15	chemokine	scl27600.3.1_4	5.76	0.000
SPON2	spondin 2, extracellular matrix protein	NM_133903.2	1.54	0.045
Metabolic Process
ALDH3A1	aldehyde dehydrogenase family 3, subfamily A1	NM_007436.1	2.10	0.013
ALDH1A1	aldehyde dehydrogenase family 1, subfamily A1	scl011668.12_94	2.55	0.027
GSTA3	glutathione S-transferase, alpha 3	scl18127.10.1_92	2.14	0.014
GSTO1	glutathione S-transferase omega 1	NM_010362.1	2.92	0.004
Hemopoiesis
CXCL15	chemokine	scl27600.3.1_4	5.76	0.000
Regulation of Transcription
IRX5	Iroquois related homeobox 5	NM_018826.2	1.94	0.025
OTX1	orthodenticle homolog 1	scl40460.6_595	1.72	0.038
IRX3	Iroquois related homeobox 3	scl34499.5.1_0	1.69	0.024
FOXA1	forkhead box A1	scl42430.2_236	1.84	0.015

Table 4.

List of genes downregulated in the pulmonary arteries of 2 mo old male SERT+ mice compared with 2 mo old male wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Transport: Oxygen Transport
HBB-B1	hemoglobin, beta adult major chain	NM_008220.2	2.41	0.019
Signal Transduction
LGR6	leucine-rich repeat-containing G protein-coupled receptor 6	scl00329252.1_132	1.86	0.011
Oxidation Reduction
JARID1B	lysine (K)-specific demethylase 5B	scl17448.26_107	1.65	0.032
SC4MOL	sterol-C4-methyl oxidase-like	NM_025436.1	2.25	0.003
Hemopoiesis
PICALM	phosphatidylinositol binding clathrin assembly protein	scl32408.23_56	3.80	0.001
HBB-B1	hemoglobin, beta adult major chain	NM_008220.2	2.41	0.019
Regulation of Transcription
TEF	thyrotroph embryonic factor	scl0002562.1_0	1.82	0.022
DBP	D site albumin promoter binding protein	NM_016974.1	2.45	0.002
PER2	period homolog 2	NM_011066.1	1.89	0.013

Hierarchal cluster analysis between the four normoxic groups (258 genes in total) revealed distinct gene expression patterns between female SERT+ and female WT that were not apparent in the identical male SERT+ and WT comparison.

Genotypic differences in hypoxic SERT+ mice.

We were also interested in investigating the genotypic differences associated with exaggerated hypoxia-induced PAH in female SERT+ mice. Following exposure to chronic hypoxia, female SERT+ mice exhibited a greater than twofold increase in the number of differentially expressed genes compared against the identical normoxic comparison. In total, 316 genes were differentially expressed. We observed that 254 genes were increased (Table 5), while the remaining 62 genes showed decreased expression (Table 6). When arranged by biological processes, 53% of genes were assigned to a total of 26 distinct pathways. Moreover, a significant number of these dysregulated pathways observed in chronically hypoxic female SERT+ mice have been previously associated with PAH including apoptosis, inflammation, transcription, and metabolism (Fig. 4).

Table 5.

