University of Southern California Dissertations and Theses

Global analysis of the molecular activities defining maturation-stage amelogenesis

(USC Thesis Other)

Global analysis of the molecular activities defining maturation-stage amelogenesis

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
1

GLOBAL ANALYSIS OF THE MOLECULAR ACTIVITIES
DEFINING MATURATION-STAGE AMELOGENESIS

By

Kaifeng Yin

A Dissertation Presented to the
FACULTY OF THE USC HERMAN OSTROW SCHOOL OF DENTISTRY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CRANIOFACIAL BOLOGY)

August 2015

Copyright 2015 Kaifeng Yin

2
Dedication

To my wife and my parents
Jing
Chongye & Jianmei
Wei & Xiuzhen

To my mentor
Dr. Michael L. Paine

3
Acknowledgements

I would like to address my sincere gratitude to my mentor—Dr. Michael L. Paine, for his
support on the whole project and his guidance through the graduate program.

I would like to extend my gratitude to other members in my committee: Dr. Joseph G.
Hacia, Dr. Malcolm L. Snead, Dr. Glenn T. Sameshima and Dr. Janet Moradian-Oldak.

I would also like to thank all members in Paine lab and Snead lab.

Finally, a special thanks to my wife—Dr. Jing Guo.

4
Table of Contents

Dedication ................................................................................................................................................... 2
Acknowledgements ..................................................................................................................................... 3
Lists of Symbols and Abbreviations ........................................................................................................... 6
Lists of Tables ............................................................................................................................................ 9
Lists of Figures ......................................................................................................................................... 11
Abstract..................................................................................................................................................... 13
Chapter One: Introduction ........................................................................................................................ 16
1.1 Enamel maturation .......................................................................................................................... 16
1.1.1 Amelogenesis and enamel maturation ...................................................................................... 16
1.1.2 Ion transport & pH regulation .................................................................................................. 17
1.1.3 Matrix turnover ........................................................................................................................ 17
1.1.4 Apoptosis ................................................................................................................................. 17
1.1.5 Amelogenesis imperfecta (AI) ................................................................................................. 18
1.2 miRNA regulation ........................................................................................................................... 19
1.2.1 Mechanisms of miRNA regulation ........................................................................................... 19
1.2.2 Global profiling techniques of miRNA regulation ................................................................... 20
1.2.3 miRNA in dental development ................................................................................................. 22
1.3 Slc26a gene family .......................................................................................................................... 25
Chapter Two: Genome-wide analysis of miRNA and mRNA transcriptomes during amelogenesis ......... 27
2.1 Abstract ........................................................................................................................................... 27
2.2 Background ..................................................................................................................................... 29
2.3 Methods .......................................................................................................................................... 32
2.4 Results ............................................................................................................................................ 40
2.5 Discussion ....................................................................................................................................... 67
2.5 Conclusion ...................................................................................................................................... 72
Chapter Three: SLC26A Gene Family Participate in pH Regulation During Enamel Maturation............. 73
3.1 Abstract ........................................................................................................................................... 73
3.2 Introduction ..................................................................................................................................... 74
3.3 Experimental procedures ................................................................................................................. 77
3.4 Results ............................................................................................................................................ 86
3.5 Discussion ..................................................................................................................................... 108
3.6 Conclusion .................................................................................................................................... 115

5
Chapter Four: Conclusions ..................................................................................................................... 116
Additional Files ...................................................................................................................................... 127
References .............................................................................................................................................. 131

6
Lists of Symbols and Abbreviations

ABD AGO-binding domain
AEC 3-amino-9-ethycarbazole
AGO Argonaute
AI Amelogenesis imperfecta
At% Atomic percentage
BEM Backscattered electron microscopy
B-H Benjamini Hochberg
BSA Bovine serum albumin
CAs Carbonic anhydrases
CFTR Cystic fibrosis conductance transmembrane regulator
Co-IP Co-immunoprecipitation
DAB 3`-diaminobenzidine
DCP2 Decapping enzyme 2
DEGs Differentially expressed genes
EDS Energy dispersive X-ray spectroscopy
ECM Extracellular matrix
EMPs Enamel matrix proteins
FBS Fetal bovine serum
FC Fold changes
FDR False discovery rate
PGS Partek® Genomics Suite
GO Gene ontology

7
GW182 Glycine-tryptophan protein of 182 kDa
IACUC Institutional Animal Care and Use Committee
IPA Ingenuity Pathway Analysis
KEGG Kyoto Encyclopedia of genes and genomics
LNA-DIG Locked nucleic acid-digoxygenin
LUX Light Upon eXtension
miRISC miRNA induced silencing complex
NCBI National Center for Biotechnology Information
OHCs Cochlear outer hair cells
OMCD Outer medullary collecting duct
PABP Poly(A)-binding protein
PAM2 PABP-interacting motif 2
PBS Phosphate buffered saline
PCR Polymerase chain reaction
PFA Paraformaldehyde
qPCR quantitative polymerase chain reaction
RNU6-2 U6 Small nuclear 2
SD Silencing domain
SEM Scanning electron microscopy
SCGC Southern California Genotyping Consortium
SLC4s Solute carrier family 4
SLC9s Solute carrier family 9
SLC26s Solute carrier family 26

8
T
m
Melting temperature
XRN1 Cytoplasmic exonuclease exoribonuclease enzyme 1
3`-UTR Three prime untranslated region

9
Lists of Tables

Table 1. Genes enriched in the GO category “carboxylic acid transmembrane transporter
activity” and their predicted miRNA regulators All fold changes represent the ratio of
maturation to secretory phase expression values. ......................................................................... 50
Table 2. Genes enriched in the GO category “ATPase activity, coupled to transmembrane
movement of ions, phosphorylative mechanism” and their predicted miRNA regulators.
All fold changes represent the ratio of maturation to secretory phase expression values. ......... 511
Table 3. Genes enriched in the GO category of “endosome membrane” and their predicted
miRNA regulators. All fold changes represent the ratio of maturation to secretory phase
expression values. ......................................................................................................................... 52
Table 4. Genes enriched in the GO category of “lysosome” and their predicted miRNA
regulators. All fold changes represent the ratio of maturation to secretory phase expression
values. ........................................................................................................................................... 53
Table 5. Genes enriched in the GO category of “calcium ion binding” and their predicted
miRNA regulators. All fold changes represent the ratio of maturation to secretory phase
expression values. ......................................................................................................................... 54
Table 6. Genes enriched in the GO category of “cytokine activity” and their predicted
miRNA regulators. All fold changes represent the ratio of maturation to secretory phase
expression values. ......................................................................................................................... 55
Table 7. Genes enriched in the GO category of “calcium ion transmembrane transporter
activity” and their predicted miRNA regulators. All fold changes represent the ratio of
maturation to secretory phase expression values. ......................................................................... 56

10
Table 8. Average fold changes of Slc26s during maturation-stage relative to secretory-
stage based on genome-wide mRNA transcriptome analysis. (N/A expression not detected;
α=0.05) .......................................................................................................................................... 87
Table 9. Rat and mouse specific primers for qPCR and cDNA analyses. ................................. 117
Table 10. qPCR analysis of gene expression changes in Slc26a1 null animals. ........................ 119
Table 11. qPCR analysis of gene expression changes in Slc26a7 null animals. ........................ 122
Table 12. Antibodies used for western blot, immunoperoxidase immunostaining,
immunofluorescence and co-immunoprecipitation analyses. ..................................................... 125
Table 13. Enamel and dentin Vickers microhardness of mutant animals. ................................. 126

11
Lists of Figures

Figure 1. Hierarchical clustering analysis of the expression levels of miRNAs with highest
variability (Coefficient of Variance (CV) > 0.15) across all samples.. ........................................ 42
Figure 2. Fold-changes of 39 up-regulated miRNAs (FDR<0.05) at maturation stage
relative to secretory stage based on miRNA qPCR array analysis. .............................................. 43
Figure 3. Fold-changes of 20 down-regulated miRNAs (FDR<0.05) at maturation stage
relative to secretory stage based on miRNA qPCR array analysis. .............................................. 44
Figure 4. Flow chart depicting the strategy used to select miRNA target genes for pathway
analyses. ........................................................................................................................................ 46
Figure 5. Highest scoring miRNA-mRNA interaction network, “Cellular Movement,
Neurological Diseases, Organismal Injury and Abnormalities.”. ................................................. 58
Figure 6. Expression patterns of seven miRNAs in secretory- and maturation-stage enamel
organs as shown by in situ hybridization.. .................................................................................... 61
Figure 7. (A) The expression levels of miR-153 in LS8 cells before and after transfection
with miR-153 mimics (blue) and inhibitors (red). (B) Relative luciferase reporter activities
in luciferase reporter assay validating the interaction of miR-153 with 3`-UTR of Lamp1.
Luciferase activities are presented in ratios of Firefly (F) to Renilla (R) luciferase reporter
activities. For the data presented, the amount of luciferase reporter vector used was 700ng,
and the concentrations of miR-153 mimics and inhibitors in final transfection complex
were 20pM and 0.2nM, respectively............................................................................................. 65
Figure 8. (A) The expression levels of miR-31 in LS8 cells before and after transfection
with miR-31 mimics (blue) and inhibitors (red). (B) Relative Luciferase reporter activities
in luciferase reporter assay validating the interaction of miR-31 with 3`-UTR of Tfrc. For

12
the data presented, the amount of luciferase reporter vector used was fixed at 700ng, and
the concentrations of miR-31 mimics and inhibitors in final transfection complex were
20pM and 0.2nM respectively. ..................................................................................................... 66
Figure 9. Reference lines for animal dissection.. ......................................................................... 78
Figure 10. Real-time PCR analysis of Slc26a1, Slc26a6 and Slc26a7 expression during
amelogenesis.. ............................................................................................................................... 88
Figure 11. Wester blot analysis of Slc26a1, Slc26a6 and Slc26a7 .............................................. 89
Figure 12. Immunoperoxidase immunostaining of Slc26a1, Slc26a6 and Slc26a7 in
secretory- and maturation-stage enamel organ.. .......................................................................... 92
Figure 13. Co-localization analysis of Slc26a1, Slc26a6, Slc26a7 with Ae2, Lamp1 and
Cftr. ............................................................................................................................................... 96
Figure 14. Co-immunoprecipitation (Co-IP) assay of Cftr with Slc26a1, Slc26a6 and
Slc26a7.. ........................................................................................................................................ 97
Figure 15. Micro-CT analysis of Slc26a1
-/-
and Slc26a7
-/-
mandibles.. ..................................... 100
Figure 16. SEM images of mature enamel in Slc26a1
-/-
and Slc26a7
-/-
animals.. ...................... 101
Figure 17. EDS analysis of mature enamel Slc26a1
-/-
and Slc26a7
-/-
animals. .......................... 102
Figure 18. Up-regulated genes in Slc26a1-/- and Slc26a7-/- animals compared with wild
types. ........................................................................................................................................... 106
Figure 19. Downregulated genes in Slc26a1-/- and Slc26a7-/- animals compared with
wild types.. .................................................................................................................................. 107
Figure 20. Schematic diagram depicting the distribution of major pH regulators in
maturation-stage ameloblasts.. ................................................................................................... 113

13
Abstract

Amelogenesis is the developmental process of dental enamel formation. Amelogenesis involves
two major functional stages—secretory and maturation. The transition of ameloblasts from
secretory to maturation stage, which is characterized by both morphological and functional
changes, results in the formation of mature enamel with ordered crystallite structures. Although
researcher today have a very clear idea of the molecular activities that define secretory-stage
amelogenesis, the molecular events that define enamel maturation remain understudied.

In chapter two, we set out to investigate the potential role of miRNA regulation in
maturation-stage tooth development. We conducted genome-wide miRNA and mRNA
transcript expression profiling analyses of secretory-stage and maturation-stage enamel
organs obtained from rat incisors. We identified a group of stage-specific miRNAs and
identified candidate gene targets based on bioinformatic prediction. Two maturation-stage-
related genes, Lamp1 and Tfrc, were verified by luciferase reporter assay to be the target
genes of miRNA regulators. The results indicated a dynamic expression pattern of
miRNAs during the transition from secretory-stage to maturation-stage enamel
mineralization, and suggest that miRNAs can influence key processes of enamel
maturation.

Based on the bioinformatic analyses in chapter two, Slc26 gene family were identified to be
involved in the pH regulation process during enamel maturation. Among the 11 members in
Slc26 gene family, Slc26a1, Slc26a6 and Slc26a7 are the only members whose transcripts are
significantly up-regulated during maturation-stage enamel formation when compared to

14
secretory-stage. In chapter three, we investigated the function roles of Slc26a1, Slc26a6 and
Slc26a7 during enamel maturation. We conducted quantitative real-time PCR and Western blot
analyses to show that Slc26a1, Slc26a6 and Slc26a7 are all significantly up-regulated at
maturation-stage (relative to secretory-stage) at both miRNA and protein levels. The subsequent
immunolocalization assays showed that in the maturation-stage ameloblasts the gene products of
Slc26a1, Slc26a6 and Slc26a7 localize to the apical region of cytoplasmic membrane, similar
to the localization of pattern of Cftr in maturation-stage ameloblasts. In addition, the
distribution of Slc26a7 was also seen within the cytoplasmic/subapical region of ameloblast,
presumably on the lysosomal membrane. From the protein complex pulled down using an
antibody to Cftr; Slc26a1, Slc26a6 and Slc26a7 were each separately detected by the
subsequent immunoblotting, suggesting the direct interaction of each of these three proteins
Slc26 proteins with Cftr. Compared with wild-type (WT) animals, Slc26a1-/- or Slc26a7-/-
animals did not show any significant abnormalities in the mature enamel phenotype (density
and structure). However, many gene transcripts examined by real-time PCR – such as Car2
(carbonic anhydrase 2), Cftr, Slc4a4/NBCe1, Slc4a9/Ae4, Slc26a9 and Alpl (alkaline
phosphatase) – showed significant up-regulation in the enamel organ cells of Slc26a1-/- and
Slc26a7-/- animals when compared to age and sex-matched wild-type controls. Collectively,
these data indicated that Slc26a1, Slc26a6 and Slc26a7 are actively involved in ion transport
related to pH regulation processes during enamel maturation and their functional roles may be
achieved, at-least in part, by forming complex interactions with Cftr.

15
Chapter One: Introduction

1.1 Enamel maturation

1.1.1 Amelogenesis and enamel maturation

Amelogenesis is the developmental process of dental enamel formation. Enamel development
involves two major functional stages-secretory and maturation for the ameloblast cells [1]. In the
secretory stage, ameloblast are highly polarized cells that synthesize and secrete a number of
structural Enamel Matrix Proteins (EMPs) [2-4]. The enamel tissue develops during the secretory
stage in a protein-rich environment maintained at near-neutral pH conditions. In the extracellular
space, thin Hap crystals elongate under the influence of EMPs and are also influenced by the
movements of the ameloblasts away from the dentine. Secretory-stage ameloblasts present a tight
barrier for diffusion of mineral ions, as suggested by differential concentrations of various
elements between tissue and enamel fluid [5, 6]. During the maturation stage, ameloblasts
undergo cyclical and rapid changes in morphology from ruffle-ended to smooth ended cell
phenotypes. Smooth-ended ameloblast (SA) with no apical tight junctions are presumed to be
impermeable to ion movement across the basal pole [3, 6], whereas ruffle-ended ameloblasts
(RA) are tightly bound by junctional complexes at their apices and show considerable
endocytotic activities. This dynamic permeability pattern allows for bi-directional diffusion of
small molecules into and, removal from the enamel area [3, 7]. As ameloblasts transition from
secretory stage to maturation stage of amelogenesis, their gene expression profiles dramatically
change [8]. Ameloblast function during enamel maturation can be classified into functions that

16
encompass ion transport [3], acid-base balance [9], EMP debris removal [10, 11] and
apoptosis[8]. By now, a majority of the detailed mechanisms of ameloblast-directed enamel
maturation are yet to be elucidated.

1.1.2 Ion transport & pH regulation

During the maturation-stage, the extracellular pH values vary considerable, ranging from neutral
to acidic conditions, with a return to physiologic pH levels at the end of maturation-stage [4, 12].
Changes in extracellular pH values require sophisticated regulatory mechanisms from
ameloblast, so as to maintain the acid-base balance in the microenvironment suitable for crystal
nucleation and growth [13]. In addition, extracellular EMPs at maturation-stage are internalized
by ameloblasts, and degraded through endosome-lysosome pathways [10, 14]. Elaborate pH
regulation machinery is also required in the endocytotic process to create an acidic luminal
environment for hydrolytic enzyme activation [15]. The process of pH control mediated by
enamel organ cells are facilitated by ion flows across biological membranes. Although there is
still uncertainty about the working schemes of pH regulation during amelogenesis [3, 13, 16-18],
the involvement of multiple ion channels or exchangers such as, carbonic anhydrases (CAs),
cystic fibrosis conductance transmembrane regulator (CFTR), Chloride Channel (CLCNs),
Solute Carrier family 4 (SLC4s) and Solute Carrier family 9 (SLC9s) in ameloblast-mediated pH
homeostasis has been widely accepted [7, 10, 16, 19-38].

1.1.3 Matrix turnover

During maturation-stage tooth development, most matrix proteins previously secreted into the
extracellular space are removed and replaced by hydroxyapatite-based minerals which contribute

17
to 96% of mature mammalian enamel. In order for the enamel organ to take up organic enamel
matrix, ameloblast-derived proteinases, such as MMP20 (expressed during secretory stage) and
KLK4 (expressed during maturation stage), must function first to degrade the matrix proteins
into smaller fragments [39, 40]. The subsequent removal of EMP fragments can be achieved by
passive diffusion across the dentin enamel junction (DEJ) [41, 42], nonspecific internalization [3,
12, 43, 44] or receptor-mediated endocytosis through the cytoplasmic surface of the apical pole
of ameloblasts [10, 45-48]. Recent studies focused on the endocytotic capabilities of both
smooth-ended and ruffled-ended ameloblasts during maturation stage, which involves the
trafficking of coated vesicles, and suggested the key role of clathrin, Adaptor proteins,
Lysosomal-associated membrane proteins, ATPase subunits, Chloride channels, Cathepsins and
Transferrin receptors [10, 14, 49].

1.1.4 Apoptosis

Programmed cell death—apoptosis is an important mechanism in the mammalian prenatal and
postnatal development. From the beginning of tooth formation and the completion of root
development, a balance between cell division and cell death is well maintained to control an
appropriate number of functional cells. Apoptotic pathways in epithelial and mesenchymal cells
have been described in animal models [50-55]. Several apoptosis molecules including Caspases,
Bcl-2 family, p21, p53, p63, p73, FasL, NF-kappa B and TGFβ signaling have been identified to
be essential during bud formation, tooth morphogenesis, reduction of tooth lamina and silencing
of the enamel knot signaling center [56]. However, little is known about the apoptosis of
ameloblast during enamel formation. Previous study showed that 50% of the ameloblasts
between late secretory and late maturation stages undergo cell death, with 25% occurring

18
immediately at the start of maturation and the remaining 25% throughout the subsequent regions
of the maturation zone [3]. In another study, apoptosis was detected in differentiating pre-
ameloblasts, in a few late secretory ameloblasts, in transitional ameloblast and adjacent stratum
intermedium cells and in maturation-stage ameloblasts [50]. Based on the genome-wide miRNA
and mRNA transcriptome analyses conducted by our group [8, 57], several candidate miRNAs
and predicted target genes are suggested to be the potential regulators of apoptotic signaling
pathways during secretory and maturation stage enamel formation and their functional roles
warrant further exploration.

1.1.5 Amelogenesis imperfecta (AI)

Amelogenesis imperfecta (AI) is a group of hereditary conditions that affect the structures and
appearances of dental enamel often in conjunction with changes in other intra-oral and/or extra-
oral tissues. AI shows a wide range of phenotype variability, which can be classified into
hypoplastic, hypocalcified, hypomaturation and hypomaturation-hypoplasia with taurodontism,
depending on the clinical characteristics [58]. To date, only a limited group of genes have been
identified to be directly associated with AI in human, such as AMEL[59, 60], ENAM [59, 61],
MMP20 [62], KLK4 [63], FAM83H [64, 65], WDR72 [66-69], STIM1and SLC24A4 [70].
Nevertheless, any malfunction of the intrinsic components in the genetic networks that regulate
amelogenesis could result in AI-like phenotypes, and thus should be considered as the potential
causes of AI. In that sense, large numbers of genes including non-coding RNAs that are
identified by genome-wide transcriptome analysis during amelogenesis warrant further
investigation to establish their links with enamel health [8, 57].

19
1.2 miRNA regulation

1.2.1 Mechanisms of miRNA regulation

MiRNAs are a group of small non-coding RNAs whose mature products are 21~22 nucleotides
in length [71, 72]. Through either canonical or non-canonical pathway, miRNA genes are
transcribed from RNA polymerase II promoters into primary miRNAs (pri-miRNAs) with a
typical hairpin structure [73]. In the nucleus, pri-miRNAs are processed by Drosha (canonical), a
member of RNase III family, or mRNA splicing machinery (non-canonical) to generate
precursor miRNAs (pre-miRNAs). The ~70 nucleotides-long pre-miRNAs are then exported via
exportin 5 into cytoplasm where they are further processed by a second RNase III enzyme—
Dicer. Mature miRNAs regulate the expression of target genes by directly binding to the 3` UTR
of their target mRNAs in form of miRNA induced silencing complex (miRISC), which results in
either mRNAs degradation or translational repression. A majority (>80%) of miRNAs in
mammals affect their gene targets at the mRNA level by decreasing the stability of target
mRNAs[74].

MiRISC consists of miRNA and two proteins—Argonaute (AGO) and glycine-tryptophan
protein of 182 kDa (GW182), and AGO proteins are the key catalytic enzymes within the
complex [75]. The N-terminal GW-repeat-containing region within the AGO-binding domain
(ABD) of GW182 confers it the ability to bind to specific Argonaute proteins. The base-paring of
miRNA (seed sequence) to complementary or partially complementary binding site that is
predominantly located in the 3` UTR of the target mRNA leads the AGO-GW182 complex to
recognize their mRNA target. GW182 interacts with poly(A)-binding protein (PABP), which

20
binds to the poly(A) tail of target mRNA, through the highly conserved PABP-interacting motif
2(PAM2) located in the Mid region of its bipartite silencing domain (SD). GW182 also interacts
with poly(A) nucleases 2 (PAN2)-poly(A) nucleases 3 (PAN3) and CCR4-NOT deadenylase
complexes. The binding of GW182 to PAN3 and NOT1, which are subunits of the PAN2-PAN3
and CCR4-NOT deadenylase complexes, requires the motifs of the SD that are predicted to be
unstructured. The assembly of miRISC on mRNA target causes translational repression. The
repressed mRNA is then adenylated. The translational repression of mRNA target has been
shown to occur prior to the complete adenylation [76, 77]. Depending on the specific cell type
and/or target, the deadenylated mRNA could be stored in its translational repressed stated [78-
81]. However, in animal cells, the deadenylated mRNA is generally decapped by the mRNA
decapping enzyme 2 (DCP2) and subject to rapid 5` to 3` degradation induced by cytoplasmic
exonuclease exoribonuclease enzyme 1 (XRN1) [82-86].

1.2.2 Global profiling techniques of miRNA regulation

Unlike the traditional high throughput profiling methods applied in RNA and DNA detection and
quantification, miRNA profiling approaches can be more challenging due to the unique
properties of mature miRNAs. First, mature miRNA are short in length (~22 nucleotides) and
lack in common sequences, such as poly(A) tail. As a result, the traditional primers or universal
primers cannot be used for reverse transcription and PCR. Second, long-chain pri-miRNAs and
pre-miRNAs share common RNA sequences with their corresponding mature miRNAs, making
it difficult for the hybridization-based detection approaches to discriminate between different
miRNA species. Third, the same mature miRNA from different biological samples can vary in
their lengths due to the presence of isomiRs. In addition, mature miRNAs within one family can

21
show sequence variability as little as one nucleotide. These properties of mature miRNA pose
higher requests to the sensitivity and specificity of profiling techniques. Finally, the difference in
G-C content of mature miRNAs tends to be amplified owing to the short length, which leads to a
wide variance in melting temperature (T
m
) for annealing reactions and miRNA-specific bias [87].
In spite of the challenges, three genome-wide profiling approaches are currently available: Real-
time reverse transcription quantitative PCR (qPCR), microarrays and next-generation sequencing
(RNA-Seq).

Real-time PCR based miRNA profiling requires the conversion of small RNAs into cDNA.
Priming the reverse transcription can be achieved either by enzymatically adding a poly(A) tail
or by generating a reverse transcription primer binding site using a stem-loop primer[88]. The
reaction products are then detected by one of the common technologies: SYBR Green, Taqman
probes, Molecular Beacon, Light Upon eXtension (LUX) and HybProbes[88]. Due to the
development of multi-well plates and the availability of user-friendly analyzing algorithms,
highly parallel qPCR detections of miRNAs are allowed to be incorporated into the routine
bench work. However, compared to the other two profiling approaches—microarray and RNA-
Seq, real-time PCR is only medium in throughput. As a result, qPCR detection of miRNAs is
still mainly applied as a validation tool rather than an initial large-scale exploration approach.
Another issue is that, real-time PCR can also generate false calls and thus is not an infallible
validation method of microarray analysis [89]. The ‘industry standard’ status of qPCR warrants
further interrogation with the advances in RNA-Seq and the corresponding algorithms.

