AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
Page 1 of 54
1
Title
2
In utero vitamin D deficiency increases airway smooth muscle mass and impairs lung
3
function
4
Authors
5
1
6
Hart, 2Graeme R Zosky
7
Author contributions
8
Conception and design of research: REF, PHH, SG, GRZ; Acquisition of data: REF, ACJ;
9
Data analysis: REF, AB, AG. Interpretation of results: REF, AB; Drafting the manuscript for
10
important intellectual content: REF; Revision of manuscript: REF, AB, SG, PHH, GZ;
11
Approval of final version of manuscript: REF, AB, ACJ, AG, SG, PHH, GRZ.
12
Affiliations
13
1
14
Australia;
15
2
16
Australia.
17
Corresponding author
18
Rachel E Foong
19
Tel no.: +618 9489 7819
20
Fax no.: +618 9489 7700
21
Email:
[email protected] Rachel E Foong, 1Anthony Bosco, 1 Anya C Jones, 1Alex Gout, 1Shelley Gorman, 1Prue H
Telethon Kids Institute, the University of Western Australia, Perth, Western Australia,
School of Medicine, Faculty of Health Science, University of Tasmania, Hobart, Tasmania,
1
Copyright © 2015 by the American Thoracic Society
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
1
Running title
2
In utero vitamin D deficiency causes lung deficits
3
Funding Sources
4
This study was funded by the National Health and Medical Research Council of Australia
5
(1042235). RF is a recipient of an Australian Postgraduate Award, Stan and Jean Perron Top-
6
Up Award and Asthma Foundation Western Australia Top-Up Award. AB is supported by a
7
BrightSpark Foundation McCusker Fellowship. ACJ is a recipient of an Australian
8
Postgraduate Award and a Top-Up Award from the University of Western Australia. SG is a
9
recipient of the BrightSpark Foundation Fellowship.
10
11
12
13
14
15
16
17
18
19
20
2
Copyright © 2015 by the American Thoracic Society
Page 2 of 54
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
Page 3 of 54
1
Abstract
2
We have previously demonstrated increased airway smooth muscle (ASM) mass and airway
3
hyperresponsiveness (AHR) in whole-life vitamin D-deficient female mice. In this study we
4
aimed to uncover the molecular mechanisms contributing to altered lung structure and
5
function. RNA was extracted from lung tissue of whole-life vitamin D-deficient and -replete
6
female mice, and gene expression patterns were profiled by RNA-Seq. The data showed that
7
genes
8
morphogenesis, Wnt signalling, and inflammation were differentially expressed in vitamin D-
9
deficient mice. Network analysis suggested that differentially expressed genes were
10
connected by the hubs MMP9, NFKBIA, EGFR and EP300. Given our findings that
11
developmental pathways may be altered, we investigated if the timing of vitamin D exposure
12
(in utero versus postnatal) had an impact on lung health outcomes. Gene expression was
13
measured in in utero or postnatal vitamin D-deficient mice, as well as whole-life vitamin D-
14
deficient and -replete mice at 8-weeks of age. Baseline lung function, AHR and airway
15
inflammation were measured and lungs fixed for lung structure assessment using
16
stereological methods, and quantification of ASM mass. In utero vitamin D deficiency was
17
sufficient to increase ASM mass, baseline airway resistance, and alter lung structure. There
18
were increased neutrophils but decreased lymphocytes in bronchoalveolar lavage. Expression
19
of inflammatory molecules S100A9 and S100A8 was mainly increased in postnatal vitamin
20
D-deficient mice. These observations suggest that in utero vitamin D deficiency can alter
21
lung structure and function, and increase inflammation, contributing to symptoms in chronic
22
diseases such as asthma.
involved
in
embryonic
organ
development,
pattern
23
24
Key words
3
Copyright © 2015 by the American Thoracic Society
formation,
branching
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
1
Airway remodeling, inflammation, RNA-Seq, mouse model, vitamin D
2
Introduction
3
Vitamin D deficiency is associated with chronic lung diseases such as asthma and chronic
4
obstructive pulmonary disease (COPD) (1, 2). Asthma and COPD are both characterized by
5
airway hyperresponsiveness (AHR) and airway remodeling (3), including an increase in
6
airway smooth muscle (ASM) mass (4). We have recently demonstrated increased ASM mass
7
in mice exposed to whole-life vitamin D deficiency (5). The increases in ASM were
8
associated with gene expression levels of transforming growth factor (TGF)-β signalling
9
molecules in vitamin D-deficient embryos at embryonic day 17.5, indicating that vitamin D
10
deficiency may impinge on developmental pathways in early life. These observations support
11
previous studies showing that low serum vitamin D levels are associated with increased ASM
12
mass in children with severe asthma (6). Similarly, in vitro studies show that 1,25-
13
hydroxyvitamin D (1,25(OH)2D3), the active metabolite of vitamin D, inhibits the
14
proliferation of ASM cells (7), providing evidence that vitamin D has the capacity to
15
modulate airway remodeling.