List of genes upregulated in the pulmonary arteries of 2 mo old hypoxic female SERT+ mice compared with 2 mo old hypoxic female wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Apoptosis
PGLYRP1	peptidoglycan recognition protein 1	scl33021.4.288_87	3.30	0.000
CIDEC	cell death-inducing DFFA-like effector c	NM_178373.2	1.84	0.000
SRGN	serglycin	scl019073.1_109	1.32	0.048
KRT8	keratin 8	scl0016691.1_37	3.13	0.000
CIDEA	cell death-inducing DNA fragmentation factor, alpha subunit-like effector A	NM_007702.1	2.38	0.000
Induction of Apoptosis
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.67	0.004
CIDEC	cell death-inducing DFFA-like effector c	NM_178373.2	1.84	0.000
ERN2	endoplasmic reticulum	scl30713.22.1_242	2.08	0.003
Brown Fat Cell Differentiation
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	1.97	0.000
ADIPOQ	adiponectin, C1Q and collagen domain containing	scl49310.3_131	1.87	0.025
UCP1	uncoupling protein 1	NM_009463.2	2.60	0.000
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.67	0.004
BC054059	cDNA sequence BC054059	scl19994.5.1_11	1.86	0.004
NUDT7	nudix NM_024446.2 1.57
MRAP	melanocortin 2 receptor accessory protein	NM_029844.1	1.96	0.001
ALDH6A1	aldehyde dehydrogenase family 6, subfamily A1	NM_134042.1	1.65	0.021
PPARG	peroxisome proliferator activated receptor gamma	NM_011146.1	1.75	0.000
Carbohydrate Metabolic Process
AMY1	amylase 1, salivary scl077379.3_13 1.66
PDK4	pyruvate dehydrogenase kinase, isoenzyme 4	scl29310.11_209	2.95	0.005
PYGL	liver glycogen phosphorylase	NM_133198.1	1.86	0.000
GPD1	glycerol-3-phosphate dehydrogenase 1	NM_010271.2	2.00	0.001
CHST1	carbohydrate	NM_023850.1	1.42	0.031
PPP1R3C	protein phosphatase 1, regulatory	NM_016854.1	1.54	0.025
KLB	klotho beta	scl27771.5.1_89	1.87	0.005
Cell Adhesion
MYBPC2	myosin binding protein C, fast-type	NM_146189.1	1.34	0.031
LGALS3BP	lectin, galactoside-binding, soluble, 3 binding protein	scl39273.6_263	1.66	0.011
1110049B09RIK	RIKEN cDNA 1110049B09 gene	scl42544.15.6_29	1.39	0.017
CDH5	cadherin 5	NM_009868.4	1.75	0.030
CD93	CD93 antigen	scl18542.4.1_65	1.99	0.034
Cell-Cell Adhesion
CDH5	cadherin 5	scl33446.12_65	1.75	0.030
CD93	CD93 antigen	scl18542.4.1_65	1.99	0.034
Chemotaxis
CYSLTR1	cysteinyl leukotriene receptor 1	NM_021476.2	1.33	0.041
CMTM8	CKLF-like MARVEL transmembrane domain containing 8	NM_027294.1	1.34	0.047
CXCL12	chemokine	scl0001073.1_120	1.77	0.007
Defense Response to Bacterium
PGLYRP1	peptidoglycan recognition protein 1	scl33021.4.288_87	3.30	0.000
HAMP2	hepcidin antimicrobial peptide 2	NM_183257.1	5.38	0.000
FCER1G	Fc receptor, IgE, high affinity I, gamma polypeptide	scl15940.5.1_15	2.29	0.016
H2-K1	histocompatibility 2, K1, K region	scl0014972.1_210	1.61	0.016
DNA Replication
POLN	DNA polymerase N	scl0272158.1_149	1.69	0.031
SUPT16H	suppressor of Ty 16 homolog	NM_033618.1	1.76	0.021
POLK	polymerase	scl43651.15_178	1.29	0.047
Immune Response
PGLYRP1	peptidoglycan recognition protein 1	scl33021.4.288_87	3.30	0.000
CXCL12	chemokine	scl0001073.1_120	1.77	0.007
CFD	complement factor D	NM_013459.1	1.68	0.007
CD300LG	CD300 antigen like family member G	scl40868.8_408	1.72	0.049
H2-T23	histocompatibility 2, T region locus 23	NM_010398.1	1.37	0.036
H2-K1	histocompatibility 2, K1, K region	scl0014972.1_210	1.61	0.016
Inflammatory Response
REG3G	regenerating islet-derived 3 gamma	NM_011260.1	2.97	0.049
CHST1	carbohydrate	NM_023850.1	1.42	0.031
KNG1	kininogen 1	NM_023125.2	1.79	0.006
PPARG	peroxisome proliferator activated receptor gamma	NM_011146.1	1.75	0.000
Integrin-Mediated Signaling Pathway
ADAM9	a disintegrin and metallopeptidase domain 9	NM_007404.1	1.39	0.018
Lipid Biosynthetic Process
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	1.97	0.000
PCX	pyruvate carboxylase	scl000483.1_20	1.74	0.002
ACLY	ATP citrate lyase	NM_134037.2	1.81	0.009
FASN	fatty acid synthase	scl014104.1_1	1.97	0.004
ELOVL6	ELOVL family member 6, elongation of long chain fatty acids	scl00170439.1_29	2.86	0.000
ELOVL5	ELOVL family member 5, elongation of long chain fatty acids	NM_134255.2	1.39	0.031
DGAT2	diacylglycerol O-acyltransferase 2	scl31009.10_67	2.20	0.000
Lipid Metabolic Process
HSD11B1	hydroxysteroid 11-beta dehydrogenase 1	scl000857.1_11	1.55	0.054
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	1.97	0.000
HADHB	hydroxyacyl-Coenzyme A dehydrogenase/3-ketoacyl-Coenzyme A thiolase/enoyl-Coenzyme A hydratase	NM_145558.1	1.52	0.001
CPT1B	carnitine palmitoyltransferase 1b, muscle	NM_009948.1	1.94	0.013
AACS	acetoacetyl-CoA synthetase	NM_030210.1	1.47	0.028
CPT2	carnitine palmitoyltrasferase 2	scl000022.1_12	1.56	0.006
ACADVL	acyl-Coenzyme A dehydrogenase, very long chain	scl40004.19.1_140	1.41	0.042
ACAA2	acetyl-Coenzyme A acyltransferase 2	scl0002163.1_25	1.40	0.033
ACADL	acyl-Coenzyme A dehydrogenase, long-chain	NM_007381.2	2.12	0.000
ACSM3	acyl-CoA synthetase medium-chain family member 3	scl000249.1_5	1.67	0.010
DGAT2	diacylglycerol O-acyltransferase 2	scl31009.10_67	2.20	0.000
PNPLA2	patatin-like phospholipase domain containing 2	scl8719.1.1_106	1.79	0.006
LPL	lipoprotein lipase	scl0016956.1_234	1.86	0.007
ADIPOR2	adiponectin receptor 2	scl28480.7_231	1.50	0.041
CIDEA	cell death-inducing DNA fragmentation factor, A alpha subunit-like effector	NM_007702.1	2.38	0.000
Metabolic Process
HSD11B1	hydroxysteroid 11-beta dehydrogenase 1	scl000857.1_11	1.55	0.005
PCX	pyruvate carboxylase	scl000483.1_20	1.74	0.002
ACLY	ATP citrate lyase	NM_134037.2	1.81	0.009
FASN	fatty acid synthase	scl014104.1_1	1.97	0.004
AMY1	amylase 1, salivary	scl077379.3_13	1.66	0.005
AGPAT2	1-acylglycerol-3-phosphate O-acyltransferase 2	NM_026212.1	2.76	0.002
ALAS2	aminolevulinic acid synthase 2, erythroid	scl54562.12.1_64	1.29	0.034
HADHB	hydroxyacyl-Coenzyme A dehydrogenase/3-ketoacyl-Coenzyme A thiolase/enoyl-Coenzyme A hydratase	NM_145558.1	1.52	0.001
AACS	acetoacetyl-CoA synthetase	NM_030210.1	1.47	0.028
GPD1	glycerol-3-phosphate dehydrogenase 1	NM_010271.2	2.00	0.000
EPHX2	epoxide hydrolase 2, cytoplasmic	scl45408.20.1_29	1.48	0.021
ACO2	aconitase 2, mitochondrial	NM_080633.1	1.89	0.000