22
Microarray analysis is among the most widely applied and established techniques for miRNA
profiling. After extraction and enrichment of miRNAs from biological samples, fluorescent
labeling of miRNAs is carried out enzymatically [89]. Chemically labeling strategies are also
available to avoid the potential substrate sequence bias induced by enzymatic methods [90].
Because microarray analysis is largely hybridization-based, low specificity and lack of ability to
perform absolute quantification of miRNA abundance are the major limitations of this method.
However, due to the lower cost and the higher throughput compared with other profiling
approaches, microarrays are perhaps the most favored miRNA profiling methods when large
amount of comparisons are required between biological samples from multiple sources[91].

RNA-Seq requires the preparation of a cDNA library based on the total miRNAs obtained from
samples of interest. The quality control steps of miRNA templates are much stricter in order to
insure a sufficient number of readable sequence clusters [92]. Therefore, a higher input of
samples is generally needed for RNA-Seq compared with the other two profiling approaches.
After sequencing, the raw sequences are filtered to reduce the sequencing errors. The usable
sequence reads then undergo genome alignment and precursor alignment to identify known and
novel miRNAs, as well as other small RNAs, such as piRNAs and snoRNAs [92]. Distinguishing
similar miRNAs with high accuracy and identifying novel miRNAs are the major advantages of
RNA-Seq, making it the most promising miRNAs profiling technique [87].

1.2.3 miRNA in dental development

The miRNAs involvement in the tooth genetic network fine-tuning was first suggested in 2008
[93]. To date, there are two functional studies that used the deletion of Dicer-1 to analyze

23
miRNA function during tooth development[94, 95]. The epithelial deletion of Dicer-1 which was
triggered by K14 promoter (K14-Cre), did not induce any embryonic tooth defect[94], whereas
the earlier epithelial deletion triggered by Pitx-2-Cre, or mesenchymal deletion under the control
of Wnt1-Cre, led to more drastic phenotypes[95]. The craniofacial defects caused by the Wnt1-
Cre were too severe to analyze any tooth phenotype, demonstrating the requirement of miRNA
fine-tuning in the neural crest derivatives[95]. However, the epithelial deletion of Dicer-1
through Pitx-2-Cre led to multiple branched and enamel-free incisors and cuspless molars[95]. In
vitro studies showed that miR-34a regulate human dental papilla cell differentiation by targeting
NOTCH and TGF-beta signaling [96]. MiR-143 and miR-145 control odontoblast differentiation
and dentin formation through KLF4 and OSX transcriptional factor signaling pathways [97].
Furthermore, the expression of miRNA showed a dynamic property during tooth development.
Based on microarray profiling, 8 miRNAs were identified to be both stage and tissue specific in
murine tooth formation [94]. MiR-140, miR-31, miR-875-5p and miR-141 were expressed
mainly during tooth morphogenesis (E16), whereas miR-689, miR-720, miR-711 and miR-455
were prevalent at the cytodifferentiation stage (E18). A most recent study, that combined both
deep sequencing and microarray to detect the expression profiles in the bud, cap, early bell and
late bell stages of developing lower deciduous molars obtained from miniature pigs, identified
166 differentially expressed miRNAs in the four stages [98]. The following bioinformatics
analysis suggested the key role of 18 miRNAs, including let-7f, miR-128, miR200b and miR-
200c, in tooth development. In addition, compared to ameloblast in adult mouse incisor, the
epithelial stem cell niches-the labial and lingual cervical loop also showed different miRNA
expression profiles [99]. These observations indicated that miRNAs are dynamically involved in
tooth development by fine-tuning tooth morphogenesis and patterning, as well as terminal cell

24
differentiation and tissue homeostasis. Therefore, it is highly possible that enamel maturation, as
an important stage during tooth formation might also be subject to miRNA regulation.

25
1.3 Slc26a gene family

The SLC26 gene family encodes multifunctional anion exchangers and anion channels
transporting a broad range of substrates, including Cl
-
, HCO
3
-
, sulfate, oxalate, I
-
and formate
[100]. SLC polypeptides are characterized by N-terminal cytoplasmic domain, 10-14
hydrophobic transmembrane spans, and C-terminal cytoplasmic sulfate transporter and anti-
sigma factor antagonist (STAS) domain, and appear to be homo-oligomeric [101]. Mammalian
SLC26 gene family comprises 11 genes, SLC26A1-A11 [100]. The eleven mammalian SLC26
genes are expressed throughout the body, and the expression patterns and tissue distributions
vary from member to member. Human SLC26 mutations underlie autosomal recessive
chondrodysplasias, chloride diarrhea and deafness[100]. Additional phenotypes exhibited by
mice deficient for individual SLC26 polypeptides include urolithiasis, hepatotoxicity, distal renal
tubular acidosis and male infertility [102-106].
SLC26A1/SAT1 was characterized as an anion exchanger (Cl
-
, SO
4
2-
) also transporting oxalate,
glyoxylate and Cl
-
[107, 108]. The transporter resides at the basolateral membrane of
hepatocytes, enterocytes and proximal tubular epithelial cells [109]. Although no human disease
has been linked to mutations or polymorphisms in the SLC26A1 gene, Slc26a1
-/-
mice exhibit
hyposulfatemia, hypersulfaturia, and urolitiasis and nephrocalcinosis in the setting of
hyperoxaluria [105, 110]. Mouse Slc26a6/Pat1/Cfex was identified as a mouse kidney protein
with Cl
-
/formate exchange activity [111]. SLC26A6 expression is also noted in intestine,
pancreas, heart, muscle, stomach, esophagus and placenta. SLC26A6 localizes to the apical
membrane of kidney proximal tubule and small intestine villi [111-113]. SLC26A6 can transport
Cl
-
/ HCO
3
-
exchange, Cl
-
/ OH
-
exchange, Cl
-
/ oxalate exchange, and Cl
-
/ formate exchange [111,
112, 114]. The Slc26a6
-/-
mice exhibit hyperoxaluria [115, 116]. Proximal tubular luminal

26
oxalate-dependent Cl
-
resorption is also abolished in Slc26a6
-/-
mouse[115]. SLC26A7 has been
reported to exhibit both Cl
-
/HCO
3
-
exchange activity and an anion channel of Cl
-
[117, 118].
SLC26A7 is predominantly expressed in the kidney and stomach. In the kidney, Slc26a7
colocalizes with AE1 on the basolateral or subapical membrane and endosome of A-intercalated
cells in renal outer medullary collecting duct (OMCD) [118, 119]. In the stomach, Slc26a7 co-
localizes with AE2 on the basolateral membrane of acid secreting parietal cells [117]. Deletion
of Slc26a7 causes distal renal tubular acidosis which is manifested by metabolic acidosis and
alkaline urine pH, and impaired gastric acid secretion[120]. The process of enamel maturation
involves multiple ions transport across plasma and endosome membrane. Thus, there is a need to
obtain a better understanding between the function of SLC26 gene family and amelogenesis.

27
Chapter Two: Genome-wide analysis of miRNA and mRNA transcriptomes
during amelogenesis

2.1 Abstract

Background:
In the rodent incisor during amelogenesis, as ameloblast cells transition from secretory stage to
maturation stage, their morphology and transcriptome profiles change dramatically. Prior whole
genome transcriptome analysis has given a broad picture of the molecular activities dominating
both stages of amelogenesis, but this type of analysis has not included miRNA transcript
profiling. In this study, we set out to document which miRNAs and corresponding target genes
change significantly as ameloblasts transition from secretory- to maturation-stage amelogenesis.

Results:
Total RNA samples from both secretory- and maturation-stage rat enamel organs were subjected
to genome-wide miRNA and mRNA transcript profiling. We identified 59 miRNAs that were
differentially expressed at the maturation stage relative to the secretory stage of enamel
development (False Discovery Rate (FDR)<0.05, fold change (FC)≥1.8). In parallel,
transcriptome profiling experiments identified 1,729 mRNA transcripts that were differentially
expressed in the maturation stage compared to the secretory stage (FDR<0.05, FC ≥1.8). Based
on bioinformatics analyses, 5.8% (629 total) of these differentially expressed genes (DEGS)
were highlighted as being the potential targets of 59 miRNAs that were differentially expressed
in the opposite direction, in the same tissue samples. Although the number of predicted target
DEGs was not higher than baseline expectations generated by examination of stably expressed

28
miRNAs, Gene Ontology (GO) analysis showed that these 629 DEGS were enriched for ion
transport, pH regulation, calcium handling, endocytotic, and apoptotic activities. Seven
differentially expressed miRNAs (miR-21, miR-31, miR-488, miR-153, miR-135b, miR-135a
and miR298) in secretory- and/or maturation-stage enamel organs were confirmed by in situ
hybridization. Further, we used luciferase reporter assays to provide evidence that two of these
differentially expressed miRNAs, miR-153 and miR-31, are potential regulators for their
predicated target mRNAs, Lamp1 (miR-153) and Tfrc (miR-31).

Conclusions:
In conclusion, these data indicate that miRNAs exhibit a dynamic expression pattern during the
transition from secretory-stage to maturation-stage tooth enamel formation. Although they
represent only one of numerous mechanisms influencing gene activities, miRNAs specific to the
maturation stage could be involved in regulating several key processes of enamel maturation by
influencing mRNA stability and translation.

29
2.2 Background

Amelogenesis is the developmental process of dental enamel formation. Amelogenesis involves
two major functional stages, secretory and maturation, and these stages are clearly demarcated
by a transition zone in the continuously growing rodent incisor teeth [1]. The transition of
ameloblasts from secretory to maturation stage, characterized by both morphological and
functional changes, results in the formation of mature enamel with ordered crystallite structures.
Gene dysregulation at any stage of amelogenesis can result in a group of hereditary conditions
called Amelogenesis Imperfecta (AI) that adversely affect the structure and appearance of enamel
[58, 121-124]. Although researchers today have a very clear idea of the molecular activities that
define secretory-stage amelogenesis [1], the molecular events that define enamel maturation
remain understudied.

MicroRNAs (miRNA) are a class of small non-coding RNAs that regulate the expression of
target genes by directly binding to their target mRNAs. To date, there are two functional studies
that used the deletion of Dicer-1 to analyze miRNA function during tooth development [125,
126]. The epithelial deletion of Dicer-1, using the keratin 14 gene promoter-Cre recombinase
combination (K14-Cre), does not induce embryonic tooth defects [126], whereas the earlier
epithelial deletion of Dicer-1, triggered by Pitx2-Cre, or mesenchymal deletion under the control
of Wnt1-Cre, led to a severe dental phenotype [125]. In vitro studies showed that miR-34a
regulates human dental papilla cell differentiation by targeting NOTCH and TGF-beta signaling
[127]. MiR-143 and miR-145 control odontoblast differentiation and dentin formation through
KLF4 and OSX transcriptional factor signaling pathways [128].

30
changes in miRNA levels have been observed during tooth development. Based on microarray
profiling studies, 8 miRNAs have been identified to be both stage- and tissue-specific in murine
tooth formation [126]. That is, miR-140, miR-31, miR-875-5p and miR-141 were expressed
mainly during tooth morphogenesis identified at embryonic day 16 (E16), whereas miR-689,
miR-720, miR-711 and miR-455 were prevalent at the cytodifferentiation stage (E18) [126]. A
more recent study that combined both deep sequencing and microarray approaches to elucidate
the miRNA expression profiles in the bud, cap, early bell and late bell stages of developing lower
deciduous molars of miniature pigs identified 166 miRNAs expressed differentially across the
four stages [98]. A subsequent bioinformatic prediction suggested that 18 of these miRNAs play
key roles during tooth development, including let-7f, miR-128, miR-200b and miR-200c [98].
Two epithelial stem cell niches, located in the labial and lingual cervical loop regions, have been
identified and shown to have different miRNA expression profiles [99]. Together these
observations indicate that miRNAs are dynamically involved in tooth development by fine-
tuning tooth morphogenesis and patterning, as well as terminal cell differentiation and tissue
homeostasis.

To investigate the potential role of miRNA regulation in maturation-stage tooth development, we
conducted genome-wide miRNA and mRNA transcript expression profiling analyses of
secretory-stage and maturation-stage enamel organs obtained from rat incisors. We identified a
group of stage-specific miRNAs and identified candidate gene targets based on bioinformatic
prediction. Two maturation-stage-related genes, Lamp1 and Tfrc, were verified by luciferase
reporter assay to be the target genes of miRNA regulators. The results indicated a dynamic
expression pattern of miRNAs during the transition from secretory-stage to maturation-stage

31
enamel mineralization, and suggest that miRNAs can influence key processes of enamel
maturation.

32
2.3 Methods

Animal dissection and total RNA isolation
All vertebrate animal studies complied with Institutional and Federal guidelines (Institutional
Animal Care and Use Committee (IACUC) protocol number 11736). We used rat incisors as the
source of total RNA, because the reference line that separates the secretory- and maturation-stage
enamel organs along the enamel surface has been well established in rats [3, 129]. Four male
Wistar Hannover rats, 4-week-old, weighing 100-110g, were sacrificed for their mandibles. After
being frozen and kept in liquid nitrogen overnight, the mandibles were subsequently lyophilized
over the following 24h. The bone encasing the enamel surface of incisors was then carefully
removed and the exposed multicellular layer, which contains mostly the secretory- and
maturation-stage enamel organs, was collected into separate RNase-free Eppendorf tubes.
Dissection procedures followed previously described protocols [129, 130]. The total RNA
including miRNA was extracted from secretory- and maturation-stage enamel organs using
miRNeasy Mini Kit (Qiagen, Valencia, CA, USA). The enamel organs isolated from the four rats
were fully processed and analyzed separately (RNA extraction, miRNA qPCR, whole genome
array analysis and bioinformatics).

Sample quality control by quantitative real-time PCR analysis
Using previously described methods, the expression of two stage-specific genes, Odam (most
highly expressed during maturation stage) and Enam (most highly expressed during secretory
stage), were checked by real-time PCR to ensure the accuracy of dissections and quality of total
RNA collected from each individual sample [16, 129]. cDNA used for real-time PCR analysis of
Odam and Enam was synthesized using miScript II RT Kit with miScript HiFlex Buffer

33
(Qiagen). Real-time PCR reactions were performed with iQ SYBR® Green supermix (Bio-rad
Life Sciences, Hercules, CA) and rat-specific primers (Odam-Forward: 5`-
ATCAATTTGGATTTGTACCACA-3’, Odam-Reverse: 5`-
CGTCGGGTTTATTTCAGAAGTGA-3’, Enam-Forward: 5`-
TGCAGAAATACAGCTTCTCCT-3’, Enam-Reverse: 5`-CATTGGCATTGGCATGGCA-3’,
Actb-Forward: 5`-AGTGTGACGTTGACATCCGTA-3’, Actb-Reverse: 5`-
GCCAGGGCAGTAATCTCCTTCT-3’). For all RNA sample pairs (four rats each with
secretory-stage and maturation-stage enamel organs) quantitative real-time PCR for Odam
showed an increase in expression in maturation-stage by >130 fold, while Enam expression was
clearly evident in the secretory stage, but negligible in the maturation stage. These results
indicated that the samples were suitable to use for additional genome profiling experiments.

Genome-wide miRNA and mRNA profiling analysis
Genome-wide miRNA profiling analysis was conducted based on the Rat miRNome miScript
miRNA PCR Array (V16.0, 384-well; Qiagen), which profiles the expression of the 653 most
abundantly expressed and best characterized miRNA sequences in the rat miRNA genome as
annotated by the miRBase Release 16. cDNA was prepared using miScript II RT Kit with
miScript HiSpec Buffer (Qiagen). MiScript SYBR Green PCR Kit (Qiagen) was used for the
real-time PCR reactions on the miScript miRNA PCR Array and the real-time instrument was a
LightCycler 480 (Roche Applied Science, IN, USA).

On the exact same RNA samples used for miRNA expression profiling analysis, we conducted
genome-wide mRNA transcriptome analysis using RatRef-12-v1 Expression BeadChips

34
(Illumina, Inc., San Diego, CA). This platform interrogates approximately 22,000 transcripts
selected primarily from the National Center for Biotechnology Information (NCBI) RefSeq
database (Release 16). All sample preparation, hybridization, BeadChips processing, and data
acquisition were performed at the Southern California Genotyping Consortium (SCGC),
according to the manufacturers’ recommended protocols.

MiRNA and mRNA expression profiling analyses were conducted on four animals, with each
animal providing both secretory- and maturation-stage RNA samples. Each individual RNA
sample was analyzed both for miRNA expression (triplicate technical replicates to ensure
quality) and for mRNA expression. In total, 24 miRNA (8 individual samples in triplicate) and 8
mRNA global expression data sets were generated in this study. Original gene expression data
files are available for download from the Gene Expression Omnibus
(http://www.ncbi.nlm.nih.gov/geo/) under GEO accession record GSE59401.

MiRNA and mRNA gene expression data processing
Raw miRNA expression data generated on miRNome miScript miRNA PCR Arrays were
processed using SAS 9.2 statistical software. All raw data were acquired in the form of Ct
values. The raw Ct values were normalized to the average Ct values of the six internal controls
located in the last row of each 384-well plate. For the purpose of downstream analyses, we
assigned a single expression score to each miRNA based on the average of data obtained from
three replicates. All processed data from all 24-miRNA expression profiling experiments are
provided in Additional File 1.

35
Raw mRNA gene expression data generated using RatRef-12-v1 Expression BeadChips were
processed using the R statistical package to produce normalized logarithm-transformed gene
expression scores [130]. Processed data from all 8 mRNA expression profiling experiments are
provided in Additional File 2. Differences in the expression levels of each mRNA and miRNA
between secretory- and maturation-stage tooth development were evaluated using two-tail
Student t test, and the type I error was controlled using the Benjamini Hochberg (B-H) method.
In Additional File 3, we provide a summary of the numbers of differentially expressed mRNAs
and miRNAs based on a variety of commonly used FC and FDR criteria. Hierarchical clustering
analyses of mRNA and miRNA data sets were based on Euclidean distance and average linkage
metrics and conducted using Partek
®
Genomics Suite version 6.5 (PGS) (Figure 1).

Pathway analysis
We used Ingenuity Pathway Analysis (IPA) (Ingenuity Systems, Redwood City, CA) to predict
gene targets for differentially expressed miRNAs identified by qPCR array analysis. In order to
refine further the number of predicted targets for differentially expressed miRNAs, we used IPA
to compare the list of predicted gene targets and the list of maturation stage-specific genes
identified by mRNA transcriptome profiling on BeadChips. Genes in common between these
two lists were uploaded to WebGestalt for Gene Ontology (GO) and KEGG (Kyoto
Encyclopedia of Genes and Genomes) pathway analysis [131, 132] (Figure 2). The selection
criteria for enriched GO categories were B-H adjusted P<0.05 for Fisher’s exact test and
minimum 5 genes in enriched categories. IPA also was used to analyze the functional
relationships between differentially expressed miRNAs and these overlapping genes (Figure 2).

36
To begin to estimate baseline expectations of how miRNA level could influence mRNA levels in
our system, we generated a list of stably expressed miRNAs present in both the secretory and
maturation stages (herein defined as having raw Ct values ≤38 stages and whose relative
expression levels showed no statistical differences between the two stages (FDR≥0.05)). As
above, we used IPA to predict mRNA targets for these stably expressed miRNAs and compared
this list of predicted targets with the differentially expressed mRNAs identified through mRNA
transcriptome profiling on BeadChips.

In situ hybridization analysis of selected miRNAs expression
The mandibles were dissected out from euthanized Wistar Hannover rats (~100g body weight ，4
weeks-old), with the surrounding soft tissues removed. The hemimandibles were then fixed in
4% paraformaldehyde (PFA) at 4° C overnight, decalcified in 10% EDTA (pH 7.4) at 4 ° C for 10
weeks, and embedded in paraffin for sectioning. Sagittal sections of 7µ m were prepared and
LNA-DIG miRNA detection probes, including U6 probes (positive control) and scrambled
probes (Exiqon, Inc., MA, USA), were utilized for miRNA in situ hybridization. All the
procedures for the in situ hybridization analyses were performed following the one-day protocol
recommended by the manufacturers, with the probe concentration and substrate incubation time
adjusted for each of tested probes individually. Information regarding probe sequences and the
complete in situ hybridization protocol can be found on the website of Exiqon
(www.exiqon.com).

37
Cell culture and luciferase reporter assay
Mouse ameloblast-like LS8 cells [14, 133] were used as the host cells for exogenous miRNA
mimics, miRNA inhibitors and luciferase reporter vectors in the luciferase reporter assay. LS8
cells were cultured in low-glucose DMEM medium (Gibco® Life Technologies, Grand Island,
NY, USA) supplemented with 10% Fetal Bovine Serum (FBS) at 37 ° C in a 5% CO
2

atmosphere. LS8 cells were seeded in 12-well cell culture plates to achieve a confluence of
approximately 30% the day before transfection. Lipofectamine® LTX with Plus
TM
Reagent (Life
Technologies) was diluted in FBS-free low-glucose DMEM medium at a concentration
recommended by the manufacturers.

Luciferase reporter vectors containing 3’-UTR of target genes and/or miRNA mimics/inhibitors
were mixed into diluted transfection reagents to form a transfection complex. Immediately
before transfection, the cell culture medium was changed to FBS-free low-glucose DMEM
medium. FBS-free low-glucose DMEM medium was removed 3h after transfection, and cells
were then incubated in low-glucose DMEM medium with 10% FBS. 48h after transfection, cells
were lysed using the passive lysis buffer provided in the Dual-Luciferase® Reporter Assay
System (Promega, San Luis Obispo, CA, USA). Luciferase Assay Reagent (LARII) and Stop &
Go reagent were then added to the cell lysates sequentially. The luciferase reporter activities
were detected using a Turner Biosystems Luminometer TD-20/20 according to the
manufacturers’ recommended protocol. Each luciferase reporter assay was conducted with
triplicate technical replicates.

38
Two kinds of dual luciferase reporter vectors containing the 3’-UTR of different mouse-specific
target genes (Lamp1 and Tfrc) were purchased from GeneCopoeia (Catalog # MmiT029570-
Lamp1, MmiT030401-Tfrc). MiRNA mimics and inhibitors for miR-153 and miR-31 were also
mouse-specific and were obtained from Qiagen (Catalog # MSY0000163-miR-153 mimic,
MIN0000163-miR-153 inhibitor, MSY0000538-miR-31 mimic, MIN0000538-miR-31 inhibitor).
For verifying the relations between miR-153 (mature miRNA sequence: 5`-
UUGCAUAGUCACAAAAGUGAUC-3’) and Lamp1 expression, the setup of experimental
groups involved LS8 cells: 1) transfected with luciferase reporter vector (3`-UTR of Lamp1); 2)
co-transfected with miR-153 mimics and luciferase reporter vector (Lamp1 3`-UTR); 3) co-
transfected with miR-153 inhibitors and luciferase reporter vector (Lamp1 3`-UTR). For miR-31
(mature miRNA sequence: 5`-AGGCAAGAUGCUGGCAUAGCUG-3’) and Tfrc, the
experimental groups involved LS8 cells: 1) transfected with luciferase reporter vector (Tfrc 3’-
UTR); 2) co-transfected by miR-31 mimics and luciferase reporter vector (Tfrc 3’-UTR); 3) co-
transfected by miR-31 inhibitors and luciferase reporter vector (Tfrc 3’-UTR).

The amount of luciferase reporter vector was stabilized at 700ng per transfection, to achieve
optimal DNA transfection efficiency. The tested concentrations of miRNA mimics in final
transfection complex (after being added into FBS-free cell culture medium in a 12-well plate)
were 20pM, 60pM and 120pM, while the tested concentrations of miRNA inhibitors were
0.2nM, 0.6nM and 1.2nM (note that it is recommended by the manufacturers that the
concentration of miRNA inhibitors should be 10 times that of miRNA mimics).

39
Renilla luciferase activities were normalized to the firefly luciferase activities. For each
verification experiment (miR-153 with Lamp1 or miR-31 with Tfrc), two-tail Student t-test was
used to detect the statistical differences in normalized luciferase activities between experimental
groups 1 and 2, and between groups 1 and 3. In real-time PCR analyses of miRNA levels in LS8
cells following transfection, the raw Ct values were normalized to those of RNU6-2. The level of
miRNA at each time point after transfection was calculated relative to the level of RNU6-2 using
the ΔCt method. The significance level for all statistical analyses mentioned above was P<0.05.

Real-time PCR analysis of cellular miRNA levels
Prior to luciferase reporter assays, the levels of miR-153 and miR-31 in LS8 cells were detected
using miRNA real-time PCR analysis at different time points: 0h, 6h, 24h and 48h following
transfection by miRNA mimics or inhibitors. The levels of miR-153 and miR-31 were calculated
relative to the level of RNA U6 Small Nuclear 2 (RNU6-2), as recommended by the
manufacturers. The protocols for RNA extraction, cDNA synthesis and real-time PCR reactions
were similar to the protocols of qPCR array analyses stated above. The mouse-specific primers
of miR-153, miR-31 and RNU6-2 were purchased from Qiagen (Catalog # MS00011214-miR-
153 primers, MS00001407-miR-31 primers, MS00033740-RNU6-2 primers), and the primer
sequences were not disclosed by the manufacturers.