16
Experimental animal studies suggest that vitamin D affects lung development. The
17
maturation of fetal rat type II alveolar epithelial cells (AECs), which are important for the
18
synthesis of surfactant and represent a key event in fetal lung development, is induced by
19
1,25(OH)2D3 (8). Alveolar epithelial-mesenchymal interactions that are important for alveolar
20
development are also influenced by 1,25(OH)2D3 (9). Furthermore, genetic studies show a
21
high number of vitamin D regulated genes are represented in the developing human and
22
murine lung transcriptome (10). One-third of these vitamin D regulated genes were
23
differentially expressed in asthmatic children compared to non-asthmatic sibling controls
4
Copyright © 2015 by the American Thoracic Society
Page 4 of 54
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
Page 5 of 54
1
suggesting that vitamin D deficiency during lung development may lead to asthma later in
2
life.
3
A recent study from our group investigating the effects of maternal vitamin D status on
4
postnatal lung function outcomes found impaired lung function and increased asthma risk in
5
children exposed to maternal vitamin D deficiency (11). This was consistent with our earlier
6
work showing a causal association between early life vitamin D deficiency and lung structure
7
and function in a mouse model (12). Based on our recent observation that whole-life vitamin
8
D deficiency increases ASM mass, we sought to investigate pathways that might be altered
9
by vitamin D deficiency in the adult lung using RNA sequencing (RNA-Seq). We focused on
10
female mice because vitamin D deficiency induced physiological changes are only present in
11
female mice (5). Moreover, the maternal effects of vitamin D deficiency on postnatal lung
12
function are strongest in girls (11). Given the existing evidence supporting a link between
13
early life vitamin D deficiency and lung health, we also investigated the relative impact of in
14
utero and/or postnatal vitamin D deficiency on ASM, lung mechanics, AHR, inflammation
15
and gene expression patterns to determine when lung structure and function are the most
16
sensitive to the effects of vitamin D deficiency.
17
Materials and Methods
18
Mouse model
19
A mouse model of vitamin D deficiency was established by dietary manipulation as described
20
previously (5). To compare the effects of in utero and postnatal vitamin D deficiency, pups
21
were cross-fostered to produce in utero vitamin D-deficient (VitD-|+) or postnatally vitamin
22
D-deficient (VitD+|-), as well as whole-life vitamin D-deficient (VitD-|-) or whole-life
23
vitamin D-replete (VitD+|+) mice. All studies were carried out on female mice when they
24
were 8-weeks of age. All procedures were approved by the Telethon Kids Institute Animal 5
Copyright © 2015 by the American Thoracic Society
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
1
Ethics Committee and conformed to National Health and Medical Research Council of
2
Australia guidelines. Further details are provided in the online supplement.
3
RNA-Seq protocol and data analysis
4
Total RNA was extracted from lung tissue of whole-life vitamin D-deficient and -replete
5
mice using TRIzol (Life Technologies, Victoria, Australia) followed by RNeasy (Qiagen,
6
Victoria, Australia). Library preparation (Illumina TruSeq RNA Sample Prep Kit v2) and
7
sequencing (50bp single-end using an Illumina HiSeq2000) was performed by the Australian
8
Genome Research Facility Ltd. The raw sequencing results are available at the NCBI Short
9
Read Archive under accession SRP046087. Further details regarding the RNA-Seq data
10
analysis is provided in the online supplement.
11
Reverse transcription quantitative PCR (RT-qPCR)
12
A subset of differentially expressed genes identified in the RNA-Seq experiment was
13
validated using RT-qPCR. Reverse transcription was carried out using the QuantiTect
14
Reverse Transcription Kit (Qiagen). Real-time PCR primer assay sequences were obtained
15
from Primerbank (13) and the primer list included in the online supplement. QuantiTect
16
SYBR Green was used for qPCR on the ABI7900HT instrument. Relative standard curves
17
were prepared from serially diluted PCR products (14).
18
Lung function and methacholine challenge
19
Baseline lung function was assessed using the low-frequency forced oscillation technique
20
(15). After lung function measurements, mice were transferred to a flexiVent system
21
(SCIREQ, Montreal, Canada) for assessment of responsiveness to methacholine (β-
22
methacholine chloride, Sigma-Aldrich, Missouri, USA) as described previously (5). Further
23
details are provided in the online supplement.
6
Copyright © 2015 by the American Thoracic Society
Page 6 of 54
AJRCMB Articles in Press. Published on 13-April-2015 as 10.1165/rcmb.2014-0356OC
Page 7 of 54
1
Differential cell counts
2
Bronchoalveolar lavage (BAL) was collected from mice by washing 500 µL of sterile saline
3
into the lungs 3 times. Total cell collected in the BAL were counted using a haemocytometer.
4
The remaining cells were cytospun and stained with Leischman’s stain for differential cell
5
counts.
6
Histology and immunohistochemistry
7
Lung structure was assessed in independent groups of mice following ATS/ERS guidelines
8
(16). The right lung was used for lung stereology as described previously (12). Transverse
9
sections from the left lung were stained with Masson’s Trichrome for measurement of ASM
10
mass (5). Immunohistochemical staining was performed using primary antibodies for
11
S100A9, S100A8, phosphorylated IκBα, phosphorylated EGFR and MMP9. Further details
12
are provided in the online supplement.
13
Statistical analyses
14
Between group comparisons were made using t tests or 2-way ANOVAs with Holm-Sidak
15
posthoc tests. Pearson’s correlation tests were performed to assess correlations between AHR
16
and inflammatory parameters. Statistical analyses were performed using SigmaPlot software
17
(SysStat Software, Illinois, USA) and Prism (GraphPad Software, California, USA). Data are
18
shown as mean (SD). P values of