UAP1	UDP-N-acetylglucosamine pyrophosphorylase 1	NM_133806.2	3.48	0.000
ACADVL	acyl-Coenzyme A dehydrogenase, very long chain	scl40004.19.1_140	1.41	0.042
GSTA3	glutathione S-transferase, alpha 3	scl18127.10.1_92	2.43	0.000
GSTO1	glutathione S-transferase omega 1	NM_010362.1	2.01	0.034
ACAA2	acetyl-Coenzyme A acyltransferase 2	scl0002163.1_25	1.40	0.033
EYA3	eyes absent 3 homolog	NM_010166.2	2.10	0.002
NAT8L	N-acetyltransferase 8-like	scl27919.3_374	1.47	0.040
ALDH6A1	aldehyde dehydrogenase family 6, subfamily A1	NM_134042.1	1.65	0.021
ACADL	acyl-Coenzyme A dehydrogenase, long-chain	NM_007381.2	2.12	0.000
ACSM3	acyl-CoA synthetase medium-chain family member 3	scl000249.1_5	1.67	0.010
ACSS1	acyl-CoA synthetase short-chain family member 1	NM_080575.1	1.47	0.012
PNPLA2	patatin-like phospholipase domain containing 2	scl8719.1.1_106	1.79	0.006
GSTA4	glutathione S-transferase, alpha 4	NM_010357.1	2.61	0.000
Muscle Contraction
MYBPC2	myosin binding protein C, fast-type	NM_146189.1	1.34	0.031
TBX20	T-box 20	NM_194263.1	1.57	0.034
PPARG	peroxisome proliferator activated receptor gamma	NM_011146.1	1.75	0.000
Oxidation Reduction
CYP2S1	cytochrome P450, family 2, subfamily s, polypeptide 1	NM_028775.2	1.56	0.034
HSD11B1	hydroxysteroid 11-beta dehydrogenase 1	scl000857.1_11	1.55	0.005
SCD1	stearoyl-Coenzyme A desaturase 1	scl52445.7_23	1.97	0.000
GPX2	glutathione peroxidase 2	NM_030677.1	1.53	0.045
FASN	fatty acid synthase	scl014104.1_1	1.97	0.000
GPX3	glutathione peroxidase 3	NM_008161.1	1.55	0.007
CYP2E1	cytochrome P450, family 2, subfamily e, polypeptide 1	NM_021282.1	2.09	0.003
GPD2	glycerol phosphate dehydrogenase 2, mitochondrial	NM_010274.2	1.52	0.007
ETFDH	electron transferring flavoprotein, dehydrogenase	NM_025794.1	1.64	0.047
DLD	dihydrolipoamide dehydrogenase	NM_007861.2	1.41	0.046
GPD1	glycerol-3-phosphate dehydrogenase 1	NM_010271.2	2.00	0.000
ALDH3A1	aldehyde dehydrogenase family 3, subfamily A1	NM_007436.1	1.92	0.003
ACADVL	acyl-Coenzyme A dehydrogenase, very long chain	scl40004.19.1_140	1.41	0.046
CYP2F2	cytochrome P450, family 2, subfamily f, polypeptide 2	scl32906.13.1_13	8.61	0.000
CYP2A5	cytochrome P450, family 2, subfamily a, polypeptide 5	NM_009997.1	1.85	0.022
CYP4A12B	cytochrome P450, family 4, subfamily a, polypeptide 12B	scl013118.12_302	3.42	0.000
ALDH6A1	aldehyde dehydrogenase family 6, subfamily A1	NM_134042.1	1.65	0.021
ACADL	acyl-Coenzyme A dehydrogenase, long-chain	NM_007381.2	2.12	0.000
PRDX2	peroxiredoxin 2	NM_011563.2	6.22	0.000
Response to Toxin
EPHX2	epoxide hydrolase 2, cytoplasmic	scl45408.20.1_29	1.48	0.021
CES3	carboxylesterase 3	scl34490.14.1_30	2.18	0.002
CYP2F2	cytochrome P450, family 2, subfamily f, polypeptide 2	scl32906.13.1_13	8.61	0.000
PON1	paraoxonase 1	NM_011134.1	2.09	0.001
Signal Transduction
RERG	RAS-like, estrogen-regulated, growth-inhibitor	NM_181988.1	2.02	0.001
GPR109A	G protein-coupled receptor 109A	NM_030701.1	1.61	0.009
CYSLTR1	cysteinyl leukotriene receptor 1	NM_021476.2	1.33	0.042
ELTD1	EGF, latrophilin seven transmembrane domain containing 1	NM_133222.1	2.08	0.001
FCER1G	Fc receptor, IgE, high affinity I, gamma polypeptide	scl15940.5.1_15	2.29	0.016
Small GTPase-Mediated Signal Transduction
KNDC1	kinase noncatalytic C-lobe domain	scl31927.18.1_9	2.41	0.002
RAB25	RAB25, member RAS oncogene family	scl21969.5.1_66	1.31	0.047
G3BP2	GTPase activating protein	scl0023881.1_86	1.33	0.021
Temperature Homeostasis
GPX2	glutathione peroxidase 2	NM_030677.1	1.53	0.045
ACADVL	acyl-Coenzyme A dehydrogenase, very long chain	scl40004.19.1_140	1.41	0.042
ACADL	acyl-Coenzyme A dehydrogenase, long-chain	NM_007381.2	2.12	0.001
CIDEA	cell death-inducing DNA fragmentation factor, alpha subunit-like effector A	NM_007702.1	2.38	0.000
Transcription
FOS	FBJ osteosarcoma oncogene	NM_010234.2	2.34	0.000
SUPT16H	suppressor of Ty 16 homolog	NM_033618.1	1.76	0.021
KLF5	Kruppel-like factor 5	scl45215.1.1_294	1.99	0.001
CEBPB	CCAAT/enhancer binding protein	NM_009883.1	1.67	0.002
TBX20	T-box 20	NM_194263.1	1.57	0.034
PPARG	peroxisome proliferator activated receptor gamma	NM_011146.1	1.75	0.000
ZFP367	zinc finger protein 367	NM_175494.2	1.22	0.031
EYA3	eyes absent 3 homolog	NM_010166.2	2.10	0.002
Transport
UCP1	uncoupling protein 1	NM_009463.2	2.60	0.000
SLC5A6	solute carrier family 5	scl26744.20.688_4	1.38	0.024
APOC1	apolipoprotein C-I	NM_007469.2	1.95	0.001
NDUFB4 NADH	dehydrogenase (ubiquinone) 1 beta subcomplex 4	scl48493.1_0	1.34	0.028
ETFB	electron transferring flavoprotein, beta polypeptide	NM_026695.2	1.29	0.044
ETFA	electron transferring flavoprotein, alpha polypeptide	NM_145615.2	1.42	0.043
ETFDH	electron transferring flavoprotein, dehydrogenase	NM_025794.1	1.64	0.047
CPT1B	carnitine palmitoyltransferase 1b, muscle	NM_009948.1	1.94	0.013
CPT2	carnitine palmitoyltrasferase 2	scl000022.1_12	1.56	0.007
RAB25	RAB25, member RAS oncogene family	scl21969.5.1_66	1.31	0.047
FXYD3	FXYD domain-containing ion transport regulator 3	NM_008557.1	1.84	0.003
GABRP	gamma-aminobutyric acid	scl0001520.1_108	2.35	0.022
ATP5K	ATP synthase, H+ transporting, mitochondrial F1F0 complex, subunit e	scl011958.2_29	1.49	0.033
G3BP2	GTPase activating protein	scl0023881.1_86	1.33	0.021
MFI2	antigen p97	scl0001844.1_62	1.45	0.014
SLC25A1	solute carrier family 25	NM_153150.1	1.71	0.002
MTCH2	mitochondrial carrier homolog 2	NM_019758.2	1.41	0.033
Triglyceride Metabolic Process
APOC1	apolipoprotein C-I	NM_007469.2	1.95	0.001
Other
CISH	cytokine inducible SH2-containing protein	scl012700.3_170	1.60	0.004
STMN2	stathmin-like 2	scl23400.7_310	1.65	0.000
SOCS3	suppressor of cytokine signaling 3	NM_007707.2	1.41	0.022
TUBA8	tubulin, alpha 8	scl29555.5_307	1.71	0.020
DNAHC2	dynein, axonemal, heavy chain 2	scl40034.27.1_30	1.57	0.004
SCGB1A1	secretoglobin, family 1A, member 1	NM_011681.1	9.55	0.000
HSPA5	heat shock 70 kDa protein 5	NM_022310.2	1.75	0.005
DUSP1	dual specificity phosphatase 1	NM_013642.1	1.64	0.021
DUSP23	dual specificity phosphatase 23	NM_026725.2	1.45	0.044
BMPER	BMP-binding endothelial regulator	NM_028472.1	1.95	0.004