40
2.4 Results

miRNAs are differentially expressed in enamel organs as they transition from secretory
stage to maturation stage in tooth development
The miRNA expression profiles of maturation-stage enamel development were compared to
those of secretory-stage enamel development using total RNA samples obtained from the enamel
organs of rat incisors. Although we considered multiple statistical cut-offs (Additional Files 3
and 4), herein we assigned differential expression based on ≥1.8-FC and <5% FDR between the
two developmental stages. This provided a robust set of differentially expressed miRNAs for
follow-up analysis.

Hierarchical clustering analysis of the most variably expressed miRNAs across all samples
showed that the maturation and secretory stages of enamel development have distinct miRNA
expression profiles (Figure 1). All the maturation-stage samples and all the secretory-stage
samples were placed into non-overlapping groups. A total of 59 out of 653 miRNAs were
identified as being differentially expressed within maturation-stage enamel organs when
compared to secretory-stage enamel organs. Among these 59 stage-specific miRNAs, 39 were
up-regulated during maturation stage (relative to secretory stage) (Figure 2, Additional File 1)
while 20 were down-regulated (Figure 3, Additional File 1). All raw Ct values and relevant
statistics can be found in the supplemental information (Additional File 1).

41

42
Figure 1. Hierarchical clustering analysis of the expression levels of miRNAs with highest
variability (Coefficient of Variance (CV) > 0.15) across all samples. All four pairs of samples are
labeled in acronyms (Maturation-stage RNA sample 1-4: M1-4, Secretory-stage sample 1-4: S1-
4). The color scale from red to green indicates relative abundance of miRNAs from higher to
lower.

43

Figure 2. Fold-changes of 39 up-regulated miRNAs (FDR<0.05) at maturation stage relative to
secretory stage based on miRNA qPCR array analysis.

44

Figure 3. Fold-changes of 20 down-regulated miRNAs (FDR<0.05) at maturation stage relative
to secretory stage based on miRNA qPCR array analysis.

45
Differentially expressed genes are identified at mRNA level

The eight total RNA samples used for parallel genome-wide transcript profiling (i.e., four
secretory-stage samples and four maturation-stage samples) were from the same samples as those
used for miRNA expression profiling such that the two data sets (genome-wide transcript and
miRNA expression profiling) could be matched with each other. To be consistent with our
miRNA data analysis and also identify robust signals, we used the same criteria for differential
gene expression as above (≥1.8-FC, <5% FDR) between the two developmental stages. A total of
1,729 genes were differentially expressed (701 down- and 1,028 up-regulated) in the maturation-
stage enamel organs compared to the secretory-stage enamel organs (Additional File 2).

Identifying differentially expressed genes that are predicted targets of the differentially
expressed miRNAs
The strategy we used to identify targets of differentially expressed miRNAs is provided (Figure
4). The list of differentially expressed miRNAs was uploaded to IPA to generate predicted gene
targets using integrated prediction algorithms (TargetScan, TarBase, miRecords and
Ingenuity® Knowledge Base). A total of 8,492 unique candidate target genes were predicted to
be regulated by the 39 up-regulated (maturation- relative to secretory-stage) miRNAs. Likewise,
6,518 unique candidate target genes were predicted to be regulated by the 20 down-regulated
(maturation- relative to secretory-stage) miRNAs. In parallel, we used the same computational
methods to predict 16492 unique gene targets for the 516 stably expressed miRNAs in these
maturation and secretory stage samples (see Methods).

46

Figure 4. Flow chart depicting the strategy used to select miRNA target genes for pathway
analyses.

47
Next, we sought to refine the candidate relationships between predicted gene targets of
differentially expressed miRNAs with our microarray-based measurements of mRNA levels.
Approximately 5.8% (629/10,786) of the candidate target genes were differentially expressed in
the expected direction in our analysis (i.e. over-expressed miRNA coincides with reduced
expression of candidate target mRNA and vice versa). Next, we compared the list of predicted
gene targets of up-regulated miRNAs with the observed list of down-regulated mRNAs during
enamel maturation (Figure 4). A total of 299 observed down-regulated genes were identified as
the potential targets for the 39 up-regulated miRNAs (Additional File 5). A total of 141 out of
these 299 genes were predicted to be regulated by multiple miRNAs (Additional File 5;
highlighted in bold). Conversely, we compared the predicted gene targets of down-regulated
miRNAs with the observed list of up-regulated mRNAs during amelogenesis. A total of 330 up-
regulated genes were identified as potential targets for the 20 down-regulated miRNAs
(Additional File 6). Among these 330 genes, 105 were predicted to have multiple miRNA
regulators (Additional File 6; highlighted in bold).

To place our observations in perspective, we compared the list of predicted gene targets of stably
expressed miRNAs with the list of differentially expressed mRNAs (maturation- relative to
secretory-stage). Approximately 8.1% of the candidate target mRNAs (1341 total: 828 up-
regulated and 513 down-regulated) were identified to be the potential targets for the stably
expressed miRNAs (Additional Files 19-21). Thus, it is not remarkable that 5.8% of the
candidate target genes of differentially expressed miRNAs are differentially expressed in the
expected direction. There are numerous potential reasons for this including (i) transcription
factors will play a dominant role in influencing mRNA levels in accordance with our prior

48
studies [129, 130] and (ii) the magnitude of the miRNA expression changes observed in our
study where at least half of the protein-coding genes in the rodent genome were implicated as
potential targets of differentially expressed miRNAs.

Thus, outside of translational regulation outside the scope of the current study, we propose that
miRNAs more likely serves to fine-tune the levels of transcripts that are rapidly up-regulated or
down-regulated during amelogenesis. For example, if the immediate up-regulation of a gene
transcription is needed, then excessive mRNA may result, requiring immediate miRNA-targeting
to reach optimal levels. In such cases, increased levels of specific targeting miRNA may be
required to achieve this fine-tuning of highly up-regulated transcripts. To begin to address this
issue, we identified 41 highly up-regulated mRNAs (≥5 fold, <5% FDR) during amelogenesis
that could be paired with predicted targeting miRNAs that were also differentially expressed
(≥1.8-fold, <5%FDR) during amelogenesis (Additional File 7). A subset of these transcripts was
subject to follow-up experiments described below.

Candidate miRNA regulated genes are enriched in key processes involved in enamel
maturation
Next, we used various pathway analyses to explore relationships among the differentially
expressed mRNAs that were candidate targets of differentially expressed miRNAs. GO analysis
of the 330 up-regulated genes identified as potential targets for the 20 down-regulated miRNAs
highlighted 120 enriched categories (Additional File 8). The categories most highly relevant to
maturation-stage tooth development included: carboxylic acid transmembrane transporter
activity (11 genes) (Table 1); ATPase activity, coupled to transmembrane movement of ions,

49
phosphorylative mechanism (5 genes) (Table 2); endosome membrane (18 genes) (Table 3);
lysosome (14 genes) (Table 4); calcium ion binding (27 genes) (Table 5); and cytokine activity
(11 genes) (Table 6). GO analysis of the 299 down-regulated genes identified as potential targets
for the 39 up-regulated miRNAs highlighted 82 enriched categories (Additional File 9). The
categories that are seemingly most relevant to maturation-stage tooth development included:
calcium ion transmembrane transporter activity (11 genes) (Table 7); and extracellular matrix
part (8 genes), cation channel complex (7 genes), Golgi membrane (20 genes), GTPase activity
(11 genes), regulation of cell-cell adhesion (7 genes) and cell junction (23 genes) (Additional
File 9).

50
Table 1. Genes enriched in the GO category “carboxylic acid transmembrane transporter
activity” and their predicted miRNA regulators All fold changes represent the ratio of maturation
to secretory phase expression values.
Human
Gene
Symbol
mRNA
Fold change
Predicted miRNA
regulator(s)
miRNA
Fold change
SLC1A1 46.7 rno-miR-298 -4.3
SLC6A8 5.6 rno-miR-135a -8.9
SLC6A14 40.6 rno-miR-3085 -4.3
SLC16A6 3.1 rno-miR-135a/376b-3p -8.9/-2.8
SLC22A5 1.8 rno-miR-138 -2.7
SLC23A2 4.3 rno-miR-138 -2.7
SLC25A15 2.6 rno-miR-138/3085 -2.7/-4.3
SLC26A1 39.8 rno-miR-138/298 -2.7/-4.3
SLC26A7 7.8 rno-miR-135a/153/298 -8.9/-3.7/-4.3
SLC36A1 1.9 rno-miR-298/3085/346 -4.3/-4.3/-3.6
SLC38A1 2.0 rno-miR-138/153 -2.7/-3.7

51
Table 2. Genes enriched in the GO category “ATPase activity, coupled to transmembrane
movement of ions, phosphorylative mechanism” and their predicted miRNA regulators. All fold
changes represent the ratio of maturation to secretory phase expression values.

Human
Gene symbol
mRNA
Fold change
Predicted miRNA
regulator(s)
miRNA
Fold change

ATP2A2 2.2 rno-miR-298 -4.3
ATP1B2 2.2 rno-miR-298/410 -4.3/-2.3
ATP1B3 2.4 rno-miR-153 -3.7
ATP6V1C1 2.5 rno-miR-135a/153 -8.9/-3.7
ATP6V1E1 3.0 rno-miR-135a -8.9

52
Table 3. Genes enriched in the GO category of “endosome membrane” and their predicted
miRNA regulators. All fold changes represent the ratio of maturation to secretory phase
expression values.

Human
Gene symbol
mRNA
Fold change

Predicted miRNA
regulator(s)
miRNA
Fold change

ATP6V0A1 2.2 rno-miR-138 -2.7
CD68 4.6 rno-miR-135a -8.9
CFTR 4.9 rno-miR-298/3085/376b-3p -4.3/-4.3/-2.8
ECE1 2.8 rno-miR-138 -2.7
EHD3 4.6 rno-miR-138/153 -2.7/-3.7
EHD4 3.3 rno-miR-376b-3p -2.8
FCGR1A 1.8 rno-miR-3085 -2.7
LAMP1 3.2 rno-miR-153 -3.7
MYD88 2.4 rno-miR-138/298/3085/135a -2.7/-4.3/-4.3/-
8.9
PARM1 11.1 rno-miR-203 -2.2
PMEPA1 4.9 rno-miR-410 -2.3
RAB21 1.9 rno-miR-410 -2.3
SLC11A2 3.0 rno-miR-203 -2.2
SLC26A7 7.8 rno-miR-135a/153/298 -8.9/-3.7/-4.3
TFRC 28.6 rno-miR-490* -8.6
TMBIM1 5.5 rno-miR-3085 -4.3
VPS37B 2.1 rno-miR-3085/376b-3p -4.3/-2.8
ZNRF2 1.9 rno-miR-153 -3.7

53
Table 4. Genes enriched in the GO category of “lysosome” and their predicted miRNA
regulators. All fold changes represent the ratio of maturation to secretory phase expression
values.

Human
Gene symbol
mRNA
Fold change

Predicted miRNA
regulator(s)
miRNA
Fold change

BGN 2.3 rno-miR-3085 -4.3
CD68 4.6 rno-miR-135a -8.9
CTSS 2.0 rno-miR-203 -2.2
FMOD 3.5 rno-miR-203/138 -2.2/-2.7
LAMP1 3.2 rno-miR-153 -3.7
P2RY2 2.4 rno-miR-135a -8.9
PDGFRB 2.8 rno-miR-3085 -4.3
PON2 1.9 rno-miR-376b-3p -2.8
SDC3 1.8 rno-miR-138 -2.7
SLC11A2 3.0 rno-miR-203 -2.2
SLC36A1 1.9 rno-miR-298/3085/346 -4.3/-4.3/-3.6
STS 2.3 rno-miR-138 -2.7
TMBIM1 5.5 rno-miR-3085 -4.3
ZNRF2 1.9 rno-miR-153 -3.7

54
Table 5. Genes enriched in the GO category of “calcium ion binding” and their predicted
miRNA regulators. All fold changes represent the ratio of maturation to secretory phase
expression values.

Human
Gene symbol
mRNA
Fold change

Predicted miRNA
regulator(s)
miRNA
Fold change

ANXA8L2 5.3 rno-miR-298/3085 -4.3/-4.3
ATP2A2 2.2 rno-miR-298 -4.3
BMP1 2.0 rno-miR-138 -2.7
CDH13 2.0 rno-miR-153 -3.7
CDH17 2.2 rno-miR-298 -4.3
CHP1 1.8 rno-miR-135a/298 -8.9/-4.3
CIB2 4.2 rno-miR-346/153 -3.6/-3.7
DSG2 2.0 rno-miR-153 -3.7
DUOX1 2.4 rno-miR-298 -4.3
EFHD2 2.8 rno-miR-138/153 -2.7/-3.7
EHD3 4.6 rno-miR-138/153 -2.7/-3.7
EHD4 3.3 rno-miR-376b-3p -2.8
FAT3 2.9 rno-miR-153/203/3085 -3.7/-2.2/-4.3
FCN1 2.6 rno-miR-3085 -4.3
GCH1 5.0 rno-miR-490* -8.6
LCP1 3.0 rno-miR-135a -8.9
MAN1A1 4.5 rno-miR-135a/3085 -8.9/-4.3
MEGF6 2.8 rno-miR-135a/3085 -8.9/-4.3
MMP14 1.9 rno-miR-298/3085/410 -4.3/-4.3/-2.3
PLCB3 2.1 rno-miR-298 -4.3
RUNX1 2.2 rno-miR-298/3085/410/135a -4.3/-4.3/-2.3/-8.9
RYR3 2.1 rno-miR-153 -3.7
SCUBE1 2.4 rno-miR-298 -4.3
SPARCL1 2.6 rno-miR-153 -3.7
STAT3 2.3 rno-miR-410 -2.3
STIM2 3.8 rno-miR-153/154* -3.7/-2.5
SULF2 2.3 rno-miR-138 -2.7

55
Table 6. Genes enriched in the GO category of “cytokine activity” and their predicted miRNA
regulators. All fold changes represent the ratio of maturation to secretory phase expression
values.

Human
Gene symbol
mRNA
Fold change

Predicted miRNA
regulator(s)
miRNA
Fold change

BMP1 2.0 rno-miR-138 -2.7
CCL19 24.7 rno-miR-298 -4.3
CSF1 2.1 rno-miR-3085 -4.3
CX3CL1 15.6 rno-miR-298 -4.3
GAB1 1.9 rno-miR-153/376b-3p/410 -3.7/-2.8/-2.3
IL1B 2.0 rno-miR-3085 -4.3
INHBA 4.8 rno-miR-135a -8.9
TNFSF11 2.4 rno-miR-3085/410 -4.3/-2.3
TNFSF13 2.4 rno-miR-298 -4.3
WNT5A 3.0 rno-miR-410 -2.3
VEGFA 3.2 rno-miR-203/410 -2.2/-2.3

56
Table 7. Genes enriched in the GO category of “calcium ion transmembrane transporter activity”
and their predicted miRNA regulators. All fold changes represent the ratio of maturation to
secretory phase expression values.

Human
Gene symbol
mRNA
Fold change

Predicted miRNA
regulator(s)
miRNA
Fold change

ATP2B1 -5.5 rno-miR-223/27a/384-
5p/488/489
2.6/2.0/19.2/6.9/67.
0
ATP2B3 -3.3 rno-miR-351 2.2
CACFD1 -1.8 rno-miR-137 14.1
CACNA1D -2.0 rno-miR-137/384-5p/489 2.2/19.2/67.0
CACNA1G -1.8 rno-miR-137/702-5p 14.1/2.6
CACNB2 -4.1 rno-miR-351/137/27a/384-
5p/31/9
2.2/14.1/2.0/19.2/9.
5/ 10.0
CACNB3 -1.9 rno-miR-351 2.2
CHRNA10 -5.5 rno-miR-351 2.2
NCS1 -2.7 rno-miR-223 2.6
SLC24A3 -2.4 rno-miR-137 14.1
SLC8A3 -3.8 rno-miR-21/489 5.2/67.0

57
We also conducted KEGG analysis and IPA Core Analysis for each group of differentially
expressed genes with and without involvement of miRNAs (Additional Files 10-18). In general,
we did not find noteworthy categories relevant to maturation-stage tooth development.
Nevertheless, we did find KEGG pathways, including calcium signaling, ECM-receptor
interaction, lysosome and endocytosis, that could be relevant to enamel maturation (Additional
Files 10 and 11), and several IPA canonical pathways that are highly conserved during
amelogenesis (Additional Files 12-15). The IPA-derived interaction networks of differentially
expressed miRNAs and their predicted gene targets showed clearly that miRNAs are regulatory
hubs in these enriched functional pathways (Figure 5). In addition, several genes were targets of
multiple miRNAs and served to connect these hubs. The details regarding the miRNA-mRNA
interaction networks are provided in the Additional Files (Additional Files 12-18).

58

Figure 5. Highest scoring miRNA-mRNA interaction network, “Cellular Movement,
Neurological Diseases, Organismal Injury and Abnormalities.” Figure was generated based on
IPA software analysis of up-regulated miRNAs (maturation/secretory (M/S)) and their predicted
gene target candidates that were down-regulated (M/S) with a top score of 110 and involvement
of 62 molecules. The miRNA hubs, the genes that have multiple potential miRNA regulators,
and the experimentally validated genes are colored red, blue and orange, respectively. The
remaining genes in the network are colored green.

59
The expression of stage-specific miRNAs confirmed by in situ hybridization
LNA-DIG miRNA probes were used to detect miRNA expression in enamel organs along the
enamel surface of rat incisors (Figure 6). Mandible section slides stained with U6 LNA-DIG
served as the positive controls (Figure 6B) and scrambled controls showed little background
(Figures 6C and 6D). The expression patterns of seven selected miRNAs (miR-21, miR-31, miR-
488, miR-153, miR-135b, miR-135a and miR-298) were examined (Figure 6E – 6R). In situ
hybridization analyses of miR-21, miR-31 and miR-488 generated higher signal intensities in
maturation-stage enamel organ cells than in secretory-stage enamel organ cells (Figure 6E – 4J),
and this data correlates well (same directional change) with the miRNA qPCR array data/fold
increase (i.e., miR-21, miR-31, miR-488 increased by 5.2, 9.5 and 6.9 fold, respectively)
(Additional File 1). In contrast, as seen in the in situ slides, there was a decrease in the
expression levels of miR-153, miR-135b, miR-135a and miR-298 from secretory-stage to
maturation-stage tooth development (Figure 6K – 6R). This trend of expression correlates well
with the miRNA qPCR array data (i.e., miR-153, miR-135b, miR-135a and miR-298 changed by
-3.7, -12.4, -8.9 and -4.3 fold, respectively) (Additional File 1).

60

61
Figure 6. Expression patterns of seven miRNAs in secretory- and maturation-stage enamel
organs as shown by in situ hybridization. (A) Negative control. Samples were incubated without
any miRNA detection probes. (B) Positive control. U6 detection probes were used for
incubation. (C) Scrambled control for secretory-stage enamel organ. (D) Scrambled control for
maturation-stage enamel organ. (E), (G), (I), (K), (M), (O) and (Q) The expression of miR-21,
miR-31, miR-488, miR-153, miR-135b, miR-135a and miR-298 in secretory-stage enamel organ.
MiR-21, miR-31 and miR-488 are down-regulated, while miR-153, miR-135b, miR-135a and
miR-298 are up-regulated in secretory-stage enamel formation compared to maturation-stage.
(F), (H), (J), (L), (N), (P) and (R) The expression of miR-21, miR-31, miR-488, miR-153, miR-
135b, miR-135a and miR-298 in maturation-stage enamel organ, showing altered maturation-
stage expression patterns. Images shown at 20x magnification. S, secretory stage. M, maturation
stage.

62
Lamp1 and Tfrc are potential gene targets of miRNA regulators
According to previous bioinformatic studies, miR-153 is a predicted regulator of LAMP1, while
miR-490* is a predicted regulator of TFRC (Table 3, Additional File 6). However, these target
predictions for differentially expressed miRNAs were based mainly on human genome
predictions; thus, we searched TargetScan (http://www.targetscan.org/) for the predicted miRNA
regulators of these two genes within the mouse genome. We found that the prediction of miR-
153 as the regulator of LAMP1/Lamp1 was consistent between the human and mouse genomes.
However, miR-490* was not listed as one of the predicted regulators for mouse-specific Tfrc.
One of the miRNAs identified as being highly up-regulated in maturation-stage enamel organ
cells is miR-31 (  10 fold increase when compared to secretory-stage), and bioinformatic
prediction identifies miR-31 as a potential regulator of Tfrc in all vertebrate genomes. As a
result, we decided to use miR-31 for subsequent in vitro verification that Tfrc could be subjected
to miRNA regulation at some level in the mouse genome.

The expression levels of miR-153 and miR-31 in LS8 cells at different time points, following the
transfection of corresponding miRNA mimics or inhibitors, were first checked separately using
quantitative real-time PCR. Before LS8 cells were transfected by miR-153 mimics, the
expression of miR-153 could not be detected (Figure 7A). During the first 6h after transfection
with miR-153 mimics, there was a sharp increase in the intracellular level of miR-153, which
decreased continuously from 6h through 48h. At 48h, the level of miR-153 returned to the
original, non-detectable level. Because there was almost no endogenous miR-153 expression,
miR-153 inhibitors did not change the intracellular level of miR-153, which remained at zero
throughout the experiment (Figure 7A). By comparison, a higher level of intracellular miR-31

63
was detected in LS8 cells before transfection with miR-31 mimics, indicating that miR-31 was
intrinsically expressed in LS8 cells (Figure 8A). The changes in the expression of intracellular
miR-31 were relatively subtle following transfection with miR-31 mimics, while the introduction
of miR-31 inhibitors repressed the intracellular miR-31 almost immediately (within 10 min) after
transfection. The intracellular miR-31 level following transfection with the miR-31 inhibitors
remained undetectable for the 48h of observation.

In the luciferase reporter assay, exogenously introduced miR-153 mimics caused a decrease of
24.6% in the luciferase reporter activities (P=0.007) (Figure 7B), indicating a direct interaction
between the miR-153 mimics and the 3`-UTR of exogenously added Lamp1 (contained in
luciferase reporter vector). Introduction of miR-153 inhibitors into Lamp1 3’-UTR transfected
LS8 cells failed to cause statistically significant changes in luciferase reporter activities
(P=0.776) (Figure 7B). This is consistent with the array analyses, in which miR-153 was down-
regulated, and as a direct result Lamp1 was up-regulated, during enamel maturation.

Exogenously added miR-31 mimics showed a small inhibitory effect, however it was not
statistically significant (P=0.284) (Figure 8B). This may suggest that the initial levels of
endogenous miR-31 are so high that the targeting activity of exogenous miR-31 mimics on the
3’-UTR of Tfrc is minimal. When miR-31 inhibitors were added to Tfrc 3’-UTR LS8 cells there
was a 77.8% increase in luciferase reporter activities (P=0.008) (Figure 8B). Taken together
these results would suggest that the Tfrc 3’-UTR luciferase vector is functional, and that the
miR-31 mimics effectively target the Tfrc transcript. This data on miR-31 directed Tfrc transcript
down-regulation is also consistent with the gene expression microarray analyses, in which both

64
Tfrc and miR-31 were up-regulated during enamel maturation. That is, Tfrc is up-regulated by
approximately 30 fold during maturation (Additional File 6), which is an extreme change in
expression, and may require some “fine-tuning” and downward adjustments. The most
immediate response from cells to fine-tune this Tfrc transcriptional up-regulation may be to
activate processes that target Tfrc mRNAs for degradation, which may include gene up-
regulation of miR-31 transcription.

Clearly, relating cell culture-based experiments to in vivo-derived data poses problems for
interpretation, but these data sets for Lamp1 and miR-153 interactions, as well as for Tfrc and
miR-31 interactions, warrant further investigation.

65

Figure 7. (A) The expression levels of miR-153 in LS8 cells before and after transfection with
miR-153 mimics (blue) and inhibitors (red). (B) Relative luciferase reporter activities in
luciferase reporter assay validating the interaction of miR-153 with 3`-UTR of Lamp1.
Luciferase activities are presented in ratios of Firefly (F) to Renilla (R) luciferase reporter
activities. For the data presented, the amount of luciferase reporter vector used was 700ng, and
the concentrations of miR-153 mimics and inhibitors in final transfection complex were 20pM
and 0.2nM, respectively.

66

Figure 8. (A) The expression levels of miR-31 in LS8 cells before and after transfection with
miR-31 mimics (blue) and inhibitors (red). (B) Relative Luciferase reporter activities in
luciferase reporter assay validating the interaction of miR-31 with 3`-UTR of Tfrc. For the data
presented, the amount of luciferase reporter vector used was fixed at 700ng, and the
concentrations of miR-31 mimics and inhibitors in final transfection complex were 20pM and
0.2nM respectively.

67
2.5 Discussion

The involvement of miRNAs in tooth development was first suggested in 2008 [93]. Since then,
studies investigating miRNA expression profiles during amelogenesis have covered mainly the
early developmental stages of odontogenesis prior to the secretory and maturation stages of
amelogenesis [93, 98, 126]. The bioinformatics analyses from these earlier murine and porcine
studies have suggested that key roles of these stage- and/or tooth-specific miRNAs relate
primarily to cell differentiation, tooth morphogenesis and patterning [93, 98, 126]. In our study,
we conducted genome-wide miRNA expression profiling analysis in secretory- and maturation-
stage enamel organs obtained from rats, and identified 59 (out of 653) miRNAs that were
differentially expressed between secretory- and maturation-stage tooth development. This is
consistent with the dynamic expression pattern of miRNAs during tooth formation across
species, as noted in prior studies [93, 98, 126], indicating possible regulatory roles of miRNAs in
the later stages (enamel organ maturation) of enamel development.