Table 6.

List of genes downregulated in the pulmonary arteries of 2 mo old hypoxic female SERT+ mice compared with 2 mo old hypoxic female wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Brown Fat Cell Differentiation
MB	myoglobin	NM_013593.2	1.89	0.003
Carbohydrate Metabolic Process
IGF2	insulin-like growth factor 2	scl30469.7_1	1.82	0.015
RPE	ribulose-5-phosphate-3-epimerase	scl0227227.1_0	2.13	0.019
Cell Adhesion
VCAN	versican	scl013003.1_89	1.11	0.025
STAB1	stabilin 1	NM_138672.1	1.39	0.030
SELP	selectin, platelet	NM_011347.1	1.38	0.021
ITGA3	integrin alpha 3	NM_013565.2	1.25	0.017
ITGB1	integrin beta 1	NM_010578.1	1.72	0.008
Cell-Cell Adhesion
TNXB	tenascin XB	NM_031176.1	1.71	0.021
Cellular Iron Ion Homeostasis
LTF	lactotransferrin	NM_008522.2	11.84	0.000
ALAS2	aminolevulinic acid synthase 2, erythroid	scl54562.12.1_64	1.29	0.033
HAMP2	hepcidin antimicrobial peptide 2	NM_183257.1	5.38	0.000
MFI2	antigen p97	scl0001844.1_62	1.45	0.014
Chemotaxis
CCL21B	chemokine	scl0018829.1_65	2.04	0.026
DNA Replication
NFIC	nuclear factor I/C	scl068530.1_6	2.03	0.005
RBBP4	retinoblastoma binding protein 4	scl24919.4.1_260	2.09	0.021
Heart Development
MB	myoglobin	NM_013593.2	1.89	0.004
MYL2	myosin, light polypeptide 2, regulatory, cardiac, slow	scl27267.9.1_12	6.36	0.001
VCAN	versican	scl013003.1_89	1.11	0.025
OSR1	oxidative-stress responsive 1	scl35223.18_513	1.53	0.007
Immune Response
H2-EA	histocompatibility 2, class II antigen E alpha	NM_010381.2	1.17	0.030
CCL21B	chemokine	scl0018829.1_65	2.04	0.026
Inflammatory Response
STAB1	stabilin 1	NM_138672.1	1.39	0.030
CCL21B	chemokine	scl0018829.1_65	2.04	0.026
SELP	selectin, platelet	NM_011347.1	1.38	0.021
Integrin-Mediated Signaling Pathway
ITGA3	integrin alpha 3	NM_013565.2	1.25	0.018
ITGB1	integrin beta 1	NM_010578.1	1.72	0.009
Lipid Biosynthetic Process
PRKAG2	protein kinase, AMP-activated, gamma 2 noncatalytic subunit	NM_145401.1	2.38	0.021
Lipid Metabolic Process
PTPN11	protein tyrosine phosphatase, nonreceptor type 11	NM_011202.2	1.88	0.040
TNXB	tenascin XB	NM_031176.1	1.71	0.022
Metabolic Process
RPE	ribulose-5-phosphate-3-epimerase	scl0227227.1_0	2.13	0.019
Muscle Contraction
TTN	titin	scl19104.8.1_3	1.18	0.037
ZBTB7A	zinc finger and BTB domain containing 7a	scl0016969.1_242	3.43	0.002
Oxidation Reduction
4933406E20RIK	RIKEN cDNA 4933406E20 gene	NM_028944.2	1.24	0.009
Signal Transduction
RAP2C	RAP2C, member of RAS oncogene family	scl54266.5_48	2.00	0.025
Transcription
CREBBP	CREB binding protein	scl48815.9.1_11	2.07	0.033
SKI	superkiller viralicidic activity 2-like (S.cerevisiae)	scl23441.8_64	1.98	0.003
RBBP4	retinoblastoma binding protein 4	scl24919.4.1_260	2.09	0.021
NFIC	nuclear factor I/C	scl068530.1_6	2.03	0.005
Transport
MB	myoglobin	NM_013593.2	1.89	0.004
LTF	lactotransferrin	NM_008522.2	11.84	0.000
TRAM1	translocating chain-associating membrane protein 1	NM_028173.1	1.72	0.036
RAMP1	receptor	scl17654.5.1_10	1.54	0.034
RAB17	RAB17, member RAS oncogene family	NM_008998.2	2.33	0.006
Triglyceride Metabolic Process
PTPN11	protein tyrosine phosphatase, nonreceptor type 11	NM_011202.2	1.88	0.040
TNXB	tenascin XB	NM_031176.1	1.71	0.022
Other
GUCY1A3	guanylate cyclase 1, soluble, alpha 3	scl0060596.1_205	1.03	0.011
BMX	BMX nonreceptor tyrosine kinase	NM_009759.2	1.62	0.030
KIF1B	kinesin family member 1B	scl0002773.1_49	2.29	0.021
ZBTB7A	zinc finger and BTB domain containing 7a	scl0016969.1_242	3.43	0.002
PRRX1	paired related homeobox 1	scl018933.1_11	1.78	0.021

Fig. 4.

Hierarchical cluster analysis of the differentially expressed genes in female and male WT and SERT+ mice following exposure to chronic hypoxia (A). Representation of the differentially expressed genes in female SERT+ mice (B) and male SERT+ mice (C), arranged by biological processes.

In contrast, a large number of these changes were not apparent in hypoxic male SERT+ mice. A total of 145 genes were differentially expressed in male SERT+ mice, with 87 showing increased expression (Table 7) and 58 showing decreased expression (Table 8). When categorized by biological processes, 42% of these genes were assigned a biological function; 12 categories were represented in total.

Table 7.

List of genes upregulated in the pulmonary arteries of 2 mo old hypoxic male SERT+ mice compared with 2 mo old hypoxic male wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Transport
LBP	lipopolysaccharide binding protein	scl20002.16.7_10	1.92	0.019
LTF	lactotransferrin	NM_008522.2	6.63	0.000
APOC1	apolipoprotein C-I	NM_007469.2	1.95	0.020
SLC4A1	solute carrier family 4	NM_011403.1	2.28	0.004
HBB-B1	hemoglobin, beta adult major chain	NM_008220.2	3.05	0.032
ABCC9	ATP-binding cassette, sub-family C	scl0001165.1_35	2.25	0.015
SLC1A3	solute carrier family 1	NM_148938.2	1.85	0.011
LCN2	lipocalin 2	NM_008491.1	1.65	0.021
Signal Transduction, Protein Binding
FGL1	fibrogen-like protein 1	scl34887.7.1_10	2.48	0.007
ANGPTL4	angiopoietin-like 4	scl50042.7_106	1.89	0.019
TNC	tenascin C	scl0002731.1_70	1.77	0.018
Immune Response
CCL7	chemokine	scl41159.3.1_10	2.02	0.014
CFD	complement factor D	NM_013459.1	1.78	0.042
CCL2	chemokine	scl020296.2_11	2.03	0.014
CLEC4D	C-type lectin domain family 4, member d	scl017474.5_5	1.54	0.034
C3	complement component 3	scl49743.39.1_15	1.74	0.042
C1S	complement component 1, s subcomponent	NM_144938.1	3.34	0.003
PRG4	PREDICTED: proteoglycan 4	scl000882.1_25	2.34	0.003
SPON2	spondin 2, extracellular matrix protein	NM_133903.2	1.98	0.019
CXCL14	chemokine	scl43911.4.1_38	1.52	0.031
Proteolysis
DPEP2	dipeptidase 2	scl34381.6_178	1.40	0.047
CTSK	cathepsin K	NM_007802.2	1.56	0.014
CFD	complement factor D	NM_013459.1	1.78	0.042
CPXM1	carboxypeptidase X 1	NM_019696.1	1.25	0.020
CTSC	cathepsin C	NM_009982.2	1.75	0.020
C1S	complement component 1, s subcomponent	NM_144938.1	3.34	0.003
HP	haptoglobin	NM_017370.1	3.13	0.002
Cell Adhesion
CNTN2	contactin 2	scl0021367.1_277	2.19	0.017
CPXM1	carboxypeptidase X 1	NM_019696.1	1.25	0.020
COL8A2	collagen, type VIII, alpha 2	scl24964.1.1958_52	1.98	0.018
CYR61	cysteine rich protein 61	NM_010516.1	1.66	0.039
TNC	tenascin C	scl0002731.1_70	1.77	0.018
SPP1	secreted phosphoprotein 1	NM_009263.1	2.51	0.006
SPON2	spondin 2, extracellular matrix protein	NM_133903.2	1.98	0.019
COMP	cartilage oligomeric matrix protein	scl33728.21.1_0	1.85	0.025
FN1	fibronectin 1	scl16639.44.189_5	1.57	0.013
Apoptosis
CIDEC	cell death-inducing DFFA-like effector c	NM_178373.2	1.90	0.046
COMP	cartilage oligomeric matrix protein	scl33728.21.1_0	1.85	0.025
Lipid Metabolic Process
SLC27A3	solute carrier family 27	scl21918.10.1_222	1.74	0.031
LPL	lipoprotein lipase	scl0016956.1_234	1.71	0.018
Innate Immune Response
LBP	lipopolysaccharide binding protein	scl20002.16.7_10	1.92	0.019
CFD	complement factor D	NM_013459.1	1.78	0.042
C3	complement component 3	scl49743.39.1_15	1.74	0.042
C1S	complement component 1, s subcomponent	NM_144938.1	3.34	0.003
Brown Fat Cell Differentiation
ADIPOQ	adiponectin, C1Q and collagen domain containing	scl49310.3_131	2.30	0.003
LRG1	leucine-rich alpha-2-glycoprotein 1	NM_029796.2	2.31	0.004
Skeletal System Development
RUNX1	runt related transcription factor 1	scl48188.1.1_190	2.07	0.011
COL1A1	procollagen, type I, alpha 1	scl012842.26_28	2.15	0.007
Blood Vessel Development
COL3A1	procollagen, type III, alpha 1	NM_009930.1	1.57	0.039
COL1A1	procollagen, type I, alpha 1	scl012842.26_28	2.15	0.007

Table 8.