Identifying target genes for miRNAs is essential to connect the stage-specific miRNA regulators
to biological functions. Computational prediction provides a tool to generate list of candidate
miRNA target genes [134-138]. However, the list of predicted genes for miRNAs can be
intimidating (herein, over ten thousand), making it impractical for further investigation. To refine
the number of predicted target genes for the 59 differentially expressed miRNAs, we conducted a
parallel alignment analysis between the list of predicted genes and the list of stage-specific genes
identified by genome-wide transcript expression profiling. The alignments generated three
workable lists of predicted genes for the secretory- and maturation-stage-specific miRNAs
(Additional Files 5-7). These were up-regulated miRNAs and their potential down-regulated

68
mRNA targets (39 and 299 respectively; Additional File 5); down-regulated miRNAs and their
potential up-regulated mRNA targets (20 and 330 respectively; Additional File 6); and up-
regulated miRNAs and their most highly up-regulated potential mRNA targets (15 and 41
respectively; Additional File 7). Recent studies demonstrated that a majority (> 80%) of
mammalian miRNAs affect their gene targets at the mRNA level by decreasing the stability of
target mRNAs [74] and the method of pairing inversed expression profiles of miRNAs and
mRNAs has been successfully used in previous miRNA expression profiling studies [139-146].
Nevertheless, we were still at risk of including false positive candidates and/or excluding valid
candidates from the final lists due to the intrinsic complexity of miRNA regulation [74, 147-
150]. Another potential problem with miRNA target prediction is that, because the genetic
information about miRNA target prediction is relatively sparse in rodents compared to humans,
all procedures and software involved in target prediction are based on homologous human
miRNAs that may also increase the false discovery and false negative rates.

The functions of ameloblasts during maturation-stage tooth development include matrix
turnover, calcium handling, pH regulation and ion transport [129]. GO analysis of the predicted
gene targets for differentially expressed miRNAs highlighted the functional categories that
overlap with all the key processes during enamel maturation. For example, among the 18 genes
significantly enriched in the category “endosome membrane” (Table 3), up-regulation of Cftr
and Lamp1 at both the mRNA and protein levels (maturation relative to secretory) have been
shown previously [10, 20, 129]. Two other examples listed in Table 3, Slc26a7 and Tfrc, have
also been confirmed as being up-regulated at both the mRNA and protein levels (maturation
relative to secretory) by our group (data not included). Lamp1 and Tfrc are considered to be the

69
potential ameloblast membrane-bound receptors for EMP debris [10, 14, 49, 151, 152], and are
involved with trafficking between the plasma membrane and endosomal/lysosomal structures
through associations with one or more adaptor protein complexes. CFTR functions as a regulator
of pH during rapid crystal growth and is critical for completion of enamel mineralization [19, 20,
22, 153]. CFTR is expressed most highly in maturation-stage ameloblasts; furthermore, in Cftr-
deficient animals that exhibit hypomineralised enamel, only the maturation-stage ameloblasts are
structurally affected [19, 20, 22, 153]. Apart from these relatively well-studied genes, Slc11a2
(listed in Table 3), which is involved in iron transport in epithelial tissues, is regulated by Cd61
in maturation-stage ameloblasts [154]. Another gene of note, listed in the category of “cytokine
activity” (Table 6), is Wnt5a. Wnt5a-induced cell death is crucial in determining the tooth size
during murine tooth development [155]. In addition, enriched in the category “carboxylic acid
transmembrane transporter activity” are 11 SLC gene family members (Table 1). The mRNA
expression levels of Slc1a1, Slc6a8, Slc25a15, Slc26a1 [129] and the other solute carrier (SLC)
gene family members (listed in Table 1) are significantly up-regulated in the enamel organ cells
during enamel maturation, and highlight the importance of SLC-mediated chloride-bicarbonate
exchange in enamel maturation. It is noteworthy that GO analysis of the miRNA-regulated genes
indicated that 27 genes up-regulated during maturation stage genes were significantly enriched in
the category of “calcium ion binding” (Table 5), while 11 genes down-regulated during
maturation stage were also enriched in the category of “calcium ion transmembrane transporter
activity” (Table 7). These data lend support to the complexity of genetic networks controlling
enamel maturation, or even the whole process of amelogenesis.

70
Experimental validation would be required for the identification of miRNA target genes in vivo
following computational prediction of miRNA and mRNA interaction. To date, there is still no
consensus about the working schemes of experimental validation of miRNA targets in vivo.
However, in addition to computational prediction, at least three criteria should be met before
confirming the miRNA regulator for a given gene [156]: 1) miRNA/mRNA coexpression; 2)
miRNA effect on target proteins; 3) miRNA effect on biological functions. While all of these
criteria might not be met under all conditions, nevertheless it is advisable that as many be
achieved as possible. In our efforts to validate the regulatory relations between Lamp1 and miR-
153, as well as between Tfrc and miR-31, we first demonstrated the coexpression of miRNAs
and mRNAs of target genes during maturation-stage development. In the case of Lamp1 and
miR-153 there was an inverse relationship, with higher Lamp1 mRNA levels and lower miR-153
levels noted in the maturation stage (Table 4). In the case of Tfrc and miR-31, the highest levels
of expression of both were noted in maturation-stage amelogenesis (Additional Files 1, 2 and 7).
The interaction between the seeding sequence of the miRNA (miR-153 and miR-31) and the 3`-
UTR of the target mRNA (Lamp1 and Tfrc) was predicted by TargetScan, and the binding site of
miR-31 to the mRNA of Tfrc was predicted to be highly conserved across vertebrates. The
effects of miR-153 and miR-31 on target proteins (Lamp1 and Tfrc respectively) was identified
indirectly by the changes in luciferase reporter assays induced by exogenously introduced
mimics and inhibitors of corresponding miRNAs into the host cells (Figures 7B and 8B). The
experiments seeking to confirm the effects of these miRNAs on biological functions is not
possible at this stage due to a lack of suitable animal models and/or appropriate organ culture
systems. As part of our future approach to studying the role of miRNAs in amelogenesis, gain-
of-function or loss-of-function studies of miRNAs should be considered in suitable primary

71
organ cultures or with the development of appropriate animal models. A caveat about
investigating the functional role of miRNAs is that there can be functional redundancy among
miRNAs. For example, miR-21 and miR-31 facilitate invasion and metastasis of colon
carcinoma cells by suppressing the same target TIAM1 in TGF-  signaling pathway[157]. In
human natural regulatory T cell FOXP3 expression is affected by both miR-21 and miR-31,
although the regulation of miR-21 is indirect[158]. Thus loss-of-function of a single miRNA,
especially the one of multi-gene family may not result in an aberrant phenotype[159].

Finally, it has been a widely accepted concept that gene expression regulation is controlled by
various factors at protein and RNA levels [160]. Although the key role of miRNA regulation has
been suggested in multiple biological processes and diseases, it cannot be the only contributor to
the intrinsic complexity of regulatory networks. In our study, 1729 genes (transcripts) showed
differential expression during maturation-stage tooth development. Among the 1729 genes, 299
up-regulated and 330 down-regulated genes were predicted to be regulated by the 59
differentially expressed miRNAs (Additional Files 5 and 6). On the other hand, the remaining
stably expressed miRNAs were predicted to be the regulators of 828 up-regulated and 513 down-
regulated genes (out of 1729). These data suggested that genetic regulators other than miRNAs,
such as transcription factors, must be dominant players in the processes of enamel maturation. In
the IPA core analysis of the miRNA-regulated candidate gene lists (299 up-regulated and 330
down-regulated), the potential transcription factors were predicted (Additional Files 12 and 13;
Upstream Regulators). It is highly possible that miRNAs and transcription factors co-regulated
the expression of these two groups of target genes, in forms of feed-forward loops (FFLs) and
feedback loops (FBLs) [160-163].

72
2.5 Conclusion

In conclusion, miRNAs are dynamically expressed when tooth development transitions from the
secretory stage to the maturation stage, and the differentially expressed miRNAs likely play a
role in the regulation of enamel maturation events by targeting genes involved in specific
activities such as pH regulation, ion transport, endocytosis and apoptosis. The data presented
here should help identify the roles of individual miRNAs in amelogenesis, and more generally
help to clarify the potential roles of miRNA-centered regulatory mechanisms in mineralized
tissues. Our data also suggest additional experiments to establish causal relationships between
miRNA and mRNA levels in genetically modified cell culture and animal models are warranted.
Recognition of miRNA-related regulation and the functions of corresponding target genes during
enamel development may also shed light on clinical diagnosis and/or treatment of diseases such
as amelogenesis imperfecta.

73
Chapter Three: SLC26A Gene Family Participate in pH Regulation During
Enamel Maturation

3.1 Abstract

The bicarbonate transport activities of Slc26a1, Slc26a6 and Slc26a7 are essential to
physiological processes in multiple organs. Although mutations of Slc26a1, Slc26a6 and Slc26a7
have not been linked to any human diseases, disruption of Slc26a1, Slc26a6 or Slc26a7
expression in animals causes severe dysregulation of acid-base balance and disorder of anion
homeostasis. Amelogenesis, especially the enamel formation during maturation-stage, requires
complex pH regulation mechanisms based on ion transport. The disruption of stage-specific ion
channels frequently results in enamel pathologies in animals. Here, using rat incisors, we
presented evidence that Slc26a1, Slc26a6 and Slc26a7 are highly expressed at maturation-stage
tooth development. In maturation-stage ameloblast, Slc26a1, Slc26a6 and Slc26a7 are co-
localized with the cystic fibrosis transmembrane conductance regulator (Cftr) to the apical region
of cytoplasmic membrane, and the distribution of Slc26a7 is also seen in the
cytoplasmic/subapical region, presumably on the lysosomal membrane. We have also examined
Slc26a1 and Slc26a7 null mice, and although no overt abnormal enamel phenotypes were
observed in Slc26a1
-/-
or Slc26a7
-/-
animals, absence of Slc26a1 or Slc26a7 appears to induce an
effective compensatory expression (gene up-regulation) of other related genes; including Cftr,
Ca2, Slc4a4, Slc4a9 and Slc26a9. Together, our data show that Slc26a1, Slc26a6 and Slc26a7
are novel participants in the extracellular transport of bicarbonate during enamel maturation, and
that their functional roles may be achieved by forming interaction units with Cftr.

74
3.2 Introduction

Enamel development involves two major functional stages-secretory and maturation [1]. In the
secretory-stage, ameloblasts synthesize and secrete a number of structural Enamel Matrix
Proteins (EMPs) [3, 4, 16]. The enamel matrix during secretory-stage is maintained at near-
neutral pH conditions in a protein-rich environment. During the maturation-stage, the
extracellular pH values vary considerable, ranging from neutral to acidic conditions, with a
return to physiologic pH levels at the end of maturation-stage [4, 12]. Changes in extracellular
pH values require sophisticated regulatory mechanisms from ameloblast, so as to maintain the
acid-base balance in the microenvironment suitable for crystal nucleation and growth [13]. In
addition, extracellular EMPs at maturation-stage are internalized by ameloblasts, and degraded
through endosome-lysosome pathways [10, 14]. Elaborate pH regulation machinery is also
required in the endocytotic process to create an acidic luminal environment for hydrolytic
enzyme activation [15]. Although there is still uncertainty about the working model of pH
regulation during amelogenesis [3, 13, 16-18], the involvement of carbonic anhydrases (CAs),
cystic fibrosis conductance transmembrane regulator (CFTR), Chloride Channel (CLCNs),
Solute Carrier family 4 (SLC4s) and Solute Carrier family 9 (SLC9s) in ameloblast-mediated pH
homeostasis has been widely accepted [7, 10, 16, 19-38].

SLC26 gene family encodes multifunctional anion exchangers and anion channels with a broad
range of substrates [164]. In mammalian, SLC26 gene family consists of 11 genes, SLC26A1-
SLC26A11. Based on our previous work of genome-wide miRNA and mRNA expression
profiling of the enamel organ cells (rat) [57], Slc26a1, Slc26a6 and Slc26a7 are the only
members among Slc26a gene family whose transcripts are significantly up-regulated during

75
maturation-stage enamel formation when compared to secretory-stage (Table 8). Slc26a1,
Slc26a6 and Slc26a7, which code for the proteins Sat1, Pat1 and Sut2 respectively, all exhibit
chloride/bicarbonate exchanger activities [108, 112, 117, 118]. Gene mutations in Slc26al,
Slc26a6 or Slc26a7 lead to multiple disorders, such as urolithiasis, hepatotoxicity, distal renal
tubular acidosis and impaired gastric secretion, induced by the disruption of ion homeostasis
[115, 116, 120, 165]. The process of enamel maturation involves pH regulation that is mediated
by multiple ion transport/exchange activities across plasma and endosome membrane [3, 7, 10,
14, 16, 19-38, 129, 130, 153, 166, 167]. Thus there is a need to obtain a better understanding
between the functional activities of SLC26A gene family members and amelogenesis.

In the present study, we first conducted quantitative real-time PCR and Western blot analyses to
show that Slc26a1, Slc26a6 and Slc26a7 are all significantly up-regulated at maturation-stage
compared with secretory-stage at both mRNA and protein levels. Based on immunolocalization
data, we show that in the maturation-stage ameloblasts the gene products of Slc26a1, Slc26a6
and Slc26a7 localize to the apical region of cytoplasmic membrane, similar to the localization
pattern of Cftr in maturation-stage ameloblasts. In addition, the distribution of Slc26a7 was also
seen within the cytoplasmic/subapical region of ameloblast, presumably on the lysosomal
membrane. From the protein complex pulled down using an antibody to Cftr; Slc26a1, Slc26a6
and Slc26a7 were each separately detected by the subsequent immunoblotting, suggesting the
direct interaction of each of these three proteins Slc26 proteins with Cftr. Compared with wild-
type (WT) animals, Slc26a1
-/-
or Slc26a7
-/-
animals did not show any significant abnormalities in
the mature enamel phenotype (density and structure). However, many gene transcripts examined
by real-time PCR – such as Car2 (carbonic anhydrase 2), Cftr, Slc4a4/NBCe1, Slc4a9/Ae4,

76
Slc26a9 and Alpl (alkaline phosphatase) – showed significant up-regulation in the enamel organ
cells of Slc26a1
-/-
and Slc26a7
-/-
animals when compared to age and sex-matched wild-type
controls. Collectively, these data indicated that Slc26a1, Slc26a6 and Slc26a7 are actively
involved in ion transport related to pH regulation processes during enamel maturation and their
functional roles may be achieved, at-least in part, by forming complex interactions with Cftr.

For many of the ion channels the assigned gene name is different to the assigned gene product.
For example Slc4a4, Slc26a1, Slc26a6 and Slc26a7 code for proteins AE2, Sat1, Pat1 and Sut2
respectively. To avoid confusion, in this paper we will refer to genes, and their respective gene
products (mRNA and protein), by their official gene ID rather than the product name.

77
3.3 Experimental procedures

Animals
All vertebrate animal studies complied with Institutional and Federal guidelines. For real-time
PCR, western blot and co-immunoprecipitation analyses, we obtained RNA and protein samples
from the enamel organs lining along the surface of rat (Wistar Hannover, 4-week, 100-110g)
incisors, because the reference line separating the secretory- and the maturation-stage enamel
organs have been well documented in rats [3, 129] (Figure 9A). The subsequent
immunohistochemistry and immunofluorescence detection of target gene products were also
conducted on sections of rat mandibles. Slc26a1
+/-
mice were purchased from the Jackson
Laboratory (stock # 012892) [165]. Slc26a1
-/-
mice were generated by breeding heterozygous
(Slc26a1
+/-
) parents. Slc26a7
+/-
mice were a kind gift from Dr. Manoocher Soleimani [120], and
bred in an identical manner to the Slc26a1 mutants. Although Slc26a7 null animals have been
documented to show distal renal tubular acidosis and impaired gastric acid secretion, they were
still reported to exhibited normal growth and survival compared to their WT littermates [120].
Both Slc26a1
-/-
and Slc26a7
-/-
mice are viable, fertile and normal in physical size compared with
their WT and heterozygous littermates. Slc26a1
-/-
and Slc26a7
-/-
animals were genotyped by PCR
using primers designed in earlier studies [120, 165].

78

Figure 9. Reference lines for animal dissection. A. In 4-week-old rat, the secretory- (S) and
maturation-stage (M) enamel organs along the enamel surface of mandibular incisor are
separated by the reference line between the 1
st
and the 2
nd
molar [3, 129]. B. Two reference lines
were used to partition secretory-stage (S) from maturation-stage (M) enamel organ. The first
reference line, vertical to the inferior border of mandibular cortical bone, divided the mesial-
distal width of the first molar into halves. The second reference line, also vertical to the inferior
to the bony border of the mandible, was located between the second and the third molar.

79
Rat tissue dissection, RNA extraction & real-time PCR analysis
For semi-quantifying the expression of Slc26a1, Slc26a6 and Slc26a7 mRNA, the secretory-
stage and maturation-stage RNA samples were obtained from the enamel organs of rat incisors.
Four 4-week old Wistar Hannover rats, each weighing 100 - 110g, were sacrificed for their
mandibles. The rat mandibles were kept in liquid nitrogen overnight and then lyophilized for
over 24 hours. After removing the cortical bone enclosing the incisors, the exposed multi-cellular
layers along the enamel surface were collected into RNase-free Eppendorf tubes. Details
regarding the dissection procedures were described previously [129, 130]. The total RNA was
extracted separately from secretory-stage and maturation-stage enamel organs using miRNeasy
Mini Kit (Qiagen, Valencia, CA, USA). The RNA sample from each of the four rats was also
processed separately. cDNA used for real-time PCR analysis was synthesized using miScript II
RT Kit with miScript HiFlex Buffer (Qiagen). In order to check the accuracy of dissection, the
expression of two-stage specific genes; Odam (highly expressed during maturation-stage) and
Enam (highly expressed during secretory-stage), were each subjected to real-time PCR analysis
to ensure accuracy in dissections [30, 129] before proceeding to examining Slc26a1, Slc26a6 and
Slc26a7, and other related gene transcript profiles. Real-time PCR reactions were performed on
CFX96 Touch
TM
Real-Time PCR Detection System (Bio-rad Life Sciences, Hercules, CA) with
iQ SYBR® Green supermix (Bio-rad Life Science) and rat-specific primers (Table 9). The raw
data acquired were in the form of Ct values, which were normalized to the Ct values of Actb ( -
actin). The  Ct method was used to calculate the fold changes in the expression of Slc26a1,
Slc26a6 and Slc26a7 (maturation-stage relative to secretory-stage) [168, 169]. Two-tail student t
test was used to detect the potential statistical differences in the expression levels of Slc26a1,

80
Slc26a6 and Slc26a7 transcripts between secretory- and maturation-stage (α=0.05). The
statistical package used was IBM SPSS Statistics 22.0 (IBM Corporation, Armonk, NY, USA).

Mouse tissue dissection, cDNA analysis and real-time PCR
The mandibles were isolated from 4-week Slc26a1
+/+
and Slc26a1
-/-
, and Slc26a7
+/+
and
Slc26a7
-/-
animals separately. The mandibles were then processed using the same procedures as
those used for rat mandibles [129, 130]. For RNA extraction from mouse mandibles, we used
different reference lines from that in rat mandibles (Figure 9). On the lingual side of the mouse
semi-mandible, the first reference line, vertical to the inferior border of mandibular cortical bone,
cut approximately half of the mesial-distal width of the first molar. The second reference line,
also vertical to the inferior to the bony border of the mandible, was located between the second
and the third molar (Figure 9B). We cropped the multi-cellular layers on the surface of incisor
enamel mesially of the first reference line as maturation-stage-derived enamel organ while the
tissues collected distally of the second reference line as secretory-stage-derived enamel organ
(Figure 9B). The dissected RNA samples from secretory- and maturation-stages were validated
by detecting the stage-specific expression of Odam and Enam (relative to Actb) using real-time
PCR. cDNA used for real-time PCR analysis was prepared using miScript II RT Kit with
miScript HiFlex Buffer (Qiagen). Real-time PCR reactions were performed on CFX96 Touch
TM
Real-Time PCR Detection System (Bio-rad Life Sciences) with iQ SYBR® Green supermix
(Bio-rad Life Science) and mouse-specific primers (Table 9). In order to verify that there were
no intact transcripts of Slc26a1 and Slc26a7 in the maturation-stage enamel organs of mutant
animals, PCR reactions were conducted using maturation-stage cDNA template and self-
designed primers. The sequences of the primers are: Slc26a1 Forward 5`-

81
cctggatattgcaaagccttcag-3`, Slc26a1 Reverse 5`-gaatcctgggaagggtcaaagc-3` (product 524 bp);
Slc26a7 Forward: 5`-cgggagcaaagaggaaaaag-3`, Slc26a7 Reverse: 5`-gtaagcaggaatgtggcactg-3`
(product 520 bp). The PCR reactions are set as follows: Slc26a1 94 ° C initial denaturation (10
min), 35 cycles-95 ° C (1 min), 59 ° C (1 min), 72 ° C (1 min), 72 ° C (8 min, final extension),
followed by a return to 4° C; Slc26a7 94 ° C initial denaturation (10 min), 35 cycles-95 ° C (1
min), 58 ° C (1 min), 72 ° C (1 min), 72 ° C (8 min, final extension), followed by a return to 4° C.
No PCR products with expected sizes were generated in mutant animals. In addition, real-time
PCR reactions were performed to detect the expression changes of the genes that are/might be
involved in maturation-stage regulation (Tables 9-11) [7, 10, 14, 19-38, 49, 129, 130, 153, 166-
172], using mouse-specific primers (Table 9) and cDNA samples from Slc26a1
+/+
and Slc26a1
-/-
,

and Slc26a7
+/+
and Slc26a7
-/-
animals separately. The procedures of data analysis for real-time
PCR are as described for rat tissue analysis. The statistical differences in the relative expression
level of each gene between the Slc26a1
+/+
and Slc26a1
-/-
,

and Slc26a7
+/+
and Slc26a7
-/-
animals
groupings were evaluated by two-tail student t test on IBM SPSS Statistics 22.0 ( =0.05).

Immunoperoxidase immunohistochemistry (IHC)
Wistar Hannover rats (100-110g body weight, 4 weeks old) were sacrificed for the mandibles.
The hemi-mandibles were then fixed in 4% paraformaldehyde (PFA) at 4 ° C overnight. 10%
EDTA (pH 7.4) was used to decalcify the samples for 10~12 weeks. 8μm sagittal sections were
prepared from paraffin-embedded samples. After the tissue sections were dewaxed and
rehydrated, endogenous peroxidase was blocked by 0.3% H
2
O
2
in methanol. Sections were
blocked by 1% bovine serum albumin (BSA) in PBS (1X, pH 7.4) and incubated overnight with
the primary antibodies against Slc26a1, Slc26a6 or Slc26a7 (antibody source is listed in Table

82
12). Tissues sections were counter-stained using Mayer`s hematoxylin after the 3-Amino-9-
ethylcarbazole (AEC)/ 3, 3'-diaminobenzidine (DAB) staining kit was applied (Table 12).
Negative controls were set up on the sagittal sections to which all the staining procedures were
applied but with no antibodies being added.

Co-localization analysis by Immunofluorescence (IF)
With the purpose of clarifying the localization of Slc26a1, Slc26a6 and Slc26a7 within the
milieu of maturation-stage ameloblast, we conducted immunofluorescence (IF) to co-localize
Slc26a1, Slc26a6 and Slc26a7 with other gene products – Ae2, Lamp1, and Cftr – whose
localization in ameloblast have been previously reported [10, 14, 20-22, 25, 30, 32, 49, 153]. The
protocols of preparing tissue sections for IF were the same as those used for immunoperoxidase
IHC. BSA-blocked tissue sections were incubated overnight with different combinations of
primary antibodies: Slc26a1 x Ae2; Slc26a1 x Lamp1; Slc26a6 x Ae2; Slc26a6 x Lamp1;
Slc26a7 x Ae2; Slc26a7 x Lamp1; and Slc26a7 x Cftr (Table 12). The co-localization analysis
was not conducted for the combinations of Slc26a1 x Cftr, or Slc26a6 x Cftr, as the antibodies to
Slc26a1, Slc26a6 and Cftr were all goat-derived and not suitable (Table 12) and no commercially
available non-goat derived Cftr antibodies suitable for rodent tissues could be identified. All
tissue sections used for co-localization analyses were stained by DAPI (Vector Laboratories;
Catalog # H-1200) the cover slips were added.