List of genes downregulated in the pulmonary arteries of 2 mo old hypoxic male SERT+ mice compared with 2 mo old hypoxic male wild-type mice, arranged by biological process

Gene Symbol	Gene Name	Accession No.	Fold Change	False Discovery Rate
Transport
UCP1	uncoupling protein 1	NM_009463.2	3.84	0.000
KCNAB1	potassium voltage-gated channel, shaker-related 1 subfamily, beta member	NM_010597.2	1.68	0.025
KCNH2	potassium voltage-gated channel, subfamily H	NM_013569.1	1.74	0.014
TOMM22	translocase of outer mitochondrial membrane 22 homolog	scl47742.5_273	3.76	0.000
RAMP1	receptor (calcitonin) activity modifying protein 1	scl17654.5.1_10	1.59	0.046
RAB17	RAB17, member RAS oncogene family	NM_008998.2	1.92	0.006
Proteolysis
MIPEP	mitochondrial intermediate peptidase	NM_027436.1	1.50	0.046
DPEP1	dipeptidase 1	NM_007876.1	1.70	0.031
CORIN	corin	NM_016869.1	1.84	0.014
Cell Adhesion
MCAM	melanoma cell adhesion molecule	NM_023061.1	1.51	0.046
PKP4	plakophilin 4	scl0003206.1_31	1.70	0.025
Apoptosis
ACTC1	actin, alpha, cardiac	NM_009608.1	1.14	0.018
CIDEA	cell death-inducing DNA fragmentation factor, alpha subunit-like effector A	NM_007702.1	1.88	0.018
PAWR	PRKC, apoptosis, WT1, regulator	scl38461.7.1_1	1.67	0.031
Lipid Metabolic Process
TNXB	tenascin XB	NM_031176.1	1.91	0.013
CIDEA	cell death-inducing DNA fragmentation factor, alpha subunit-like effector A	NM_007702.1	1.88	0.018
Heart Development
PDLIM3	PDZ and LIM domain 3	NM_016798.2	1.59	0.028
EDN1	endothelin 1	NM_010104.2	1.55	0.046
TNNT2	troponin T2, cardiac	NM_011619.1	1.62	0.004
Brown Fat Cell Differentiation
UCP1	uncoupling protein 1	NM_009463.2	3.84	0.000
Skeletal System Development
GJA5	gap junction membrane channel protein alpha 5	NM_008121.2	1.55	0.042
Blood Vessel Development
GJA5	gap junction membrane channel protein alpha 5	NM_008121.2	1.55	0.042
GJA4	gap junction membrane channel protein alpha 4	NM_008120.2	1.80	0.014

Hierarchal cluster analysis of the differentially expressed genes between the four hypoxic groups revealed distinct gene expression patterns which were unique to female SERT+ mice. This may be critical to the exaggerated hypoxia-induced PAH phenotype observed in these mice.

Quantitative RT-PCR analysis.

For validation of the microarray study, we employed qRT-PCR. To perform this, we selected three differentially expressed genes for each of the four group comparisons (Supplementary Table). Our genes of interest were FOS, CEBPβ, CYP1B1, MYL3, HAMP2, LTF, PLN, NPPA, UCP1, and C1S. In concordance with our microarray data, expression of these genes was significantly altered in relevant groups (Fig. 5). Of particular interest, qRT-PCR analysis confirmed that FOS, CEBPβ and CYP1B1 were considerably up-regulated (4-, 20-, and 8-fold, respectively) in female SERT+ mice.

Fig. 5.

Validation of microarray data using qRT-PCR analysis. qRT-PCR analysis performed on 3 differentially expressed genes (according to microarray) for each comparison. A–C: normoxic female WT mice cf. normoxic female SERT+ mice (FOS, CEBPβ, CYP1B1); D–F: hypoxic female WT mice cf. hypoxic female SERT+ mice (MYL3, CEBPβ, HAMP2); G–I: normoxic male WT mice cf. normoxic male SERT+ mice (LTF, PLN, NPPA); J–L: hypoxic WT male cf. hypoxic SERT+ male (LTF, C1S, UCP1). n = 4 and performed in triplicate. Data are expressed as means ± SE and analyzed by unpaired t-test. *P < 0.05, **P < 0.01, ***P < 0.001 cf. relevant comparison group.

CCAAT/enhancer-binding protein beta, CYP1B1, and c-FOS protein expression in female SERT+ mice.

To build on interesting gene expression differences observed in female SERT+ mice, we investigated expression of CCAAT/enhancer-binding protein beta (C/EBPβ), CYP1B1, and FOS at protein level. In agreement with our qRT-PCR findings, protein expression of C/EBPβ, CYP1B1, and c-FOS were also upregulated in the pulmonary arteries of female SERT+ mice (Fig. 6).

Fig. 6.

Representative immunoblotting and densitometric analysis confirming increased protein expression of C/EBPβ (A, B), CYP1B1 (C, D), and c-FOS (E, F) in the pulmonary arteries of female SERT+ mice compared with female WT mice. n = 4 and performed in triplicate. Data are expressed as means ± SE and analyzed by unpaired t-test. *P < 0.05 cf. female WT mice.

Serotonin and 17β-estradiol stimulate C/EBPβ, CYP1B1, and c-FOS expression in human PASMCs.