Western blot analysis
The protein samples were obtained from Wistar Hannover rats (100-110g body weight, 4 weeks
old). After the animals were euthanized and decapitalized, the mandibles were dissected and

83
isolated immediately. We used the reference line between the first and the second molar to
discriminate secretory-stage from maturation-stage enamel organ [3, 129] (Figure 9A). The bony
structures enclosing the incisor were removed and the enamel organs along the enamel surface
were collected for secretory-stage and maturation-stage separately. The samples were added to
pre-cooled Eppendorf tubes containing RIPA Lysis and Extraction buffer (Thermo Fisher
Scientific Inc., Rockford, IL, USA; Catalog # 89901) mixed with Halt Protease Inhibitors
Cocktail (Thermo Fisher Scientific; Catalog # 78429). The samples were homogenized with a
pestle (on ice), kept on ice for 30min and centrifuged at 16,000rpm for 15min while being
maintained at 4 ° C. The supernatant was collected and quantified using BCA Protein Assay Kit
(Thermo Fisher Scientific; Catalog # 23225). Protein extracts were also obtained from rat
kidneys to serve as controls. The samples were mixed with LDS Sample Loading Buffer
(Thermo Fisher Scientific; Catalog # 84788), heated for 10 min at 95 ° C, and loaded on mini-
gels (15μg each well) (Thermo Fisher Scientific; NuPAGE® Novex® 10% Bis-Tris Protein
Gels, Catalog # NP0315BOX). Electrophoresis was carried out at 120V for 2-2.5 h, and gels
were electrotransferred to nitrocellulose membrane at constant current of 0.1A for 2h. The blots
were then blocked by 5% non-fat milk powder in Tris Buffered Saline (TBS) for 1 h at room
temperature, and incubated overnight with primary antibodies to Slc26a1, Slc26a6, Slc26a7 and
-actin (internal control), respectively (Table 12). After appropriate HRP-conjugated secondary
antibody was applied for 2 h at room temperature, the blots were washed and developed with
substrate kits (Thermo Fisher Scientific; SuperSignal West Pico Chemiluminescent Substrate,
Catalog # 34077; SuperSignal West Femto Maximum Sensitivity Substrate, Catalog # 34095).
Quantification of the relative intensity of the bands was conducted using ImageJ 1.48.

84
Co-immunoprecipitation (Co-IP)
Maturation-stage enamel organ protein samples were obtained using procedures described above,
and were then lyzed in 1xTBS with 1% Triton-100. After being pre-cleared by 20ul A/G agarose
beads at 4 ° C for 1 h, the protein samples (50 -100µ g) were incubated with 1µ g of primary
antibody to Cftr (Table 12) at 4 ° C for 1 h, while being gently agitated. The sample was then
mixed with 20µ l A/G agarose beads and kept at 4 ° C overnight. The beads were washed
extensively with Tris buffer (50mM Tris, 0.1% NP-40, pH 7.4), and the bound proteins were
subject to Western blot analyses with the primary antibodies against Slc26a1, Slc26a6 and
Slc26a7 (separately), and their appropriate secondary antibody (Table 12). Protocols for Western
blot analysis have been described. Positive and negative controls were set up simultaneously
with the experimental group. The positive control was the pre-cleared total protein without
subsequent immunoprecipitation procedures, while the negative control was prepared by
skipping the step of applying the primary antibody of Cftr.

Micro-CT analysis
Mandibles were isolated from 8-week-old Slc26a1
-/-
and Slc26a7
-/-
mice and their respective age-
matched WT littermates. The number of mice sacrificed for each groups was 4 (n=4). The hemi-
mandibles were air-dried for at least 7 days before micro-CT (µ CT) analyses was conducted. The
samples were scanned with Siemens MicroCAT II at the Molecular Imaging Center (MIC) of
University of Southern California. The acquisition settings were documented as previously
described [36]. The reconstruction and the subsequent calculation of the relative density of fully
mature enamel and dentin were performed with Amira 3D Visualization and Analysis Software
5.4.3 (FEI Visualization Science Group, Burlington, MA, USA). Two-tail student t test was used

85
to evaluate the potential statistical differences in the relative density and the thickness of enamel
between Slc26a1
-/-
and their WT littermate controls, and Slc26a7
-/-
and their WT littermate
controls on IBM SPSS Statistics 22.0 ( =0.05).

Scanning Electron Microscopy (SEM) ，Energy-dispersive X-ray Spectroscopy (EDS) and
microindentation
The hemi-mandibles used for Scanning Electron Microscopy (SEM) and microindentation
analyses were extracted from 8-week-old Slc26a1
-/-
and Slc26a7
-/-
mice and their respective age-
matched WT littermates (n=6). The samples were prepared and scanned as previously
documented [29, 36]. EDS was conducted on JEOL JSM-7001F to analyze the elemental
composition of Slc26a1
-/-
and Slc26a7
-/-
enamel, with an accelerating voltage of 10 kV. The
differences in Atomic percent (At%) of elements Ca, P, O, C, Cl, Na and Mg between mutant
(Slc26a1
-/-
and Slc26a7
-/-
) and wild-type enamel were detected by two-tail student T test on IBM
SPSS Statistics 22.0 (n=6, =0.05).

86
3.4 Results

Up-regulation of Slc26a1, Slc26a6 and Slc26a7 during enamel maturation
Based on our previous work of genome-wide miRNA and mRNA transcriptome analyses using
RNA samples extracted from the secretory- and maturation-stage enamel organs of 4-week-old
rat mandibular incisors, we showed that among all the Slc26 gene family members, Slc26a1,
Slc26a6 and Slc26a7 are the only three that are differentially expressed between the two
developmental stages (Table 8). In this current study we conducted real-time PCR analysis using
the same RNA samples used previously for genome-wide mRNA transcriptome analysis [57].
We verified that Slc26a1, Slc26a6 and Slc26a7 were all significantly up-regulated at mRNA
level during maturation-stage tooth development (relative to secretory-stage) (P<0.05; Figure
10), which is consistent with the results obtained from prior microarray analysis [57] (Table 8).
The average fold changes for Slc26a1, Slc26a6 and Slc26a7 were ~ 15.4, ~ 3.9 and ~ 8.1,
respectively (Figure 10). In order to assess Slc26a1, Slc26a6 and Slc26a7 protein expression
levels, we performed Western blot analysis using protein samples obtained from secretory- and
maturation-stage enamel organs of rat mandibular incisors, as well as from rat kidney (as a
reference control). At the protein level Slc26a1, Slc26a6 and Slc26a7 all exhibited higher
expression during maturation-stage when compared to secretory-stage (Figure 11). The average
fold changes in the protein-level expression of Slc26a1, Slc26a6 and Slc26a7 were ~ 1.5, ~ 3.5
and ~ 3.1 respectively (Figure 11), which were not as great as those calculated at mRNA levels
(Figure 10), but this was a consistent directional change (mRNA and protein) for all three genes.
In addition, Slc26a1 and Slc26a6 showed higher abundance at protein level in maturation-stage
enamel organ than in kidney, while the trend was the opposite for Slc26a7 with respect to its
relative abundance (Figure 10).

87
Table 8. Average fold changes of Slc26s during maturation-stage relative to secretory-stage
based on genome-wide mRNA transcriptome analysis. (N/A expression not detected; α=0.05)

Gene symbol Fold changes P values
Slc26a1 39.8 0.000031
Slc26a2 -1.2 0.368
Slc26a3 1.2 0.391
Slc26a4 1.0 0.825
Slc26a5 1.1 0.711
Slc26a6 5.4 0.00016
Slc26a7 7.8 0.00013
Slc26a8 1.1 0.8
Slc26a9 -1.1 0.263
Slc26a10 N/A N/A
Slc26a11 -1.5 0.278

88

Figure 10. Real-time PCR analysis of Slc26a1, Slc26a6 and Slc26a7 expression during
amelogenesis. (A) Expression levels of Slc26a1, Slc26a6 and Slc26a7 relative to Actb in the
maturation-stage amelogenesis. (B) Fold changes during maturation stage compared with
secretory stage (M/S). The expression of Slc26a1, Slc26a6 and Slc26a7 were up-regulated by
~15.4, ~3.9 and ~8.1 fold at maturation-stage relative to secretory-stage.

89

Figure 11. Western blot analysis of Slc26a1, Slc26a6 and Slc26a7. A1-C1. Protein-level
expression of Slc26a1, Slc26a6 and Slc26a7 were detected by western blot analysis using
samples obtained from both secretory and maturation-stage enamel organs (4-week-old rat
incisors). Protein samples extracted from kidney (4-week-old rat) were used as reference. The
molecular weights for Slc26a1, Slc26a6 and Slc26a7 are 75kDa, 90kDa and 72kDa, respectively.
Beta-Actin served as the control for sample loading. A2-C2. The intensities of the bands (relative
to Beta-Actin) were measured by ImageJ. The average fold changes of Slc26a1, Slc26a6 and
Slc26a7 at protein level were ~1.5, ~3.5 and ~3.1, which were not as great as those calculated at
mRNA level.

90
Localization of Slc26a1, Slc26a6 and Slc26a7 in enamel organ
Using sagittal sections prepared from 4-week-old rat mandibles, immunoperoxidase
immunostaining was performed to clarify the expression patterns of Slc26a1, Slc26a6 and
Slc26a7 in both secretory- and maturation-stage enamel organs. At secretory-stage, Slc26a1 was
mainly localized to the basal membrane of ameloblasts (Figure 12A). The localization of Slc26a6
within secretory-stage enamel organ seemed to be more diverse than Slc26a1 where it is
expressed at both the basal membrane and apical membranes of ameloblasts (Figure 12D). The
expression of Slc26a7 was barely detected by IHC in secretory-stage enamel organ (Figure 12G).
In maturation-stage enamel organ cells Slc26a1, Slc26a6 and Slc26a7 were all expressed in both
the ameloblasts and in the papillary layer (Figures 12B, 12C, 12E, 12F, 12H and 12I), but the
relative expression patterns varied. For Slc26a1, higher expression was observed in papillary
layer than in ameloblast (Figures 12B and 12C). In maturation-stage ameloblast, Slc26a1 was
localized to the apical and/or subapical domains of cytoplasmic membrane (Figures 12B and
12C). The relative distribution of Slc26a6 in maturation-stage enamel organ cells differed from
Slc26a1, with higher expression of Slc26a6 observed in the ameloblasts rather than in the
papillary layer (Figure 12E and 12F). The localization of Slc26a6 in ameloblasts also varied
between smooth-ended ameloblasts (SA) and ruffle-ended ameloblasts (RA) with a greater apical
concentration of Slc26a6 seen in SA (Figure 12E compared to 12F). Slc26a7 is expressed both in
the ameloblasts and papillary layer (Figure 12H and 12I) with a greater apical concentration is
seen in SA when compared to RA (Figure 12H compared to 12I).

91

92
Figure 12. Immunoperoxidase immunostaining of Slc26a1, Slc26a6 and Slc26a7 in secretory-
and maturation-stage enamel organ. Immunostaining procedures were applied to the sagittal
sections prepared from paraffin-embedded 4-week-old rat mandibles. A. Slc26a1 in secretory-
stage ameloblasts (S); B. Slc26a1 in smooth-ended ameloblasts at maturation-stage (M-SA); C.
Slc26a1 in ruffle-ended ameloblasts at maturation-stage (M-RA); D. Slc26a6 in secretory-stage
ameloblasts (S); E. Slc26a6 in smooth-ended ameloblasts at maturation-stage (M-SA); F.
Slc26a6 in ruffle-ended ameloblasts at maturation-stage (M-RA); G. Slc26a7 in secretory-stage
ameloblasts (S); H. Slc26a7 in smooth-ended ameloblasts at maturation-stage (M-SA); I.
Slc26a7 in ruffle-ended ameloblasts at maturation-stage (M-RA); J-L. The sections that were
incubated without antibodies being applied served as negative controls for immunostaining. All
images were collected under x20 magnification.

93
With the purpose of acquiring deeper insight into the localization of Slc26a1, Slc26a6 and
Slc26a7 in maturation-stage ameloblast, we conducted dual immunofluorescence (IF) to
establish the spatial localization each of these three genes with Slc4a2/Ae2, Lamp1 and Cftr. For
Ae2, Lamp1 and Cftr their expression patterns during enamel maturation have been previously
studied [14, 16, 20, 21, 25, 32]. Based on the IF analysis, we showed that Slc26a1 and Slc26a6
exhibited similar distribution patterns during maturation-stage, where they are both localized to
the apical/subapical region of ameloblast (Figures 13A-D). Slc26a7 expression is localized to the
apical/subapical membrane and the membrane of cytoplasmic vesicles of maturation-stage
ameloblast (Figure 13E, 13F and 13G). These data for Slc26a1, Slc26a6 and Slc26a7 are
consistent with the observations seen in IHC (Figure 12). In ameloblasts the spatial localization
of Ae2 is at the lateral membrane (Figure 13A, 13C and 13G), and Lamp1 is localized to the
peri-nuclear region within the cytoplasm (Figures 13B, 13D and 13F). Cftr is localized to the
apical membrane of maturation-stage ameloblast (Figure 13G) [20], thus Cftr, Slc26a1 and
Slc26a6 show significant colocalization patterns in maturation-stage ameloblasts.

Slc26a1, Slc26a6 and Slc26a7 interact with Cftr
As we have shown, Slc26a1, Slc26a6, Slc26a7 and Cftr are all localized to the apical membrane
of maturation-stage ameloblast, although Slc26a7 also has an expression profile including
subapical and cytoplasmic region (Figures 12 and 13). To investigate the hypothesis that the
similarity in the expression patterns of Slc26a1, Slc26a6 and Slc26a7 to Cftr on the apical
membrane of ameloblast during enamel maturation may involve a direct physical interaction, we
conducted co-immunoprecipitation (Co-IP) assays using protein samples obtained from
maturation-stage enamel organ. We confirmed that from the protein complex pulled down by

94
anti-Cftr antibody, Slc26a1, Slc26a6 and Slc26a7 are all able to be detected by the corresponding
antibodies separately (Figure 14), indicating that there exist protein-protein interactions between
Cftr and each of the three SLC26 genes studied. Positive and negative controls were included in
the co-IP assays (Figure 14).

95

96
Figure 13. Co-localization analysis of Slc26a1, Slc26a6, Slc26a7 with Ae2, Lamp1 and Cftr. (A-
B) Co-localization of Slc26a1 with Ae2 and Lamp1 in maturation-stage ameloblasts. The signals
of Slc26a1 were mainly seen on the apical membrane of maturation-stage ameloblasts (Panels A
and B; Green). By contrast, Ae2 was localized to the basolateral membrane (Panel A; Red) while
Lamp1 showed cytoplasmic and/or peri-nuclear distribution pattern in ameloblasts (Panel B;
Red). The sections were stained by DAPI to highlight the nucleus (Panels A and B; Blue). The
images were collected under confocal microscopy (x63 magnification). (C-D) Co-localization of
Slc26a6 with Ae2 and Lamp1 in maturation-stage ameloblasts. Slc26a6 exhibited a similar
expression pattern to that of Slc26a1—on the apical membrane of maturation-stage ameloblasts
(Panels C and D; Green). The fluorescence signals of Ae2 (Panel C; Red) and Lamp1 (Panel D;
Red) were used as references. The sections were stained by DAPI to highlight the nucleus
(Panels C and D; Blue). The images were collected under confocal microscopy (x63
magnification). (E-G) Co-localization of Slc26a7 with Ae2, Lamp1 and Cftr in maturation-stage
ameloblasts. The expression of Slc26a7 was found both on the apical membrane and within the
cytoplasmic region (Panels E-G; Green). There were partial overlaps in fluorescence signals of
Slc26a7 with those of Lamp1 (Panel F; Red) and Cftr (Panel G; Red), rather than Ae2 (Panel E;
Red). The sections were stained by DAPI to highlight the nucleus (Panels E-G; Blue). The
images were collected under confocal microscopy (x63 magnification).

97

Figure 14. Co-immunoprecipitation (Co-IP) assay of Cftr with Slc26a1, Slc26a6 and Slc26a7.
Co-IP was conducted using protein samples extracted from the maturation-stage enamel organ of
4-week-old rat incisors (50~100 μg initial input). The interaction complexes were pulled down
by anti-Cftr antibody (Table 12). The subsequent western blot analyses were performed using
primary antibodies to Slc26a1, Slc26a6 and Slc26a7, respectively (Table 12). The positive
control was the pre-cleared total protein without the following immunoprecipitation procedures,
and the negative control was prepared by skipping the step of applying the antibody of Cftr.

98
Enamel phenotypes of Slc26a1 null and Slc26a7 null animals
All teeth prepared for micro-CT and SEM scanning were dissected from 8-week-old animals
(Slc26a1
-/-
, Slc26a7
-/-
and their wild-type littermates). Generally, there were no remarkable
differences between the mutant (Slc26a1
-/-
or Slc26a7
-/-
) and the wild-type teeth in the aspects of
gross anatomy, microstructures and hardness of enamel (Figures 15 and 16, Table 13
microindentation). We used Amira 3D Visualization and Analysis Software 5.4.3 to delineate
enamel in scanned teeth and calculate its relative density. Moreover, we measured the thickness
of enamel of incisors from 3D-reconstructed teeth at the point where the enclosing cortical bone
of incisors terminates. These parameters obtained from Slc26a1
-/-
and Slc26a7
-/-
animals were
compared separately with those from wild-types. However, no differences with statistical
significance were detected (P>0.05) (Figure 15). Nevertheless, the elemental composition of
mutant enamel showed significant changes when compared with wild-type enamel. The Atomic
percentages (At%) of Cl increased by ~34% in both Slc26a1
-/-
(P=0.012) and Slc26a7
-/-

(P=0.035) enamel (Figure 17A-C, E). There was a decrease of ~24% in the At% of C in
Slc26a1
-/-
enamel (P=0.028), and the difference in the At% of C between Slc26a7
-/-
and wild-type
enamel was marginally significant (P=0.078) (Figure 17E). In addition, the At% of Na in
Slc26a7
-/-
enamel also decreased significantly (P=0.028) (Figure 17E).

99

100
Figure 15. Micro-CT analysis of Slc26a1
-/-
and Slc26a7
-/-
mandibles. The mandibles from wild-
type, Slc26a1 null and Slc26a7 null animals (8-week) were subject to micro-CT analysis (n=3).
The relative density and thickness of enamel on the labial incisor, where the cortical bone
enclosing just begins (A1-C2), were measured respectively. There is no statistical difference
between mutant and wild-type animals with respect to these two parameters (D-E).

101

Figure 16. SEM images of mature enamel in Slc26a1
-/-
and Slc26a7
-/-
animals. The surface
properties of enamel (incisor and molars) in Slc26a1
-/-
and Slc26a7
-/-
animals were similar to
those observed in wild-type animals (A1-B3, magnification x35). When the internal structures of
enamel (incisor) were observed in coronal section (C1-C3, magnification x5000), the mature
enamel from mutant animals showed mild disruption in rod density and diameter (C2-C3)
compared with wild-type enamel (C1). The arrangement of the enamel rod and inter-rod
structures (incisor) did not seem to be impacted by the deletion of Slc26a1 or Slc26a7 (C1-C3).

102

Figure 17. EDS analysis of mature enamel Slc26a1
-/-
and Slc26a7
-/-
animals. (A-C) EDS
spectrum of mature enamel in wild-type, Slc26a1
-/-
and Slc26a7
-/-
animals, respectively (n=6). (D)
No statistically significant differences were detected in the At% of Ca, P and O between mutant
and wild-type enamel. (E) The At% of Cl increased significantly by ~34% in both Slc26a1
-/-

(P=0.012) and Slc26a7
-/-
(P=0.035) enamel. There was a significant decrease of the At% of C in
Slc26a1
-/-
enamel (P=0.028). Similar difference in the At% of C was detected between Slc26a7
-/-

103
and wild-type enamel, but was only marginally significant (P=0.078). The At% of Na in
Slc26a7
-/-
enamel also significantly decreased (P=0.028).

104
Compensatory gene expression in Slc26a1 null and Slc26a7 null animals
Since we observed no overt abnormalities in mature enamel from either Slc26a1
-/-
or Slc26a7
-/-
animals, we sought to investigate the changes in the expression profiles of genes that have a
similar biological function to the anion exchangers Slc26a1 and Slc26a7 during maturation-stage
enamel formation in mutant animals (complete list of interrogated genes provided in (Tables 10-
11). Total RNA samples used for real-time PCR analysis were extracted from the maturation-
stage enamel organ of 4-week-old mutants (Slc26a1
-/-
and Slc26a7
-/-
) and their age-matched WT
littermates. Among the 41 genes examined, 22 genes showed differential expression (18 up-
regulated and 4 down-regulated) in Slc26a1
-/-
and/or Slc26a7
-/-
animals compared with WT
(P<0.05; Tables 10-11, Figures 18 and 19). Many of the up-regulated genes have been well
characterized as being involved in either pH regulation (i.e. Cftr, Car2, Ae2 and NBCe1) or
endocytosis (i.e. the lysosomal-associated membrane proteins Lamp1, Lamp2, Cd63 and Cd68,
and others such as Clcn7, Rab21 and Ctss) during enamel maturation (Tables 10-11, Figure 18)
[7, 10, 14, 19-38, 49, 129, 130, 153, 166-172]. It is noteworthy that Cftr, Slc4a9/Ae4, Slc26a5
and Slc26a9 exhibited the largest scales of fold changes (>5) among the up-regulated genes in
both mutant animal groups (Tables 10-11, Figure 18). This data is strong evidence that enamel
organ cells initiate a compensatory gene expression up-regulation of anion exchangers following
the deletion of either Slc26a1 or Slc26a7. Finally, the expression of some of the SLC4 and Slc26
gene family members were negligible in the enamel organs of wild-type and mutant mice strains,
and these include Slc4a5/NBCe2, Slc26a3, Slc26a8 and Slc26a10 (Tables 10-11).

105

106
Figure 18. Up-regulated genes in Slc26a1-/- and Slc26a7-/- animals compared with wild types.
Most genes that showed significant changes in expression were up-regulated (Panels A-S),
indicating a compensatory effect induced by the deletion of Slc26a1 or Slc26a7. The expression
values of differentially expressed genes were normalized to those of Beta-Actin. For all the two-
tail t tests used, the significance level was 0.05. * <0.05; ** <0.01.

107

Figure 19. Downregulated genes in Slc26a1-/- and Slc26a7-/- animals compared with wild
types. The expression values of differentially expressed genes were normalized to those of Beta-
Actin. For all the two-tail t tests used, the significance level was 0.05. * <0.05; ** <0.01.

108
3.5 Discussion

During maturation-stage tooth development, ameloblasts function to remove organic protein
debris from extracellular enamel matrix [10, 11] and to deposit inorganic ions into the enamel
area [3], so that the crystal structures in enamel achieve their terminal width and thickness,
rendering the dental enamel fully mature and functional. In the extracellular matrix, the mineral
deposition, crystal growth and protease activities are presumed to be highly pH-dependent [3,
173], while in lysosomal lumen, acidic pH is necessary for the activation of hydrolytic enzymes
and the degradation of internalized macromolecules [10, 15, 174]. Thus, these two
interconnected key processes at maturation-stage require the tight control of pH values to
maintain either luminal or extracellular acid-base balance, which is facilitated by the diffusion of
ions across various biological membranes [16, 25]. Although the detailed mechanisms of
maturation-stage pH regulation is yet to be elucidated, previous investigations have suggested
the essential role of several genes, such as CFTR, AE2, NBCe1, CA2, CA6, NHE1 and CLCN7,
in enamel maturation [7, 10, 16, 19-22, 25, 27, 29, 31, 32, 35-38, 129, 153, 166, 175]. In this
study, we provided experimental evidences to show that SLC26A1/SAT1, SLC26A6/PAT1 and
SLC26A7/SUT2 are novel candidate genes that are involved in the pH regulation process during
maturation-stage tooth development.

The cellular localization patterns of genes that participate in the pH regulation process during
enamel maturation have been shown to be quite diverse, which can be direct reflections of their
functional roles. For example, Cftr serves as a chloride channel mainly on the apical plasma
membrane of the maturation-stage ameloblast, and thus is conceived as a regulator of pH during
rapid crystal growth [20] (Figures 13 and 20); NBCe1 and Ae2 localize to the basolateral

109
membrane of ameloblast at maturation-stage, and the bicarbonate transport activity mediated by
NBCe1 and Ae2 are thought to be critical to intracellular pH homeostasis [21, 27, 32] (Figures
13 and 20); the localization of Clcn7 is identified to the intracellular organelles, presumably
lysosomal membrane, suggesting that Clcn7 might help to accumulate protons within lysosomal
lumen and form part of endocytotic apparatus in maturation-stage ameloblast [10] (Figure 20).
According to our results from immunoperoxidase immunostaining and immunofluorescence, the
distributions of Slc26a1, Slc26a6 and Slc26a7 are similar—they all localize to the apical
membrane of maturation-stage ameloblast, and in case of Slc26a7 subapical and cytoplasmic
region (Figures 12-13). This is consistent with a previous discovery that Slc26a4, another
member in Slc26 gene family, showed positive staining on the same membrane domain of
ameloblasts during enamel maturation [176]. These data indicate that the anion exchanger
activities of Slc26a1, Slc26a4, Slc26a6 and Slc26a7 on the apical membrane of maturation-stage
ameloblasts may respond to the intracellular and/or extracellular pH changes, and regulate pH
values by secreting bicarbonate into enamel matrix to neutralize protons (Figure 22). In addition
to the apical membrane, the distribution of Slc26a7 is also seen in the cytoplasmic region within
maturation-stage ameloblast (Figures 13E-13G). It is highly possible that the intracellular
localization of Slc26a7 is to the membranous structures in early and/or late endo-lysosomal
pathway (Figure 20). This is mainly evidenced by the observations that there are partial overlaps
in fluorescence signals of Lamp1 (late endo-lysosome-localized [14, 177]) and Slc26a7 from co-
localization assay and that the overlaps reside in the subapical and the middle portion instead of
the peri-nuclear region of the cytosol (Figure 13F). In other functional cell types, the localization
of Slc26a7 also demonstrates such diversity as observed in maturation-stage ameloblast. For
example, in kidney, Slc26a7 co-localizes with AE1 to the basolateral or subapical membrane,

110
and with Tfrc to the endosomal membrane of A-intercalated cells in renal outer medullary
collecting duct (OMCD) [118, 119]. The distinct distribution pattern of Slc26a7 in maturation-
stage ameloblast lends support to its potential role in regulating pH within lysosomal lumen,
which might be similar to that of Clcn7, in addition to the pH regulation presumed to be
functioning in extracellular enamel matrix.