To determine if serotonin and 17β-estradiol stimulate expression of C/EBPβ, CYP1B1, and c-FOS, we investigated expression of these in PASMCs following 24 h stimulation with serotonin and 17β-estradiol. Stimulation with serotonin or 17β-estradiol was sufficient to increase C/EBPβ, CYP1B1, and c-FOS expression in PASMCs (Fig. 7).

Fig. 7.

Representative immunoblotting and densitometric analysis confirming increased protein expression of C/EBPβ (A, B), CYP1B1 (C, D), and c-FOS (E, F) in human PASMCs following stimulation with 1 μmol/l serotonin (5-HT) or 1 nmol/l 17β-estradiol (E₂). n = 3 and performed in triplicate. Data are expressed as means ± SE and analyzed by 1-way ANOVA followed by Dunnett's post hoc test. *P < 0.05 cf. control PASMCs.

C/EBPβ, CYP1B1, and c-FOS expression in IPAH PASMCs.

To identify if these findings translate with relevance to human PAH, we investigated the expression of CEBPβ, CYP1B1, and FOS in PASMCs derived from IPAH patients. PASMCs from non-PAH donors were studied as controls. Interestingly, CEBPβ, CYP1B1, and FOS expression appeared significantly increased in mRNA extracted from IPAH PASMCs (Fig. 8). Similarly, Western blot analysis confirmed that protein expression of C/EBPβ, CYP1B1, and c-FOS is also increased in IPAH PASMCs compared with control PASMCs.

Fig. 8.

Quantitative RT-PCR analysis confirming the upregulation of CEBPβ (A), CYP1B1 (B), and FOS (C), and representative immunoblotting and densitometric analysis confirming increased protein expression of C/EBPβ (D, E), CYP1B1 (F, G), and c-FOS (H, I) in pulmonary artery smooth muscle cells (PASMCs) derived from IPAH patients compared with control PASMCs. n = 3 and performed in triplicate. Data are expressed as means ± SE and analyzed by unpaired t-test. *P < 0.05, **P < 0.01, ***P < 0.001 cf. control.

DISCUSSION

Despite increased mortality reported in men (15), the incidence of both IPAH and HPAH remains up to threefold more common in women. This is highlighted in recent epidemiological studies carried out in Scotland, France, and USA, where 60, 65, and 77% of the patients studied respectively were female (16, 28, 43). Established experimental models of PAH have failed to provide insight into this increased occurrence. Paradoxically, several experimental models of PAH exhibit male susceptibility compared with their female counterparts (12, 24, 29, 34). Here, we describe an experimental model of PAH that exhibits female susceptibility. Female SERT+ mice develop PAH and exaggerated hypoxia-induced PAH, whereas male SERT+ mice remain unaffected, when compared against their respective WT controls. We were interested in determining the genotypic differences associated with the development and progression of PAH in SERT+ mice. To investigate this, microarray analysis was performed in the pulmonary arteries of SERT+ mice at 2 mo of age, where no PAH phenotype is reported.

Through microarray analysis we have identified a large number of differentially expressed genes in the pulmonary arteries of SERT+ mice. In total, we identified 155 genes changed in female SERT+ mice, while 148 genes were changed in male SERT+ mice. Heat map analysis identified gene expression changes in females that were not apparent in males. When assigned to biological processes, we also identified that >40% of the differentially expressed genes in female SERT+ mice were directly involved in biological pathways. In total, 15 known biological pathways were dysregulated in female SERT+ mice and included oxidation-reduction, cell differentiation, regulation of transcription, apoptosis, muscle contraction, cellular calcium ion homeostasis, and glycolysis. This may be relevant to the development of PAH in SERT+ mice, as dysregulation of these pathways has been previously implicated in the pathogenesis of PAH (33, 35). Indeed, similar pathway changes have also been described in the lungs of VIP−/− mice (9) and BMPR-II mutant mice (40). In contrast to female SERT+ mice, only 25% of altered genes in male SERT+ mice were associated with biological function and as a consequence resulted in the dysregulation of pathways to a much lesser extent.

In chronic hypoxia, there were also a large number of differentially expressed genes in SERT+ mice compared with their respective WT controls. We observed a total of 316 genes altered in females, whereas less than half (154) of these were altered in males. In hypoxic female SERT+ mice, 53% of genes were associated with biological function. Similar to the normoxic female comparison, altered genes were related to apoptotic, inflammatory, transcription, and metabolic processes, all of which are well-described in PAH (13). In total, 26 biological pathways were identified as dysregulated. As expected, fewer genes were reported as changed in male SERT+ mice. These differences may help explain the exaggerated hypoxia-induced PAH phenotype in female SERT+ mice.

The female hormone 17β-estradiol is one risk factor in PAH. Decreased expression of the 17β-estradiol-metabolizing enzyme CYP1B1, resulting in altered estrogen metabolism, has been identified in female PAH patients harboring a BMPR-II mutation compared with unaffected female carriers (2). Multiple factors modulate the levels of estrogen-metabolizing enzymes in the liver and target tissues, and the biological effects of an estrogen will depend on the profile of metabolites formed and the biological activities of each of these metabolites (50). 17β-Estradiol is metabolized to both pro- and antiproliferative metabolites, and its effects will depend on its metabolism. 17β-Estradiol can be converted to estrone and subsequently metabolized to 16α-hydroxyestrone (16-OHE1) via CYP3A4. Or alternatively, 17β-estradiol is metabolized to 2-hydroxyestradiol (2-OHE) via the estrogen-metabolizing enzymes CYP1A1/2 and to a lesser extent via CYP1B1 (10, 46). 2-OHE can itself be metabolized to 2-methoxyestradiol (2-ME) via catechol O-methyltransferase (COMT). Both 2-OHE and 2-ME have antiproliferative effects on cells (44), whereas 16α-OHE1 stimulates proliferation by constitutively activating the estrogen receptor (39). Metabolism of 17β-estradiol will therefore be species, sex, and strain-dependent, and differential disruption in the balance of metabolites may therefore account for the differential effects of female hormones in different models of PAH. Consistent with this, our microarray findings show that CYP1B1 mRNA expression is increased in female SERT+ mice. In further support of this, immunoblotting confirmed that CYP1B1 protein expression is also increased in the pulmonary arteries of female SERT+ mice. Of further interest, both serotonin and 17β-estradiol stimulation increased CYP1B1 expression in PASMCs. Indeed, similar 17β-estradiol effects have been previously described in cancer cells (45). On this evidence, serotonin and 17β-estradiol may be accountable for increased CYP1B1 expression in female SERT+ mice.