The phenomenon that several ion transporters/channels with similar physiological functions and
cellular localizations interact with one another to form united protein complexes has been
reported in multiple areas of biomedical research. In most cases when Cftr interacts with Slc26s,
Cftr seems to serve as a hub for these potential interaction complexes [178-184]. One example is
that Slc26a3, Slc26a6 and Slc9a3r1 co-localize with Cftr to the midpiece of mouse sperm, and
the protein complex formed by Cftr with Slc26a3, Slc26a6 and Slc9a3r1 functions primarily to
mediate transmembrane transport of chloride, which is critical for sperm capacitation [179]. In
cochlear outer hair cells (OHCs), the physical interaction between Cftr and Slc26a5, which is
localized to the lateral membrane of OHCs, has potential electrophysiological significance [182].
Additionally, in human bronchial cell lines, functional CFTR contributes to the functions of
SLC26A9 as an anion conductance [178]. Based on our data from Co-IP, we presented that
physical interactions exist between Cftr and Slc26a1, Slc26a6, Slc26a7 in maturation-stage
ameloblast, and the functional complexes may be localized to the apical membrane where the
expression of these genes are identified (Figures 13-14, 20). At this stage of work, we did not
seek to investigate the potential interactions between Cftr and other important pH regulators,
such as Ae2, NBCe1 and Clcn7, mainly because of their distinct locations within the milieu of
ameloblast (Figure 20). However, it is equally unreasonable to rule out the possibility that

111
physical interactions involving Cftr and other related genes do exist, as these pH regulators may
exhibit mobility during ameloblast modulation cycles [7]. Therefore, it could be speculated that
Cftr might interact with a broader range of pH regulators, and that the pH regulation process
during enamel maturation might be achieved by the coordination of functional protein complexes
that are far more sophisticated than expected.

Mutations in the pH regulators that have been identified to be functional during enamel
maturation, such as Cftr, Ae2 and NBCe1, could often result in severe AI-like enamel/tooth
phenotypes [16, 19-22, 25, 27, 29, 32, 37, 38]. In our study, deletion of Slc26a1 and Slc26a7 lead
to an increase of Cl and a decrease of C in the elemental composition of enamel matrix (Figure
17E), which is consistent with our hypothesis about the functional role of Slc26a1 and Slc26a7
as anion exchangers of Cl
-
(intracellular-oriented)

and HCO
3
-
(extracellular-oriented) (Figure 20).

However, no significant abnormalities in mature enamel structures and hardness were observed
in Slc26a1 or Slc26a7 mutant animals (Figures 15 and 16), although impaired physiological
functions were documented in other organs [120, 165], suggesting that there might be functional
redundancy of Slc26a1 and Slc26a7 during amelogenesis. To investigate this hypothesis, we
conducted real-time PCR reactions to detect the potential changes of gene expression in mutant
animals. While a small number of genes were selected for qPCR analysis, selection was made
based on previous studies showing that these genes or other members within the same gene
family are involved in ion transport, pH regulation and endocytotic pathways during maturation-
stage amelogenesis [7, 10, 16, 19-25, 27-30, 32, 33, 35, 37, 38, 166, 167, 173]. The results
showed that deletion of Slc26a1 or Slc26a7 induced upregulation of multiple genes with similar
functional roles in maturation-stage pH regulation and endocytotic pathway (Tables 10-11,

112
Figure 18). The up-regulated genes can be roughly subdivided into three categories: 1) those that
have been relatively well characterized, such as Car2, Cftr, Ae2, NBCe1, Slc26a6, Clcn7,
Lamp1-Lamp4, Rab21 and Ctss [7, 10, 16, 19-25, 27-30, 32, 33, 35, 37, 38, 166, 167, 173]; 2)
those that are not expressed/differentially expressed in normal enamel maturation, such as
Slc26a2, Slc26a5, Slc26a9, Ae4 [129]; 3) other related genes, such as Alpl and Enam. Based on
the features of these up-regulated genes, it is reasonable to conclude that there were strong
compensatory reactions in response to the deletion of Slc26a1 or Slc26a7 in mutant animals.
Absence of enamel phenotypes is hardly a novel discovery when the genes involved in
maturation-stage pH regulation and endocytotic pathway are deleted, and can often be explained
by the compensatory effect from other genes [10, 176, 185]. A final point to be noted is that
compared with wild-type animals, the expression of Slc26a7 was downregulated by ~5 fold in
Slc26a1 null animals (Tables 10-11, Figure 19). Previous investigations on Car2 null animals
demonstrated that Car2 deficiency decreases the expression of Slc26a4, Slc26a7 and Ae1 at
mRNA level in kidney collecting duct [186]. The authors proposed that changes in the
expression patterns of Slc26a4, Slc26a7 and Ae1 might be attributed to increased level of
apoptosis, which could result from disturbance of pH homeostasis induced by Car2 deletion
[186]. Similar explanations may or may not apply in case of decreased Slc26a7 transcripts in
Slc26a1 null animals, and the interplay between Slc26a1 and Slc26a7 during enamel maturation
warrants further investigation.

113

Figure 20. Schematic diagram depicting the distribution of major pH regulators in maturation-
stage ameloblasts. Cftr, Slc26a1, Slc26a4, Slc26a6 and Slc26a7 are localized to the apical
membrane and Slc26a1, Slc26a6 and Slc26a7 physically interact with Cftr to form regulation
complexes; Slc26a7 is also found on the endo-lysosomal membrane; Ae2, Nhe1 and NBCe1 are
localized basolaterally; Clcn7 is expressed on the endo-lysosomal membrane; CA2 exhibits
intracellular distribution whereas CA6 functions in extracellular enamel matrix. Lamp1 in this
image was used as a marker of late lysosomes.

114
3.6 Conclusion

In summary, Slc26a1, Slc26a6 and Slc26a7 participate in maintaining the acid-base balance
during amelogenesis, despite that deletion of Slc26a1 or Slc26a7 fails to induce significant
abnormalities in enamel phenotypes. Moreover, Slc26a1, Slc26a6 and Slc26a7 contribute to the
formation of more sophisticated functional complexes involving other stage-specific pH
regulators, which may shed light on future investigations into the pathophysiological
mechanisms of enamel development and health.

115
Chapter Four: Conclusions

Amelogenesis is a process controlled by orchestrated molecular activities. During the
developmental transition from secretory stage to maturation stage, the expression patterns of
miRNAs and mRNAs showed dynamic changes. The differentially expressed miRNAs likely
play an essential role in the regulation of enamel maturation events by targeting genes involved
in specific activities such as pH regulation, ion transport, endocytosis and apoptosis. The data
presented in Chapter Two should help identify the roles of individual miRNAs in amelogenesis,
and more generally help to clarify the potential roles of miRNA-centered regulatory mechanisms
in mineralized tissues. Our data also suggest additional experiments to establish causal
relationships between miRNA and mRNA levels in genetically modified cell culture and animal
models are warranted. Recognition of miRNA-related regulation and the functions of
corresponding target genes during enamel development may also shed light on clinical diagnosis
and/or treatment of diseases such as amelogenesis imperfecta.

In Chapter Three, we investigated the functional roles of a particular group of genes—Slc26a
gene family in the ion transport and pH regulation processes during maturation-stage
amelogenesis, which is based on the genome-wide transcriptome profiling and the subsequent
bioinformatic analyses. We showed that Slc26a1, Slc26a6 and Slc26a7 participate in maintaining
the acid-base balance during amelogenesis, despite that deletion of Slc26a1 or Slc26a7 fails to
induce significant abnormalities in enamel phenotypes. Moreover, Slc26a1, Slc26a6 and Slc26a7
contribute to the formation of more sophisticated functional complexes involving other stage-
specific pH regulators, which may also shed light on future investigations into the
pathophysiological mechanisms of enamel development and health.

116
Table 9. Rat and mouse specific primers for qPCR and cDNA analyses.

Species Symbol Forward (5`-3`) Reverse (5`-3`)
Rat Odam ATCAATTTGGATTTGTACCACA CGTCGGGTTTATTTCAGAAGTGA
Rat Enam TGCAGAAATACAGCTTCTCCT CATTGGCATTGGCATGGCA
Rat Actb AGTGTGACGTTGACATCCGTA GCCAGGGCAGTAATCTCCTTCT
Rat Slc26a1 GCC AGG CAC AGT AGA CAC TT TGC TTC CAG TTA GAG CAT CC
Rat Slc26a6 CCT CAG CCA CCA TGT ACT TC AGC CTG TTT CTG GGA CTT CT
Rat Slc26a7 GCACCGCAGCAGTACAATCTGAA AAAGAATAGCGAAGGACAGC
Mouse Odam TTGACAGCTTTGTAGGCACA GACCTTCTGTTCTGGAGCAA
Mouse Enam TATGGTCTTCCACCAAGGAA TAGGCACACCATCTCCAAAT
Mouse Actb AAGAGCTATGAGCTGCCTGA TACGGATGTCAACGTCACAC
Mouse Car2 TCCCACCACTGGGGATACAG CTCTTGGACGCAGCTTTATCATA
Mouse Car6 TGGAGCTATTCAGGGGATGATG CCGTCTTCACGTCGATGGG
Mouse Cftr GCATATTGTTGGGAATCAGC ACGATTCCGTTGATGACTGT
Mouse Ae1 GGCTGCTGTCATCTTCATCT CTGAGAAGCCTAGCACAAGC
Mouse Ae2 CATGGAGACACAGATCACCA GCTGTTCCTTGACTTCCTGA
Mouse Ae3 TTCCTTGTCCGCTACATCTC TTCAGTTCCAAGCCAGTCTC
Mouse Ae4 CCAGGGCAGGGGGATTTTG CCCCAATGTCTATGCCTGAGG
Mouse NBCe1 CTCCGAGAACTACTCCGACA ACCCTGCTCCACTTTCTCTT
Mouse NBCe2 GGGTCAGGGAGCAGAGAGTTA GGAAGACTCCATAGAGCACTGG
Mouse Slc26a1 CCAACCAGCTCTCTGTCTGT GTCGGACCAGTACCAGTGTC
Mouse Slc26a2 CTACATAAAGCCTGCCCTGA TGAAAGAAGCCCATTGCTAC
Mouse Slc26a3 GTCCTCAAGCTCCACAGAAA GGCAGCTGTAGCATGAATTT
Mouse Slc26a4 GACTGCATGGCAAGCTTAAT CACCACGGAATAGACCAGAC
Mouse Slc26a5 GAAAGGCCCATCTTCAGTCATC GCCACTTAGTGATAGGCAGGAAC
Mouse Slc26a6 TTGCTGGAGCTGTATCTTCC TGTTTGCCTTCCAAAGAGAG
Mouse Slc26a7 CCCCACCGAGAAGACATTAAGC TGAACTGCCAACATTATCCCAG
Mouse Slc26a8 GCCAAGTCACTTCTGAGGAA CCATGGAAAAGTCCAGAATG

117
Mouse Slc26a9 CCCCGCTACGTGGTAGACA AGCACCTGAAAGTGTTGCGA
Mouse Slc26a10 CTGTGGCTGGAGTAACCGTG AGAAAGACGTGTAGAGTCCGA
Mouse Slc26a11 AGTGTATGCGGGAACACATGC CGAAGGCGTAAGCAATCAGAG
Mouse Slc11a2 TCAGAGCTCCACCATGACTG TGTGAACGTGAGGATGGGTA
Mouse Slc36a1 GACTACAGTTCCACAGACGTG CCATGTCATGCTACTGCTCTCT
Mouse Lamp1 TCTATGGCACTGCAACTGAA GGCTCTGTTCTTGTTCTCCA
Mouse Lamp2 AACTTCAACACCCACTCCAA AAAGGCACCTTCTCCTCAGT
Mouse Cd63 CACAGACTGGGAAAACATCC TAATTCCCAAGACCTCCACA
Mouse Cd68 ACATCAGAGCCCGAGTACAG GGTGAACAGCTGGAGAAAGA
Mouse Tfrc ATACACCCGGTTTAGCCTTG GCAATAGCTGCAAAGCAGAG
Mouse Clcn7 CCAAGGAGATTCCACACAAC CAATGAGGGCACAGATAACC
Mouse Atp6v1b2 AAGAAGCTGAGGGGTGAAGT TCTTCCAGTCTCCACTTTGC
Mouse Rab10 AGCAAGAAGGAGTGGGTCTT AGCCAGAGTGCTCAGCTTTA
Mouse Rab21 TCCGCTAAACAGAACAAAGG GGCAATGATCCACAGTTCTC
Mouse Rab24 GTGGACGTTAAGGTGGTTATGC CCCGATGGTGTTCTGATAGGG
Mouse Cltc GCTCGAGAGTCCTATGTGGA CCAAGCGTCCAAAGTTAGAA
Mouse Ap2b1 GGAGGCTGTGAAGAAAGTGA GCATCCCATGGTTCTAACTG
Mouse Mcoln1 AAACACCCCAGTGTCTCCAG ACCAGCCATTGACAAACTCC
Mouse Ctsk ATGGTGAGCTTTGCTCTGTC TATGGGCAGAGATTTGCTTC
Mouse Ctss GCCATTCCTCCTTCTTCTTC CTAGCAATTCCGCAGTGATT
Mouse Alpl TCTGCTCAGGATGAGACTCC TCCCTTTTAACCAACACCAA

118
Table 10. qPCR analysis of gene expression changes in Slc26a1 null animals.
Gene symbols WT Slc26a1 null
Relative expression values
(SD)
Relative expression values
(SD)
Fold
changes
Odam 344.89 (172.28) 998.88 (392.72) 2.9
Enam 0.11 (0.098) 0.25 (0.20) 2.3
Car2 1.85 (0.14) 4.64 (0.18) * 2.5 *
Car6 1.27 (0.30) 0.84 (0.23) -1.5
Cftr 2.51 (0.24) 13.86 (2.56) * 5.5 *
Ae1/Slc4a1 0.00042(0.00039) 0.00065(0.00020) 1.5
Ae2/Slc4a2 6.51 (1.74) 10.37 (1.43) 1.6
Ae3/Slc4a3 0.0024(0.00018) 0.0039(0.0019) 1.6
Ae4/Slc4a9 0.00044 (0.00015) 0.0026 (0.0018) * 5.9 *
NBCe1/Slc4a4 6.26 (1.27) 13.80 (1.79) * 2.2 *
NBCe2/Slc4a5 N/A N/A N/A
Slc26a1 0.13 (0.032) N/A N/A
Slc26a2 0.010 (0.0013) 0.019 (0.0038) ** 1.9 **
Slc26a3 N/A N/A N/A
Slc26a4 155.42 (1.81) 74.20 (8.57) -2.1
Slc26a5 N/A N/A N/A
Slc26a6 0.17 (0.076) 0.39 (0.28) 2.3
Slc26a7 0.018 (0.016) 0.0036 (0.000091) * -5.0 *
Slc26a8 N/A N/A N/A
Slc26a9 0.00094 (0.00020) 0.0051 (0.0043) ** 5.4 **
Slc26a10 N/A N/A N/A
Slc26a11 0.0013 (0.00051) 0.0020 (0.00048) 1.5
Slc11a2 0.011 (0.0017) 0.0080 (0.0038) -1.4
Slc36a1 0.998 (0.00066) 0.997 (0.00095) * -1.1 *

119
Lamp1 0.67 (0.058) 0.90 (0.13) * 1.3 *
Lamp2 0.97 (0.17) 2.37 (0.58) * 2.4 *
Lamp3/Cd63 0.58 (0.068) 0.72 (0.18) 1.2
Lamp4/Cd68 0.056 (0.011) 0.11 (0.039) * 2.0 *
Tfrc 1.75 (0.49) 2.36 (0.94) 1.3 *
Clcn7 0.019 (0.0030) 0.045 (0.018) * 2.4 *
Atp6v1b2 0.35 (0.090) 0.35 (0.049) 1.0
Rab10 0.029 (0.0047) 0.040 (0.010) 1.4
Rab21 0.083 (0.010) 0.13 (0.030) * 1.6 *
Rab24 0.97 (0.0062) 0.96 (0.019) 1.0
Cltc 0.92 (0.025) 0.91 (0.010) 1.0
Ap2b1 0.975 (0.0018) 0.968 (0.0033) ** -1.1 **
Mcoln1 0.0036 (0.00048) 0.014 (0.0074) 3.9
Ctsk 0.0032 (0.00057) 0.0031 (0.0014) 1.0
Ctss 0.062 (0.020) 0.11 (0.053) 1.8
Alpl 0.23(0.040) 0.58(0.094) * 2.5 *
Nhe1/Slc9a1 0.0055(0.0028) 0.0023(0.0026) -2.4

1. Real-time PCR relative expression values are normalized to Beta-Actin.
2. The relative expression values are presented in average values with standard deviation(SD).
3. Four animals in each experimental group (n=4).
4.Two-tail t test was used to detect the potential statistical difference between mutant animal (Slc26a1 null/Slc26a7 null) group and
wild-type animal group. (α=0.05)

120
5. For the expression of each gene detected, fold changes were calculated as the ratio of mutant animal group (Slc26a1 null/Slc26a7
null) to wild-type animal group.
6. * P<0.05 **P<0.01.
7. N/A not detected.

121
Table 11. qPCR analysis of gene expression changes in Slc26a7 null animals.

Gene symbols WT Slc26a7 null
Relative expression values (SD) Relative expression values (SD) Fold changes
Odam 344.89 (172.28) 532.51 (290.26) 1.5
Enam 0.11 (0.098) 0.50 (0.30) * 4.5 *
Car2 1.85 (0.14) 5.50 (0.30) * 3.0 *
Car6 1.27 (0.30) 1.82 (0.64) 1.4
Cftr 2.51 (0.24) 33.81 (4.65) * 13.5 *
Ae1/Slc4a1 0.00042(0.00039) 0.00089(0.00044) 2.1
Ae2/Slc4a2 6.51 (1.74) 14.03 (1.16) * 2.2 *
Ae3/Slc4a3 0.0024(0.00018) 0.0020(0.00092) -1.2
Ae4/Slc4a9 0.00044 (0.00015) 0.031 (0.0045) ** 70.5 **
NBCe1/Slc4a4 6.26 (1.27) 9.78 (1.26) * 1.6 *
NBCe2/Slc4a5 N/A N/A N/A
Slc26a1 0.13 (0.032) 0.11 (0.031) -1.2
Slc26a2 0.010 (0.0013) 0.022 (0.0073) * 2.2 *
Slc26a3 N/A N/A N/A
Slc26a4 155.42 (1.81) 58.26 (4.04) -2.7
Slc26a5 N/A 1.00 (0.00016) N/A
Slc26a6 0.17 (0.076) 0.47 (0.21) * 2.8 *
Slc26a7 0.018 (0.016) N/A N/A
Slc26a8 N/A N/A N/A
Slc26a9 0.00094 (0.00020) 0.078 (0.043) ** 83.0 **
Slc26a10 N/A N/A N/A
Slc26a11 0.0013 (0.00051) 0.0016 (0.00099) 1.2
Slc11a2 0.011 (0.0017) 0.0069 (0.0010) ** -1.6 **

122

1. Real-time PCR relative expression values are normalized to Beta-Actin.
2. The relative expression values are presented in average values with standard deviation(SD).
3. Four animals in each experimental group (n=4).
4.Two-tail t test was used to detect the potential statistical difference between mutant animal (Slc26a1 null/Slc26a7 null) group and
wild-type animal group. (α=0.05)
Slc36a1 0.998 (0.00066) 0.994 (0.0037) * -1.1 *
Lamp1 0.67 (0.058) 0.82 (0.12) * 1.2 *
Lamp2 0.97 (0.17) 2.63 (1.07) 2.7 *
Lamp3/Cd63 0.58 (0.068) 0.96 (0.11) ** 1.7 **
Lamp4/Cd68 0.056 (0.011) 0.12 (0.053) * 2.1 *
Tfrc 1.75 (0.49) 1.26 (0.22) -1.4
Clcn7 0.019 (0.0030) 0.037 (0.018) 1.9
Atp6v1b2 0.35 (0.090) 0.44 (0.081) 1.3
Rab10 0.029 (0.0047) 0.069 (0.037) 2.4
Rab21 0.083 (0.010) 0.12 (0.011) ** 1.4 **
Rab24 0.97 (0.0062) 0.96 (0.020) 1.0
Cltc 0.92 (0.025) 0.90 (0.029) 1.0
Ap2b1 0.975 (0.0018) 0.971 (0.0077) 1.0
Mcoln1 0.0036 (0.00048) 0.0042 (0.00032) 1.2
Ctsk 0.0032 (0.00057) 0.0046 (0.0022) 1.4
Ctss 0.062 (0.020) 0.13 (0.023) ** 2.1 **
Alpl 0.23(0.040) 0.52(0.054) ** 2.3 **
Nhe1/Slc9a1 0.0055(0.0028) 0.0069(0.0012) 1.3

123
5. For the expression of each gene detected, fold changes were calculated as the ratio of mutant animal group (Slc26a1 null/Slc26a7
null) to wild-type animal group.
6. * P<0.05 **P<0.01.
7. N/A not detected.

124
Table 12. Antibodies used for western blot, immunoperoxidase immunostaining, immunofluorescence and co-immunoprecipitation
analyses.

Manufacturer Gene product
detected
Application
Proteintech Slc26a1 (rat) immunohistochemistry (dilution 1:300)
Santa Cruz Biotechnology Slc26a1 (rat) immunofluorescence (dilution 1:400)
western blot/co-immunoprecipitation (dilution 1:100)
Abcam Slc26a6 (rat) immunohistochemistry (dilution 1:5000)
Santa Cruz Biotechnology Slc26a6 (rat) immunofluorescence (dilution 1:100)
western blot/co-immunoprecipitation (dilution 1:100)
Santa Cruz Biotechnology Slc26a7 (rat) immunohistochemistry (dilution 1:100)
Abcam Slc26a7 (rat) immunofluorescence (dilution 1:300)
western blot/co-immunoprecipitation (dilution 1:1000)
Abcam Ae2 (rat) immunofluorescence (dilution 1:100)
Abcam Lamp1 (rat) immunofluorescence (dilution 1:1000)
Santa Cruz Biotechnology Cftr (rat) immunofluorescence (dilution 1:100)
co-immunoprecipitation (amount 1µ g)
Abcam Actb (rat) western blot (dilution 1:3000)

125
Table 13. Enamel and dentin Vickers microhardness of mutant animals.

Enamel, mean (SD) Dentin, mean (SD)
WT (control for Slc26a1 null) 2.3 (0.3) 0.66 (0.02)
Slc26a1 null 2.0 (0.3) 0.62 (0.02)
WT (control for Slc26a7 null) 2.2 (0.1) 0.63 (0.02)
Slc26a7 null 2.5 (0.1) 0.58 (0.02)

The means and standard deviations (SD) of enamel and dentin microhardness are listed in unit of
GPa for mutant animals and their wild-type littermates. The sample size for each experimental
group was 6, and tests were repeated 10 times on average. Differences in hardness between wild
type and null animals were small, if even extant. It appears that biomechanical function, as
measured by hardness was unaffected by the knockout of a single transporter. Likewise, light
microscopy revealed no macro or micro-level differences in form, structure or organization
between wild-type and null animals.

126
Additional Files

Additional file 1 Normalized miRNA expression scores, fold changes and P values.
Additional file 2 Normalized (log) mRNA transcripts expression scores, fold changes and P
values.
Additional file 3 Number of differentially expressed mRNAs with a 5% FDR and different fold
changes.
Additional file 4 Volcano plot depicting the miRNA expression data. the relationships between
fold changes of miRNA expression (X-axis) and adjusted p values (Y-axis) are provided.
Differentially expressed miRNAs (maturation/secretory, Fold Changes ≥ 1.2/≤-1.2, FDR <0.05
are highlighted in red, while the non-differentially expressed miRNAs are colored in blue.
Additional file 5 Up-regulated miRNAs at maturation-stage tooth development relative
secretory-stage and their predicted gene targets expressed in opposite direction.
Additional file 6 Down-regulated miRNAs at maturation-stage tooth development relative
secretory-stage and their predicted gene targets expressed in opposite direction.
Additional file 7 Non-differentially expressed miRNAs between secretory- and maturation-
stage tooth development and their predicted target that are up-regulated during maturation- stage
(relative to secretory-stage).
Additional file 8 Non-differentially expressed miRNAs between secretory- and maturation-
stage tooth development and their predicted target that are down-regulated during maturation-
stage (relative to secretory-stage).