C/EBPβ is a transcription factor encoded by the CEBPβ gene. C/EBPβ has been previously shown to regulate inflammation, cell differentiation, and cell proliferation (31). For example, C/EBPβ is essential in the pathogenesis of multiple proliferative disorders including skin, breast, and ovarian cancer (32, 38, 51). In line with this, C/EBPβ-deficient mice appear resistant to tumorigenesis (37). The role of C/EBPβ in the development of PAH is poorly defined. Increased C/EBPβ expression has been reported in the lungs of chronically hypoxic rats (42), where it appears to stimulate inducible nitric oxide synthase expression. Reduced CEBPβ expression has also been previously reported in the lungs of SERT knockout mice (3). Conversely, our microarray data show the upregulation CEBPβ in female SERT+ mice. Increased CEBPβ mRNA expression was confirmed by qRT-PCR analysis. We also identified that C/EBPβ protein expression was increased in the pulmonary arteries of female SERT+ mice. In support of this, we observed that serotonin and 17β-estradiol increased C/EBPβ expression in human PASMCs. These findings suggest that serotonin and 17β-estradiol may stimulate C/EBPβ expression in vivo, and this contributes to the pathogenesis of PAH in female SERT+ mice.

We observed increased FOS expression in the pulmonary arteries of female SERT+ mice. FOS is a proto-oncogene that exists as an immediate early gene transcription factor and is transactivated in response to various stimuli (14). For example, FOS expression is increased in the heart following exposure to hypoxia (4). In bovine PASMCs, serotonin is also a potent inducer of FOS expression via a MAPK-dependent pathway (36). In agreement with this, we observed that serotonin stimulation also increased c-FOS expression in human PASMCs. Of interest, expression was also increased in 17β-estradiol-stimulated cells. Similar effects have also been described in rat hepatoctyes (19). In vivo, c-FOS expression is increased in the pulmonary arteries of female SERT+ mice. Here, our evidence suggests serotonin and 17β-estradiol stimulate c-FOS expression, and this may be relevant to the pathogenesis of PAH in female SERT+ mice.

With relevance to human PAH we further examined CEBPβ, CYP1B1, and FOS expression in PASMCs derived from IPAH patients. We observed that expression of these three genes (CEBPβ, CYP1B1 and FOS) was increased in IPAH PASMCs. We observed at least fivefold increases in CEBPβ and FOS mRNA expression compared with control PASMCs. Immunoblotting confirmed that the upregulation of C/EBPβ and c-FOS was also apparent at protein level. Since these genes are involved in inflammation and proliferation, both of which are essential components in disease pathogenesis (47), our findings suggest their importance in human PAH. We also observed increased CYP1B1 mRNA and protein expression in IPAH PASMCs, suggesting the importance of CYP1B1-mediated estrogen metabolism in PAH. However, these findings are inconsistent with previous studies in Epstein-Barr virus immortalized B cells derived from female BMPR-II PAH patients (48), where decreased CYP1B1 mRNA expression was described. Most likely, this is attributable to the differences in cell type investigated. This study focuses on changes in PASMCs, which represent a more physiologically relevant cell type in PAH.

We have previously reported that SERT+ mice develop elevated RVSP in the absence of RVH (21). This phenomenon is particular to normoxic mice as we, like others, have shown that mice develop RVH following exposure to hypoxia (18, 22). We are not alone in observing this phenomenon as other studies have similarly demonstrated elevated RVSP in transgenic mice in the absence of RVH. For example, mice that express BMPR-II^R899X in smooth muscle or molecular loss of BMPR-II signaling in smooth muscle demonstrate elevated RVSP with no change in RVH (41, 49). The observation that this only occurs in normoxic mice suggests that hypoxia induces an effect on RVH that may indeed be independent of RVSP.

The distal arteries are typically those most susceptible to pulmonary vascular remodeling in PAH; however, microarray analysis was performed in the proximal pulmonary arteries of mice as these were the smallest that could be practically dissected out from whole lung. Therefore, these gene changes may not be entirely representative of gene expression changes in smaller resistance arteries. For example, our microarray results show that hypoxic female SERT+ mice exhibit increased PPAR-γ expression relative to hypoxic female WT mice. However, previous observations confirm that PPAR-γ expression is reduced in the distal pulmonary arteries of PAH patients (1), and its targeted deletion in pulmonary artery smooth muscle or endothelial cells is sufficient to cause PAH in mice (7, 11). This contrast in findings may well result from the effect of hypoxia per se or an indirect compensatory change in response to PAH in SERT+ mice.

In conclusion, through microarray analysis we have identified a large number of differentially expressed genes in the pulmonary arteries of SERT+ mice. These findings offer further insight into the gender differences observed in this serotonin-dependent model of PAH. At least three of these genes (CEBPβ, CYP1B1, and FOS) are also upregulated at protein level in these mice. With relevance to human PAH, we identified that mRNA and protein expression of CEBPβ, CYP1B1, and FOS was also increased in PASMCs derived from IPAH patients. This study has described genotypic differences in a serotonin-dependent model of PAH and these findings at least in part, may be relevant to the pathogenesis observed in human PAH.

GRANTS

K. White is supported by a Capacity Building Award in Integrative Mammalian Biology funded by the Biotechnology and Biological Sciences Research Council (BBSRC), British Pharmacological Society, Knowledge Transfer Network, Medical Research Council, and Scottish Funding Council. Z. Maqbool is supported by the BBSRC. Y. Dempsie is supported by the Medical Research Council.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).