127
Additional file 9 Flow chart depicting the strategies used to select targets for differentially
(Observed) and non-differentially (Baseline) expressed miRNAs. Approximately 5.8%
(629/10,786) of the candidate target genes were differentially expressed in the expected direction
in our analysis. By comparison, approximately 8.1% of candidate target mRNAs (1341 total: 828
up-regulated and 513 down-regulated) were identified to be the potential targets for the stably
expressed miRNAs.
Additional file 10 Up-regulated miRNAs at maturation-stage tooth development relative
secretory-stage (FC > 5) and their predicted gene targets expressed in the same direction.
Additional file 11 Gene Ontology analysis of the 330 up-regulated transcripts (maturation
relative to secretory) that are predicted to be regulated by the 20 down-regulated miRNAs
(maturation relative to secretory).
Additional file 12 Gene Ontology analysis of the 299 down-regulated transcripts (maturation
relative to secretory) that are predicted to be regulated by the 39 up-regulated miRNAs
(maturation relative to secretory).
Additional file 13 KEGG analysis of the 330 up-regulated transcripts (maturation relative to
secretory) that are predicted to be regulated by the 20 down-regulated miRNAs (maturation
relative to secretory).
Additional file 14 KEGG analysis of the 299 down-regulated transcripts (maturation relative to
secretory) that are predicted to be regulated by the 39 up-regulated miRNAs (maturation relative
to secretory).

128
Additional file 15 IPA core analysis of differentially expressed miRNAs and their predicted
gene target candidates (with 140 molecules involved per network interaction ).
Additional file 16 IPA core analysis of differentially expressed miRNAs and their predicted
gene target candidates (with 70 molecules involved per network interaction).
Additional file 17 IPA core analysis of down-regulated miRNAs and their validated predicted
gene target candidates (with 140 molecules involved per network interaction).
Additional file 18 IPA core analysis of down-regulated miRNAs and their validated predicted
gene target candidates (with 140 molecules involved per network interaction).
Additional file 19 IPA Gene Network Analysis. Top gene networks for differentially expressed
miRNAs (maturation/secretory) and their validated predicted gene target candidates were
generated by IPA to show their directed interactions, with 140 (Panels A-C) and 70 (Panels D-F)
molecules involved in each network.
Additional file 20 IPA Gene Network Analysis. Top gene networks for down-regulated
miRNAs (maturation/secretory) and their validated predicted gene target candidates were
generated by IPA to show their directed interactions, with 140 (Panels A-C) and 70 (Panels D-F)
molecules involved in each network.
Additional file 21 IPA Gene Network Analysis. Top gene networks for up-regulated miRNAs
(maturation/secretory) and their validated predicted gene target candidates were generated by
IPA to show their directed interactions, with 140 (Panels A-C) and 70 (Panels D-F) molecules
involved in each network.

129
Link to additional files: http://www.biomedcentral.com/1471-2164/15/998/additional

130
References

1. Nanci A (ed.): Ten Cate`s oral histology development, structure and function., 7th
edn. St Louis, Missouri: Mosby Elsevier; 2008.
2. Lacruz RS, Nanci A, Kurtz I, Wright JT, Paine ML: Regulation of pH During
Amelogenesis. Calcified Tissue International 2010, 86(2):91-103.
3. Smith CE: Cellular and chemical events during enamel maturation. Crit Rev Oral
Biol Med 1998, 9(2):128-161.
4. Smith CE, Issid M, Margolis HC, Moreno EC: Developmental changes in the pH of
enamel fluid and its effects on matrix-resident proteinases. Adv Dent Res 1996,
10(2):159-169.
5. Takano Y: Enamel mineralization and the role of ameloblasts in calcium transport.
Connect Tissue Res 1995, 33(1-3):127-137.
6. Hubbard MJ: Calcium transport across the dental enamel epithelium. Crit Rev Oral
Biol Med 2000, 11(4):437-466.
7. Josephsen K, Takano Y, Frische S, Praetorius J, Nielsen S, Aoba T, Fejerskov O: Ion
transporters in secretory and cyclically modulating ameloblasts: a new hypothesis
for cellular control of preeruptive enamel maturation. Am J Physiol Cell Physiol
2010, 299(6):C1299-1307.
8. Lacruz RS, Smith CE, Bringas P, Chen Y-B, Smith SM, Snead ML, Kurtz I, Hacia JG,
Hubbard MJ, Paine ML: Identification of novel candidate genes involved in
mineralization of dental enamel by genome-wide transcript profiling. Journal of
Cellular Physiology 2012, 227(5):2264-2275.
9. Lacruz RS, Nanci A, Kurtz I, Wright JT, Paine ML: Regulation of pH During
Amelogenesis. Calcified Tissue International 2009, 86(2):91-103.
10. Lacruz RS, Brookes SJ, Wen X, Jimenez JM, Vikman S, Hu P, White SN, Lyngstadaas
SP, Okamoto CT, Smith CE et al: Adaptor protein complex 2-mediated, clathrin-
dependent endocytosis, and related gene activities, are a prominent feature during
maturation stage amelogenesis. J Bone Miner Res 2013, 28(3):672-687.
11. Simmer JP, Hu JC: Expression, structure, and function of enamel proteinases.
Connect Tissue Res 2002, 43(2-3):441-449.
12. Smith CE, Nanci A: Protein dynamics of amelogenesis. Anat Rec 1996, 245(2):186-
207.
13. Simmer JP, Fincham AG: Molecular mechanisms of dental enamel formation. Crit
Rev Oral Biol Med 1995, 6(2):84-108.
14. Shapiro JL, Wen X, Okamoto CT, Wang HJ, Lyngstadaas SP, Goldberg M, Snead ML,
Paine ML: Cellular uptake of amelogenin, and its localization to CD63, and Lamp1-
positive vesicles. Cell Mol Life Sci 2007, 64(2):244-256.
15. Lloyd JB: Metabolite efflux and influx across the lysosome membrane. Subcell
Biochem 1996, 27:361-386.
16. Lacruz RS, Nanci A, Kurtz I, Wright JT, Paine ML: Regulation of pH During
Amelogenesis. Calcif Tissue Int 2010, 86(2):91-103.
17. Lyman GE, Waddell WJ: pH gradients in the developing teeth of young mice from
autoradiography of [14C]DMO. Am J Physiol 1977, 232(4):F364-367.

131
18. Sasaki S, Takagi T, Suzuki M: Cyclical changes in pH in bovine developing enamel as
sequential bands. Arch Oral Biol 1991, 36(3):227-231.
19. Arquitt CK, Boyd C, Wright JT: Cystic fibrosis transmembrane regulator gene
(CFTR) is associated with abnormal enamel formation. J Dent Res 2002, 81(7):492-
496.
20. Bronckers A, Kalogeraki L, Jorna HJ, Wilke M, Bervoets TJ, Lyaruu DM, Zandieh-
Doulabi B, Denbesten P, de Jonge H: The cystic fibrosis transmembrane conductance
regulator (CFTR) is expressed in maturation stage ameloblasts, odontoblasts and
bone cells. Bone 2010, 46(4):1188-1196.
21. Bronckers AL, Lyaruu DM, Jansen ID, Medina JF, Kellokumpu S, Hoeben KA, Gawenis
LR, Oude-Elferink RP, Everts V: Localization and function of the anion exchanger
Ae2 in developing teeth and orofacial bone in rodents. J Exp Zool B Mol Dev Evol
2009, 312B(4):375-387.
22. Chang EH, Lacruz RS, Bromage TG, Bringas P, Jr., Welsh MJ, Zabner J, Paine ML:
Enamel pathology resulting from loss of function in the cystic fibrosis
transmembrane conductance regulator in a porcine animal model. Cells Tissues
Organs 2011, 194(2-4):249-254.
23. Dogterom AA, Bervoets TJ, Lyaruu DM, Woltgens JH: Carbonate incorporation and
carbonic anhydrase activity in developing hamster molars in vitro. J Biol Buccale
1983, 11(1):15-21.
24. Dogterom AA, Bronckers AL: Carbonic anhydrase in developing hamster molars. J
Dent Res 1983, 62(7):789-791.
25. Duan X: Ion channels, channelopathies, and tooth formation. J Dent Res 2014,
93(2):117-125.
26. Duan X, Mao Y, Wen X, Yang T, Xue Y: Excess fluoride interferes with chloride-
channel-dependent endocytosis in ameloblasts. J Dent Res 2011, 90(2):175-180.
27. Jalali R, Guo J, Zandieh-Doulabi B, Bervoets TJ, Paine ML, Boron WF, Parker MD,
Bijvelds MJ, Medina JF, DenBesten PK et al: NBCe1 (SLC4A4) a potential pH
regulator in enamel organ cells during enamel development in the mouse. Cell Tissue
Res 2014.
28. Lacruz RS, Hilvo M, Kurtz I, Paine ML: A survey of carbonic anhydrase mRNA
expression in enamel cells. Biochem Biophys Res Commun 2010, 393(4):883-887.
29. Lacruz RS, Nanci A, White SN, Wen X, Wang H, Zalzal SF, Luong VQ, Schuetter VL,
Conti PS, Kurtz I et al: The sodium bicarbonate cotransporter (NBCe1) is essential
for normal development of mouse dentition. J Biol Chem 2010, 285(32):24432-24438.
30. Lacruz RS, Smith CE, Moffatt P, Chang EH, Bromage TG, Bringas P, Jr., Nanci A,
Baniwal SK, Zabner J, Welsh MJ et al: Requirements for ion and solute transport,
and pH regulation during enamel maturation. J Cell Physiol 2012, 227(4):1776-1785.
31. Lin HM, Nakamura H, Noda T, Ozawa H: Localization of H(+)-ATPase and carbonic
anhydrase II in ameloblasts at maturation. Calcif Tissue Int 1994, 55(1):38-45.
32. Lyaruu DM, Bronckers AL, Mulder L, Mardones P, Medina JF, Kellokumpu S, Oude
Elferink RP, Everts V: The anion exchanger Ae2 is required for enamel maturation
in mouse teeth. Matrix Biol 2008, 27(2):119-127.
33. Paine ML, Snead ML, Wang HJ, Abuladze N, Pushkin A, Liu W, Kao LY, Wall SM,
Kim YH, Kurtz I: Role of NBCe1 and AE2 in secretory ameloblasts. J Dent Res 2008,
87(4):391-395.

132
34. Reibring CG, El Shahawy M, Hallberg K, Kannius-Janson M, Nilsson J, Parkkila S, Sly
WS, Waheed A, Linde A, Gritli-Linde A: Expression patterns and subcellular
localization of carbonic anhydrases are developmentally regulated during tooth
formation. PLoS One 2014, 9(5):e96007.
35. Wang X, Suzawa T, Ohtsuka H, Zhao B, Miyamoto Y, Miyauchi T, Nishimura R, Inoue
T, Nakamura M, Baba K et al: Carbonic anhydrase II regulates differentiation of
ameloblasts via intracellular pH-dependent JNK signaling pathway. J Cell Physiol
2010, 225(3):709-719.
36. Wen X, Kurtz I, Paine ML: Prevention of the disrupted enamel phenotype in Slc4a4-
null mice using explant organ culture maintained in a living host kidney capsule.
PLoS One 2014, 9(5):e97318.
37. Wright JT, Hall KI, Grubb BR: Enamel mineral composition of normal and cystic
fibrosis transgenic mice. Adv Dent Res 1996, 10(2):270-274; discussion 275.
38. Wright JT, Kiefer CL, Hall KI, Grubb BR: Abnormal enamel development in a cystic
fibrosis transgenic mouse model. J Dent Res 1996, 75(4):966-973.
39. Bartlett JD, Simmer JP: Proteinases in developing dental enamel. Crit Rev Oral Biol
Med 1999, 10(4):425-441.
40. Lu Y, Papagerakis P, Yamakoshi Y, Hu JC, Bartlett JD, Simmer JP: Functions of KLK4
and MMP-20 in dental enamel formation. Biol Chem 2008, 389(6):695-700.
41. Josephsen K, Warshawsky H: Radioautography of rat incisor dentin as a continuous
record of the incorporation of a single dose of 3H-labeled proline and tyrosine. Am J
Anat 1982, 164(1):45-56.
42. Katchburian E, Holt SJ: Role of lysosomes in amelogenesis. Nature 1969,
223(5213):1367-1368.
43. Nanci A, Fortin M, Ghitescu L: Endocytotic functions of ameloblasts and
odontoblasts: immunocytochemical and tracer studies on the uptake of plasma
proteins. Anat Rec 1996, 245(2):219-234.
44. Sasaki T: Endocytotic pathways at the ruffled borders of rat maturation
ameloblasts. Histochemistry 1984, 80(3):263-268.
45. Franklin DL, Severs NJ, Katchburian E: Development of the distal end and Tomes'
processes of ameloblasts observed by freeze-fracture and ultrathin section electron
microscopy. J Anat 1991, 174:103-114.
46. Sasaki T: Tracer, cytochemical, and freeze-fracture study on the mechanisms
whereby secretory ameloblasts absorb exogeneous proteins. Acta Anat (Basel) 1984,
118(1):23-33.
47. Sasaki T, Higashi S, Tachikawa T, Yoshiki S: Thin-section, tracer, and freeze-fracture
study of the smooth-ended maturation ameloblasts in rat incisors. Acta Anat (Basel)
1983, 117(4):303-313.
48. Sasaki T, Motegi N, Higashi S: [Ultrastructural and cytochemical study of the
resorptive function of the secretory ameloblasts]. Showa Shigakkai Zasshi 1984,
3(2):151-162.
49. Zou Y, Wang H, Shapiro JL, Okamoto CT, Brookes SJ, Lyngstadaas SP, Snead ML,
Paine ML: Determination of protein regions responsible for interactions of
amelogenin with CD63 and LAMP1. Biochem J 2007, 408(3):347-354.
50. Bronckers AL, Goei SW, Dumont E, Lyaruu DM, Woltgens JH, van Heerde WL,
Reutelingsperger CP, van den Eijnde SM: In situ detection of apoptosis in dental and

133
periodontal tissues of the adult mouse using annexin-V-biotin. Histochem Cell Biol
2000, 113(4):293-301.
51. Lesot H, Vonesch JL, Peterka M, Tureckova J, Peterkova R, Ruch JV: Mouse molar
morphogenesis revisited by three-dimensional reconstruction. II. Spatial
distribution of mitoses and apoptosis in cap to bell staged first and second upper
molar teeth. Int J Dev Biol 1996, 40(5):1017-1031.
52. Sasaki T: Morphology and function of maturation ameloblasts in kitten tooth germs.
J Anat 1984, 138 ( Pt 2):333-342.
53. Shigemura T, Hatakeyama N, Shibuya N, Yamazaki M, Masuda A, Chen FS, Momose Y,
Ito Y: Effects of propofol on contractile response and electrophysiological properties
in single guinea-pig ventricular myocyte. Pharmacol Toxicol 1999, 85(3):111-114.
54. Vaahtokari A, Aberg T, Thesleff I: Apoptosis in the developing tooth: association with
an embryonic signaling center and suppression by EGF and FGF-4. Development
1996, 122(1):121-129.
55. Viriot L, Peterkova R, Vonesch JL, Peterka M, Ruch JV, Lesot H: Mouse molar
morphogenesis revisited by three-dimensional reconstruction. III. Spatial
distribution of mitoses and apoptoses up to bell-staged first lower molar teeth. Int J
Dev Biol 1997, 41(5):679-690.
56. Matalova E, Svandova E, Tucker AS: Apoptotic signaling in mouse odontogenesis.
OMICS 2012, 16(1-2):60-70.
57. Yin K, Hacia JG, Zhong Z, Paine ML: Genome-wide analysis of miRNA and mRNA
transcriptomes during amelogenesis. BMC Genomics 2014, 15:998.
58. Gadhia K, McDonald S, Arkutu N, Malik K: Amelogenesis imperfecta: an
introduction. Br Dent J 2012, 212(8):377-379.
59. Wright JT, Hart TC, Hart PS, Simmons D, Suggs C, Daley B, Simmer J, Hu J, Bartlett
JD, Li Y et al: Human and mouse enamel phenotypes resulting from mutation or
altered expression of AMEL, ENAM, MMP20 and KLK4. Cells Tissues Organs 2009,
189(1-4):224-229.
60. Wright JT, Hart PS, Aldred MJ, Seow K, Crawford PJ, Hong SP, Gibson CW, Hart TC:
Relationship of phenotype and genotype in X-linked amelogenesis imperfecta.
Connect Tissue Res 2003, 44 Suppl 1:72-78.
61. Mardh CK, Backman B, Holmgren G, Hu JC, Simmer JP, Forsman-Semb K: A nonsense
mutation in the enamelin gene causes local hypoplastic autosomal dominant
amelogenesis imperfecta (AIH2). Hum Mol Genet 2002, 11(9):1069-1074.
62. Kim JW, Simmer JP, Hart TC, Hart PS, Ramaswami MD, Bartlett JD, Hu JC: MMP-20
mutation in autosomal recessive pigmented hypomaturation amelogenesis
imperfecta. J Med Genet 2005, 42(3):271-275.
63. Hart PS, Hart TC, Michalec MD, Ryu OH, Simmons D, Hong S, Wright JT: Mutation in
kallikrein 4 causes autosomal recessive hypomaturation amelogenesis imperfecta. J
Med Genet 2004, 41(7):545-549.
64. Lee SK, Hu JC, Bartlett JD, Lee KE, Lin BP, Simmer JP, Kim JW: Mutational
spectrum of FAM83H: the C-terminal portion is required for tooth enamel
calcification. Hum Mutat 2008, 29(8):E95-99.
65. Wright JT, Frazier-Bowers S, Simmons D, Alexander K, Crawford P, Han ST, Hart PS,
Hart TC: Phenotypic variation in FAM83H-associated amelogenesis imperfecta. J
Dent Res 2009, 88(4):356-360.

134
66. El-Sayed W, Parry DA, Shore RC, Ahmed M, Jafri H, Rashid Y, Al-Bahlani S, Al Harasi
S, Kirkham J, Inglehearn CF et al: Mutations in the beta propeller WDR72 cause
autosomal-recessive hypomaturation amelogenesis imperfecta. Am J Hum Genet
2009, 85(5):699-705.
67. El-Sayed W, Shore RC, Parry DA, Inglehearn CF, Mighell AJ: Hypomaturation
amelogenesis imperfecta due to WDR72 mutations: a novel mutation and
ultrastructural analyses of deciduous teeth. Cells Tissues Organs 2011, 194(1):60-66.
68. Kuechler A, Hentschel J, Kurth I, Stephan B, Prott EC, Schweiger B, Schuster A,
Wieczorek D, Ludecke HJ: A Novel Homozygous WDR72 Mutation in Two Siblings
with Amelogenesis Imperfecta and Mild Short Stature. Mol Syndromol 2012,
3(5):223-229.
69. Lee SK, Seymen F, Lee KE, Kang HY, Yildirim M, Tuna EB, Gencay K, Hwang YH,
Nam KH, De La Garza RJ et al: Novel WDR72 mutation and cytoplasmic localization.
J Dent Res 2010, 89(12):1378-1382.
70. Wang S, Choi M, Richardson AS, Reid BM, Seymen F, Yildirim M, Tuna E, Gencay K,
Simmer JP, Hu JC: STIM1 and SLC24A4 Are Critical for Enamel Maturation. J
Dent Res 2014, 93(7 suppl):94S-100S.
71. Bartel DP: MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004,
116(2):281-297.
72. Kim VN: MicroRNA biogenesis: coordinated cropping and dicing. Nat Rev Mol Cell
Biol 2005, 6(5):376-385.
73. Li Z, Rana TM: Therapeutic targeting of microRNAs: current status and future
challenges. Nat Rev Drug Discov 2014, 13(8):622-638.
74. Guo H, Ingolia NT, Weissman JS, Bartel DP: Mammalian microRNAs predominantly
act to decrease target mRNA levels. Nature 2010, 466(7308):835-840.
75. Li Z, Rana TM: Molecular mechanisms of RNA-triggered gene silencing
machineries. Acc Chem Res 2012, 45(7):1122-1131.
76. Bazzini AA, Lee MT, Giraldez AJ: Ribosome profiling shows that miR-430 reduces
translation before causing mRNA decay in zebrafish. Science 2012, 336(6078):233-
237.
77. Djuranovic S, Nahvi A, Green R: miRNA-mediated gene silencing by translational
repression followed by mRNA deadenylation and decay. Science 2012,
336(6078):237-240.
78. Iwasaki S, Kawamata T, Tomari Y: Drosophila argonaute1 and argonaute2 employ
distinct mechanisms for translational repression. Mol Cell 2009, 34(1):58-67.
79. Wakiyama M, Takimoto K, Ohara O, Yokoyama S: Let-7 microRNA-mediated mRNA
deadenylation and translational repression in a mammalian cell-free system. Genes
Dev 2007, 21(15):1857-1862.
80. Zdanowicz A, Thermann R, Kowalska J, Jemielity J, Duncan K, Preiss T, Darzynkiewicz
E, Hentze MW: Drosophila miR2 primarily targets the m7GpppN cap structure for
translational repression. Mol Cell 2009, 35(6):881-888.
81. Wu E, Thivierge C, Flamand M, Mathonnet G, Vashisht AA, Wohlschlegel J, Fabian
MR, Sonenberg N, Duchaine TF: Pervasive and cooperative deadenylation of 3'UTRs
by embryonic microRNA families. Mol Cell 2010, 40(4):558-570.

135
82. Behm-Ansmant I, Rehwinkel J, Doerks T, Stark A, Bork P, Izaurralde E: mRNA
degradation by miRNAs and GW182 requires both CCR4:NOT deadenylase and
DCP1:DCP2 decapping complexes. Genes Dev 2006, 20(14):1885-1898.
83. Chen CY, Zheng D, Xia Z, Shyu AB: Ago-TNRC6 triggers microRNA-mediated
decay by promoting two deadenylation steps. Nat Struct Mol Biol 2009, 16(11):1160-
1166.
84. Eulalio A, Huntzinger E, Nishihara T, Rehwinkel J, Fauser M, Izaurralde E:
Deadenylation is a widespread effect of miRNA regulation. RNA 2009, 15(1):21-32.
85. Rehwinkel J, Behm-Ansmant I, Gatfield D, Izaurralde E: A crucial role for GW182 and
the DCP1:DCP2 decapping complex in miRNA-mediated gene silencing. RNA 2005,
11(11):1640-1647.
86. Piao X, Zhang X, Wu L, Belasco JG: CCR4-NOT deadenylates mRNA associated
with RNA-induced silencing complexes in human cells. Mol Cell Biol 2010,
30(6):1486-1494.
87. Pritchard CC, Cheng HH, Tewari M: MicroRNA profiling: approaches and
considerations. Nat Rev Genet 2012, 13(5):358-369.
88. Benes V, Castoldi M: Expression profiling of microRNA using real-time quantitative
PCR, how to use it and what is available. Methods 2010, 50(4):244-249.
89. Git A, Dvinge H, Salmon-Divon M, Osborne M, Kutter C, Hadfield J, Bertone P, Caldas
C: Systematic comparison of microarray profiling, real-time PCR, and next-
generation sequencing technologies for measuring differential microRNA
expression. RNA 2010, 16(5):991-1006.
90. Nelson PT, Wang WX, Wilfred BR, Tang G: Technical variables in high-throughput
miRNA expression profiling: much work remains to be done. Biochim Biophys Acta
2008, 1779(11):758-765.
91. Yin JQ, Zhao RC, Morris KV: Profiling microRNA expression with microarrays.
Trends Biotechnol 2008, 26(2):70-76.
92. Creighton CJ, Reid JG, Gunaratne PH: Expression profiling of microRNAs by deep
sequencing. Brief Bioinform 2009, 10(5):490-497.
93. Jevnaker AM, Osmundsen H: MicroRNA expression profiling of the developing
murine molar tooth germ and the developing murine submandibular salivary gland.
Arch Oral Biol 2008, 53(7):629-645.
94. Michon F, Tummers M, Kyyrö nen M, Frilander MJ, Thesleff I: Tooth morphogenesis
and ameloblast differentiation are regulated by micro-RNAs. Developmental Biology
2010, 340(2):355-368.
95. Cao H, Wang J, Li X, Florez S, Huang Z, Venugopalan SR, Elangovan S, Skobe Z,
Margolis HC, Martin JF et al: MicroRNAs Play a Critical Role in Tooth Development.
Journal of Dental Research 2010, 89(8):779-784.
96. Shi S, Wan M, Gao B, Sun F, Tang Y, Ye L, Fan Y, Klein OD, Zhou X, Zheng L:
microRNA miR-34a Regulates Cytodifferentiation and Targets Multi-signaling
Pathways in Human Dental Papilla Cells. PLoS ONE 2012, 7(11):e50090.
97. Liu H, Lin H, Zhang L, Sun Q, Yuan G, Chen S, Chen Z: miR-145 and miR-143
Regulate Odontoblast Differentiation through Targeting Klf4 and Osx Genes in a
Feedback Loop. Journal of Biological Chemistry 2013, 288(13):9261-9271.

136
98. Li A, Song T, Wang F, Liu D, Fan Z, Zhang C, He J, Wang S: MicroRNAome and
expression profile of developing tooth germ in miniature pigs. PLoS ONE 2012,
7(12):e52256.
99. Jheon AH, Li CY, Wen T, Michon F, Klein OD: Expression of microRNAs in the stem
cell niche of the adult mouse incisor. PLoS ONE 2011, 6(9):e24536.
100. Alper SL, Sharma AK: The SLC26 gene family of anion transporters and channels.
Molecular Aspects of Medicine 2013, 34(2-3):494-515.
101. Aravind L, Koonin EV: The STAS domain - a link between anion transporters and
antisigma-factor antagonists. Curr Biol 2000, 10(2):R53-55.
102. Aronson PS: Role of SLC26A6-mediated Cl(-)-oxalate exchange in renal physiology
and pathophysiology. J Nephrol 2010, 23 Suppl 16:S158-164.
103. Dorwart MR, Shcheynikov N, Yang D, Muallem S: The solute carrier 26 family of
proteins in epithelial ion transport. Physiology (Bethesda) 2008, 23:104-114.
104. Kere J: Overview of the SLC26 family and associated diseases. Novartis Found Symp
2006, 273:2-11; discussion 11-18, 261-264.
105. Markovich D: Physiological roles of renal anion transporters NaS1 and Sat1. Am J
Physiol Renal Physiol 2011, 300(6):F1267-1270.
106. Ohana E, Yang D, Shcheynikov N, Muallem S: Diverse transport modes by the solute
carrier 26 family of anion transporters. J Physiol 2009, 587(Pt 10):2179-2185.
107. Schnedler N, Burckhardt G, Burckhardt BC: Glyoxylate is a substrate of the sulfate-
oxalate exchanger, sat-1, and increases its expression in HepG2 cells. J Hepatol 2011,
54(3):513-520.
108. Xie Q, Welch R, Mercado A, Romero MF, Mount DB: Molecular characterization of
the murine Slc26a6 anion exchanger: functional comparison with Slc26a1. Am J
Physiol Renal Physiol 2002, 283(4):F826-838.
109. Regeer RR, Markovich D: A dileucine motif targets the sulfate anion transporter sat-
1 to the basolateral membrane in renal cell lines. Am J Physiol Cell Physiol 2004,
287(2):C365-372.
110. Markovich D: Slc13a1 and Slc26a1 KO models reveal physiological roles of anion
transporters. Physiology (Bethesda) 2012, 27(1):7-14.
111. Knauf F, Yang CL, Thomson RB, Mentone SA, Giebisch G, Aronson PS: Identification
of a chloride-formate exchanger expressed on the brush border membrane of renal
proximal tubule cells. Proc Natl Acad Sci U S A 2001, 98(16):9425-9430.
112. Petrovic S, Ma L, Wang Z, Soleimani M: Identification of an apical Cl-/HCO-3
exchanger in rat kidney proximal tubule. Am J Physiol Cell Physiol 2003,
285(3):C608-617.
113. Wang Z, Petrovic S, Mann E, Soleimani M: Identification of an apical Cl(-)/HCO3(-)
exchanger in the small intestine. Am J Physiol Gastrointest Liver Physiol 2002,
282(3):G573-579.
114. Aronson PS, Giebisch G: Mechanisms of chloride transport in the proximal tubule.
Am J Physiol 1997, 273(2 Pt 2):F179-192.
115. Freel RW, Hatch M, Green M, Soleimani M: Ileal oxalate absorption and urinary
oxalate excretion are enhanced in Slc26a6 null mice. Am J Physiol Gastrointest Liver
Physiol 2006, 290(4):G719-728.

137
116. Jiang Z, Asplin JR, Evan AP, Rajendran VM, Velazquez H, Nottoli TP, Binder HJ,
Aronson PS: Calcium oxalate urolithiasis in mice lacking anion transporter Slc26a6.
Nat Genet 2006, 38(4):474-478.
117. Petrovic S, Ju X, Barone S, Seidler U, Alper SL, Lohi H, Kere J, Soleimani M:
Identification of a basolateral Cl-/HCO3- exchanger specific to gastric parietal cells.
Am J Physiol Gastrointest Liver Physiol 2003, 284(6):G1093-1103.
118. Petrovic S, Barone S, Xu J, Conforti L, Ma L, Kujala M, Kere J, Soleimani M:
SLC26A7: a basolateral Cl-/HCO3- exchanger specific to intercalated cells of the
outer medullary collecting duct. Am J Physiol Renal Physiol 2004, 286(1):F161-169.
119. Xu J, Worrell RT, Li HC, Barone SL, Petrovic S, Amlal H, Soleimani M:
Chloride/bicarbonate exchanger SLC26A7 is localized in endosomes in medullary
collecting duct cells and is targeted to the basolateral membrane in hypertonicity
and potassium depletion. J Am Soc Nephrol 2006, 17(4):956-967.
120. Xu J, Song P, Nakamura S, Miller M, Barone S, Alper SL, Riederer B, Bonhagen J,
Arend LJ, Amlal H et al: Deletion of the chloride transporter slc26a7 causes distal
renal tubular acidosis and impairs gastric acid secretion. J Biol Chem 2009,
284(43):29470-29479.
121. Hu JC, Chun YH, Al Hazzazzi T, Simmer JP: Enamel formation and amelogenesis
imperfecta. Cells Tissues Organs 2007, 186(1):78-85.
122. Stephanopoulos G, Garefalaki ME, Lyroudia K: Genes and related proteins involved in
amelogenesis imperfecta. J Dent Res 2005, 84(12):1117-1126.
123. Wright JT: The molecular etiologies and associated phenotypes of amelogenesis
imperfecta. Am J Med Genet A 2006, 140(23):2547-2555.
124. Wright JT, Torain M, Long K, Seow K, Crawford P, Aldred MJ, Hart PS, Hart TC:
Amelogenesis imperfecta: genotype-phenotype studies in 71 families. Cells Tissues
Organs 2011, 194(2-4):279-283.
125. Cao H, Wang J, Li X, Florez S, Huang Z, Venugopalan SR, Elangovan S, Skobe Z,
Margolis HC, Martin JF et al: MicroRNAs play a critical role in tooth development. J
Dent Res 2010, 89(8):779-784.
126. Michon F, Tummers M, Kyyronen M, Frilander MJ, Thesleff I: Tooth morphogenesis
and ameloblast differentiation are regulated by micro-RNAs. Dev Biol 2010,
340(2):355-368.
127. Wan M, Gao B, Sun F, Tang Y, Ye L, Fan Y, Klein OD, Zhou X, Zheng L: microRNA
miR-34a regulates cytodifferentiation and targets multi-signaling pathways in
human dental papilla cells. PLoS One 2012, 7(11):e50090.
128. Liu H, Lin H, Zhang L, Sun Q, Yuan G, Zhang L, Chen S, Chen Z: miR-145 and miR-
143 regulate odontoblast differentiation through targeting Klf4 and Osx genes in a
feedback loop. J Biol Chem 2013, 288(13):9261-9271.
129. Lacruz RS, Smith CE, Bringas P, Jr., Chen YB, Smith SM, Snead ML, Kurtz I, Hacia JG,
Hubbard MJ, Paine ML: Identification of novel candidate genes involved in
mineralization of dental enamel by genome-wide transcript profiling. J Cell Physiol
2012, 227(5):2264-2275.
130. Lacruz RS, Smith CE, Chen YB, Hubbard MJ, Hacia JG, Paine ML: Gene-expression
analysis of early- and late-maturation-stage rat enamel organ. Eur J Oral Sci 2011,
119 Suppl 1:149-157.

138
131. Wang J, Duncan D, Shi Z, Zhang B: WEB-based GEne SeT AnaLysis Toolkit
(WebGestalt): update 2013. Nucleic Acids Res 2013, 41(Web Server issue):W77-83.
132. Zhang B, Kirov S, Snoddy J: WebGestalt: an integrated system for exploring gene
sets in various biological contexts. Nucleic Acids Res 2005, 33(Web Server
issue):W741-748.
133. Chen LS, Couwenhoven RI, Hsu D, Luo W, Snead ML: Maintenance of amelogenin
gene expression by transformed epithelial cells of mouse enamel organ. Arch Oral
Biol 1992, 37(10):771-778.
134. Xiao F, Zuo Z, Cai G, Kang S, Gao X, Li T: miRecords: an integrated resource for
microRNA-target interactions. Nucleic Acids Res 2009, 37(Database issue):D105-110.
135. Papadopoulos GL, Reczko M, Simossis VA, Sethupathy P, Hatzigeorgiou AG: The
database of experimentally supported targets: a functional update of TarBase.
Nucleic Acids Res 2009, 37(Database issue):D155-158.
136. Lewis BP, Burge CB, Bartel DP: Conserved seed pairing, often flanked by
adenosines, indicates that thousands of human genes are microRNA targets. Cell
2005, 120(1):15-20.
137. Grimson A, Farh KK, Johnston WK, Garrett-Engele P, Lim LP, Bartel DP: MicroRNA
targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 2007,
27(1):91-105.
138. Friedman RC, Farh KK, Burge CB, Bartel DP: Most mammalian mRNAs are
conserved targets of microRNAs. Genome Res 2009, 19(1):92-105.
139. Chen HC, Chen GH, Chen YH, Liao WL, Liu CY, Chang KP, Chang YS, Chen SJ:
MicroRNA deregulation and pathway alterations in nasopharyngeal carcinoma. Br
J Cancer 2009, 100(6):1002-1011.
140. Chen HC, Lee YS, Sieber M, Lu HT, Wei PC, Wang CN, Peng HH, Chao AS, Cheng PJ,
Chang SD et al: MicroRNA and messenger RNA analyses of mesenchymal stem cells
derived from teeth and the Wharton jelly of umbilical cord. Stem Cells Dev 2012,
21(6):911-922.
141. Clarke C, Henry M, Doolan P, Kelly S, Aherne S, Sanchez N, Kelly P, Kinsella P, Breen
L, Madden SF et al: Integrated miRNA, mRNA and protein expression analysis
reveals the role of post-transcriptional regulation in controlling CHO cell growth
rate. BMC Genomics 2012, 13:656.
142. Dong J, Carey WA, Abel S, Collura C, Jiang G, Tomaszek S, Sutor S, Roden AC,
Asmann YW, Prakash YS et al: MicroRNA-mRNA interactions in a murine model of
hyperoxia-induced bronchopulmonary dysplasia. BMC Genomics 2012, 13:204.
143. He HC, Han ZD, Dai QS, Ling XH, Fu X, Lin ZY, Deng YH, Qin GQ, Cai C, Chen JH et
al: Global analysis of the differentially expressed miRNAs of prostate cancer in
Chinese patients. BMC Genomics 2013, 14:757.
144. Huang C, Xiao X, Chintagari NR, Breshears M, Wang Y, Liu L: MicroRNA and mRNA
expression profiling in rat acute respiratory distress syndrome. BMC Med Genomics
2014, 7:46.
145. Ichimi T, Enokida H, Okuno Y, Kunimoto R, Chiyomaru T, Kawamoto K, Kawahara K,
Toki K, Kawakami K, Nishiyama K et al: Identification of novel microRNA targets
based on microRNA signatures in bladder cancer. Int J Cancer 2009, 125(2):345-352.

139
146. Tan KS, Choi H, Jiang X, Yin L, Seet JE, Patzel V, Engelward BP, Chow VT: Micro-
RNAs in regenerating lungs: an integrative systems biology analysis of murine
influenza pneumonia. BMC Genomics 2014, 15:587.
147. Du T, Zamore PD: Beginning to understand microRNA function. Cell Res 2007,
17(8):661-663.
148. Salmanidis M, Pillman K, Goodall G, Bracken C: Direct transcriptional regulation by
nuclear microRNAs. Int J Biochem Cell Biol 2014.
149. Standart N, Jackson RJ: MicroRNAs repress translation of m7Gppp-capped target
mRNAs in vitro by inhibiting initiation and promoting deadenylation. Genes Dev
2007, 21(16):1975-1982.
150. Pillai RS, Bhattacharyya SN, Filipowicz W: Repression of protein synthesis by
miRNAs: how many mechanisms? Trends Cell Biol 2007, 17(3):118-126.
151. Bartlett JD, Ganss B, Goldberg M, Moradian-Oldak J, Paine ML, Snead ML, Wen X,
White SN, Zhou YL: 3. Protein-protein interactions of the developing enamel matrix.
Curr Top Dev Biol 2006, 74:57-115.
152. Tompkins K, George A, Veis A: Characterization of a mouse amelogenin [A-4]/M59
cell surface receptor. Bone 2006, 38(2):172-180.
153. Sui W, Boyd C, Wright JT: Altered pH regulation during enamel development in the
cystic fibrosis mouse incisor. J Dent Res 2003, 82(5):388-392.
154. Yoshida T, Kumashiro Y, Iwata T, Ishihara J, Umemoto T, Shiratsuchi Y, Kawashima N,
Sugiyama T, Yamato M, Okano T: Requirement of integrin beta3 for iron
transportation during enamel formation. J Dent Res 2012, 91(12):1154-1159.
155. Cai J, Mutoh N, Shin JO, Tani-Ishii N, Ohshima H, Cho SW, Jung HS: Wnt5a plays a
crucial role in determining tooth size during murine tooth development. Cell Tissue
Res 2011, 345(3):367-377.
156. Kuhn DE, Martin MM, Feldman DS, Terry AV, Jr., Nuovo GJ, Elton TS: Experimental
validation of miRNA targets. Methods 2008, 44(1):47-54.
157. Cottonham CL, Kaneko S, Xu L: miR-21 and miR-31 converge on TIAM1 to regulate
migration and invasion of colon carcinoma cells. J Biol Chem 2010, 285(46):35293-
35302.
158. Rouas R, Fayyad-Kazan H, El Zein N, Lewalle P, Rothe F, Simion A, Akl H, Mourtada
M, El Rifai M, Burny A et al: Human natural Treg microRNA signature: role of
microRNA-31 and microRNA-21 in FOXP3 expression. Eur J Immunol 2009,
39(6):1608-1618.
159. Mallory AC, Dugas DV, Bartel DP, Bartel B: MicroRNA regulation of NAC-domain
targets is required for proper formation and separation of adjacent embryonic,
vegetative, and floral organs. Curr Biol 2004, 14(12):1035-1046.
160. Zhang HM, Kuang S, Xiong X, Gao T, Liu C, Guo AY: Transcription factor and
microRNA co-regulatory loops: important regulatory motifs in biological processes
and diseases. Brief Bioinform 2013.
161. Michon F: Tooth evolution and dental defects: from genetic regulation network to
micro-RNA fine-tuning. Birth Defects Res A Clin Mol Teratol 2011, 91(8):763-769.
162. Shalgi R, Lieber D, Oren M, Pilpel Y: Global and local architecture of the
mammalian microRNA-transcription factor regulatory network. PLoS Comput Biol
2007, 3(7):e131.

140
163. Tsang J, Zhu J, van Oudenaarden A: MicroRNA-mediated feedback and feedforward
loops are recurrent network motifs in mammals. Mol Cell 2007, 26(5):753-767.
164. Alper SL, Sharma AK: The SLC26 gene family of anion transporters and channels.
Mol Aspects Med 2013, 34(2-3):494-515.
165. Dawson PA, Russell CS, Lee S, McLeay SC, van Dongen JM, Cowley DM, Clarke LA,
Markovich D: Urolithiasis and hepatotoxicity are linked to the anion transporter
Sat1 in mice. J Clin Invest 2010, 120(3):706-712.
166. Snead CM, Smith SM, Sadeghein N, Lacruz RS, Hu P, Kurtz I, Paine ML: Identification
of a pH-responsive DNA region upstream of the transcription start site of human
NBCe1-B. Eur J Oral Sci 2011, 119 Suppl 1:136-141.
167. Toyosawa S, Ogawa Y, Inagaki T, Ijuhin N: Immunohistochemical localization of
carbonic anhydrase isozyme II in rat incisor epithelial cells at various stages of
amelogenesis. Cell Tissue Res 1996, 285(2):217-225.
168. Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time
quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001, 25(4):402-
408.
169. Schmittgen TD, Livak KJ: Analyzing real-time PCR data by the comparative C(T)
method. Nat Protoc 2008, 3(6):1101-1108.
170. Hu JC, Simmer JP: Developmental biology and genetics of dental malformations.
Orthod Craniofac Res 2007, 10(2):45-52.
171. McKee MD, Yadav MC, Foster BL, Somerman MJ, Farquharson C, Millan JL:
Compounded PHOSPHO1/ALPL deficiencies reduce dentin mineralization. J Dent
Res 2013, 92(8):721-727.
172. Yadav MC, de Oliveira RC, Foster BL, Fong H, Cory E, Narisawa S, Sah RL, Somerman
M, Whyte MP, Millan JL: Enzyme replacement prevents enamel defects in
hypophosphatasia mice. J Bone Miner Res 2012, 27(8):1722-1734.
173. Takagi T, Ogasawara T, Tagami J, Akao M, Kuboki Y, Nagai N, LeGeros RZ: pH and
carbonate levels in developing enamel. Connect Tissue Res 1998, 38(1-4):181-187;
discussion 201-185.
174. De Duve C, Wattiaux R: Functions of lysosomes. Annu Rev Physiol 1966, 28:435-492.
175. Smith CE, Nanci A, Moffatt P: Evidence by signal peptide trap technology for the
expression of carbonic anhydrase 6 in rat incisor enamel organs. Eur J Oral Sci 2006,
114 Suppl 1:147-153; discussion 164-145, 380-141.
176. Bronckers AL, Guo J, Zandieh-Doulabi B, Bervoets TJ, Lyaruu DM, Li X, Wangemann
P, DenBesten P: Developmental expression of solute carrier family 26A member 4
(SLC26A4/pendrin) during amelogenesis in developing rodent teeth. Eur J Oral Sci
2011, 119 Suppl 1:185-192.
177. Huotari J, Helenius A: Endosome maturation. EMBO J 2011, 30(17):3481-3500.
178. Bertrand CA, Zhang R, Pilewski JM, Frizzell RA: SLC26A9 is a constitutively active,
CFTR-regulated anion conductance in human bronchial epithelia. J Gen Physiol
2009, 133(4):421-438.
179. Chavez JC, Hernandez-Gonzalez EO, Wertheimer E, Visconti PE, Darszon A, Trevino
CL: Participation of the Cl-/HCO(3)- exchangers SLC26A3 and SLC26A6, the Cl-
channel CFTR, and the regulatory factor SLC9A3R1 in mouse sperm capacitation.
Biol Reprod 2012, 86(1):1-14.

141
180. Hayashi H, Suruga K, Yamashita Y: Regulation of intestinal Cl-/HCO3- exchanger
SLC26A3 by intracellular pH. Am J Physiol Cell Physiol 2009, 296(6):C1279-1290.
181. Hihnala S, Kujala M, Toppari J, Kere J, Holmberg C, Hoglund P: Expression of
SLC26A3, CFTR and NHE3 in the human male reproductive tract: role in male
subfertility caused by congenital chloride diarrhoea. Mol Hum Reprod 2006,
12(2):107-111.
182. Homma K, Miller KK, Anderson CT, Sengupta S, Du GG, Aguinaga S, Cheatham M,
Dallos P, Zheng J: Interaction between CFTR and prestin (SLC26A5). Biochim
Biophys Acta 2010, 1798(6):1029-1040.
183. Singh AK, Sjoblom M, Zheng W, Krabbenhoft A, Riederer B, Rausch B, Manns MP,
Soleimani M, Seidler U: CFTR and its key role in in vivo resting and luminal acid-
induced duodenal HCO3- secretion. Acta Physiol (Oxf) 2008, 193(4):357-365.
184. Wang Y, Soyombo AA, Shcheynikov N, Zeng W, Dorwart M, Marino CR, Thomas PJ,
Muallem S: Slc26a6 regulates CFTR activity in vivo to determine pancreatic duct
HCO3- secretion: relevance to cystic fibrosis. EMBO J 2006, 25(21):5049-5057.
185. Andrejewski N, Punnonen EL, Guhde G, Tanaka Y, Lullmann-Rauch R, Hartmann D,
von Figura K, Saftig P: Normal lysosomal morphology and function in LAMP-1-
deficient mice. J Biol Chem 1999, 274(18):12692-12701.
186. Sun X, Soleimani M, Petrovic S: Decreased expression of Slc26a4 (Pendrin) and
Slc26a7 in the kidneys of carbonic anhydrase II-deficient mice. Cell Physiol Biochem
2008, 21(1-3):95-108.

Abstract (if available)

Abstract Amelogenesis is the developmental process of dental enamel formation. Amelogenesis involves two major functional stages—secretory and maturation. The transition of ameloblasts from secretory to maturation stage, which is characterized by both morphological and functional changes, results in the formation of mature enamel with ordered crystallite structures. Although researcher today have a very clear idea of the molecular activities that define secretory‐stage amelogenesis, the molecular events that define enamel maturation remain understudied. ❧ In chapter two, we set out to investigate the potential role of miRNA regulation in maturation‐stage tooth development. We conducted genome‐wide miRNA and mRNA transcript expression profiling analyses of secretory‐stage and maturation‐stage enamel organs obtained from rat incisors. We identified a group of stage‐specific miRNAs and identified candidate gene targets based on bioinformatic prediction. Two maturation‐stage‐related genes, Lamp1 and Tfrc, were verified by luciferase reporter assay to be the target genes of miRNA regulators. The results indicated a dynamic expression pattern of miRNAs during the transition from secretory‐stage to maturation‐stage enamel mineralization, and suggest that miRNAs can influence key processes of enamel maturation. ❧ Based on the bioinformatic analyses in chapter two, Slc26 gene family were identified to be involved in the pH regulation process during enamel maturation. Among the 11 members in Slc26 gene family, Slc26a1, Slc26a6 and Slc26a7 are the only members whose transcripts are significantly up-regulated during maturation‐stage enamel formation when compared to secretory‐stage. In chapter three, we investigated the function roles of Slc26a1, Slc26a6 and Slc26a7 during enamel maturation. We conducted quantitative real‐time PCR and Western blot analyses to show that Slc26a1, Slc26a6 and Slc26a7 are all significantly up‐regulated at maturation‐stage (relative to secretory‐stage) at both miRNA and protein levels. The subsequent immunolocalization assays showed that in the maturation‐stage ameloblasts the gene products of Slc26a1, Slc26a6 and Slc26a7 localize to the apical region of cytoplasmic membrane, similar to the localization of pattern of Cftr in maturation‐stage ameloblasts. In addition, the distribution of Slc26a7 was also seen within the cytoplasmic/subapical region of ameloblast, presumably on the lysosomal membrane. From the protein complex pulled down using an antibody to Cftr

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A Western blot analysis and immunohistochemical study of 4-day old mice teeth over-expressing amelotin during amelogenesis

PDF

Integrative analysis of gene expression and phenotype data

PDF

Transcellular calcium transport in amelogenesis

PDF

Amelogenin-ameloblastin protein interaction and function in dental enamel formation

PDF

Cytomegalovirus induced amelogenesis imperfecta

PDF

Discovery of mature microRNA sequences within the protein- coding regions of global HIV-1 genomes: Predictions of novel mechanisms for viral infection and pathogenicity

PDF

Ameloblastin-protein interactions pattern enamel matrix

PDF

Enamelysin (MMP20) and Kallikrein 4 (KLK4) functions during enamel formation

PDF

Amelogenin peptides for biomimetic remineralization of enamel and dentin

PDF

Transcriptional co-activation functions of Msx homeodomain proteins by activating Hsf proteins

PDF

CCNL1, EWS, TFIP11, and RNA splicing

PDF

Role of MMP-20 in preventing protein occlusion in enamel apatite crystals: relevance in enamel biomineralization and biomimetics

PDF

Modulation of C/EBP alpha in the regulation of mouse amelogenin transcription during tooth formation

PDF

Leucine-rich amelogenin peptide induces osteogenesis in mouse embryonic stem cells

PDF

Functionalization of scaffolds with anti-BMP-2 antibody: role in antibody mediated osseous regeneration (AMOR)

PDF

Alk5 mediated TGF-β signaling acts upstream of FGF10 to regulate the proliferation and maintenance of dental epithelial stem cells

PDF

Extracellular heat shock protein-90alpha (eHsp90α): mechanisms of secretion, quantitation and function

PDF

TLR8-transferred miR-192 acts as a tumor suppressor in neuroblastoma by inhibiting CTCF

PDF

Interaction of epigenetics and SMAD signaling in stem cells and diseases

PDF

The role of neuregulin receptors in cell differentiation and the response to inflammatory cytokines in the intestinal epithelium

Asset Metadata

Creator Yin, Kaifeng (author)

Core Title Global analysis of the molecular activities defining maturation-stage amelogenesis

School School of Dentistry

Degree Doctor of Philosophy

Degree Program Craniofacial Biology

Publication Date 12/18/2015

Defense Date 05/01/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag amelogenesis,enamel maturation,microRNA,OAI-PMH Harvest,pH regulation,Slc26as

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Paine, Michael L. (committee chair), Hacia, Joseph G. (committee member), Moradian-Oldak, Janet (committee member), Sameshima, Glenn T. (committee member), Snead, Malcolm L. (committee member)

Creator Email kaifengy@usc.edu,scarfair1986@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-572083

Unique identifier UC11299498

Identifier etd-YinKaifeng-3476.pdf (filename),usctheses-c3-572083 (legacy record id)

Legacy Identifier etd-YinKaifeng-3476.pdf

Dmrecord 572083

Document Type Dissertation

Format application/pdf (imt)

Rights Yin, Kaifeng

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA