Short Communication
Caries Res 2013;47:259–263 Received: May 4, 2012
DOI: 10.1159/000346134
Mineralisation of Developmentally
Hypomineralised Human Enamel in vitro
F.A. Crombie N.J. Cochrane D.J. Manton J.E.A. Palamara E.C. Reynolds
Melbourne Dental School, Oral Health Co-operative Research Centre, The University of Melbourne,
Parkville, Vict., Australia
Key Words
Casein phosphopeptide amorphous calcium fluoride
phosphate ⴢ Developmental enamel defects ⴢ
Mineralisation ⴢ Molar-incisor hypomineralisation
Abstract
Molar-incisor hypomineralisation (MIH) is a problematic and
costly condition. Caries remineralising agents are often rec-
ommended for MIH management despite the lack of evi-
dence that these lesions have the capacity for increasing
their mineral content. Following surface layer removal 8
NaOCl pre-treatment and 14-day exposure to a CPP-ACFP so-
lution at pH 5.5, MIH lesions were analysed using transverse
microradiography and polarised light microscopy. Lesions
were highly variable but treatment with the remineralising
solution increased mineral content (1,828 8 461 vol%
min ⴢ ␮m, %R = 17.7 8 5.7) and porosity decreased demon-
strating the proof of concept that the mineral content of de-
velopmentally hypomineralised enamel can be improved
after eruption. Copyright © 2013 S. Karger AG, Basel
Molar-incisor hypomineralisation (MIH) is a prob-
lematic condition for patients and clinicians alike [Weer-
heijm et al., 2001]. The decreased mineral content, in-
creased porosity, decreased hardness, defective micro-
© 2013 S. Karger AG, Basel
0008–6568/13/0473–0259$38.00/0
Fax +41 61 306 12 34
E-Mail karger@karger.com Accessible online at:
www.karger.com www.karger.com/cre
Accepted after revision: November 26, 2012
Published online: January 29, 2013
structure, increased carbonate content, poor bonding
properties and pulpal changes associated with MIH can
manifest as extreme sensitivity, including ineffective lo-
cal analgesia, increased caries susceptibility, physical
breakdown during normal function and poor restorative
outcomes [Leppaniemi et al., 2001; Crombie, 2011]. The
ability to manage MIH with medical rather than surgical
intervention could provide benefit in terms of patient
comfort, increased caries resistance and preservation of
tooth structure, which are particularly important given
the accelerated rate at which affected teeth move through
the restoration cycle. Although caries and MIH both rep-
resent hypomineralised enamel conditions, MIH is a de-
velopmental rather than an acquired condition and the
innate capacity of the tissue to take up additional min-
eral may be different. In particular the enamel protein
content in MIH is abnormally high. These have been
identified as primarily serum proteins [Farah et al., 2010;
Mangum et al., 2010], including albumin, which has
been postulated as an inhibitor of apatite crystal growth
[Robinson et al., 1998]. Therefore deproteination of these
lesions may be useful to improve their response to min-
eralisation treatments [Robinson et al., 1990]. MIH-af-
fected enamel has also been shown to have higher car-
bonate content than sound enamel [Crombie, 2011]. As-
suming the carbonate is present as carbonated apatite
then this would increase the relative solubility of the
tooth mineral such that deposition of more stable hy-
Dr. Felicity Crombie
Melbourne Dental School
Faculty of Medicine Dentistry and Health Sciences, The University of Melbourne
720 Swanston Street, Melbourne, VIC 3010 (Australia)
E-Mail fcrombie @ unimelb.edu.au
droxyapatite and fluorapatite phases during treatment
would be beneficial.
The properties of MIH-affected enamel are highly
variable, however, common features include the full
thickness of the enamel being affected and a narrow sur-
face layer of reduced porosity and increased hardness
and mineral content [Crombie, 2011]. Promisingly, this
layer may indicate that natural post-eruptive repair of
the tissue is possible following exposure to favourable
conditions. Conversely though, this same surface layer
may complicate achieving further mineralisation chang-
es: a similar more mineralised surface layer is observed
in natural carious lesions and is thought to act as a dif-
fusion barrier to deeper remineralisation [ten Cate and
Arends, 1978; Cochrane et al., 2012]. Many in vitro rem-
ineralisation studies use shallow (^100 ␮m), artificially
created subsurface lesions on polished enamel in order
to reduce lesion variability and reduce interference by
the surface layer [Larsen and Fejerskov, 1989; Cochrane
et al., 2008].
The aim of this study was to determine if developmen-
tally hypomineralised enamel of MIH is capable of im-
provement.
Materials and Methods
Human first permanent molars diagnosed clinically as affect-
ed by MIH were collected with ethics approval from the Human
Research Ethics Committee of the University of Melbourne. Six
teeth were sterilised in 10% neutral phosphate-buffered formalin
for 2 weeks, then rinsed with distilled water and kept in a humid-
ified environment until use. Teeth were decoronated and lesions
randomly assigned to two experimental groups (n = 3). Lesions
were abraded with wetted coarse, medium, fine and superfine
Sof-LexTM discs (3M ESPE, USA), using light finger pressure for
10–15 s each removing 50–75 ␮m of enamel to eliminate the sur-
face layer prior to application of an acid-resistant nail varnish over
half of the hypomineralised lesion to ensure one half was untreat-
ed (control half) and the other half exposed to the solution (test
half). One group was subsequently pre-treated by irrigation with
0.95% w/v NaOCl (Milton solution, Milton, UK) for 2 min fol-
lowed by distilled water for 2 min. Solutions of 1% w/v casein
phosphopeptide-amorphous calcium fluoride phosphate (CPP-
ACFP) were prepared at pH 5.5 with 0.05% (w/v) sodium azide to
prevent microbial contamination [Mayne et al., 2011]. Teeth were
suspended in 2 ml of the remineralising solution at 37 ° C for 14
days, with a daily change of solution.
The teeth were embedded in methyl-methacrylate resin (Pala-
dur, Kulzer, Germany), 600-␮m sections cut using a diamond-
embedded, water-cooled saw (Minitom, Struers, Denmark) and
lapped to 100 ␮m using a RotoPol-21/RotoForce 4 lapping instru-
ment (Struers) with 1,200- and 2,400-grit paper. Samples were
radiographed beside an aluminum step wedge using Micro-
chrome High Resolution glass plates (Type 1A; Microchrome
260 Caries Res 2013;47:259–263
Technology, San Jose, Calif., USA) and nickel-filtered copper K␣
radiation as described previously [Mayne et al., 2011]. Radio-
graphic images produced were examined under transmitted light
microscopy and images captured using Image Pro Plus (Media
Cybernetics, USA) and analysed using Image ToolTM (UTHSCSA,
USA). Section thickness was measured and the equation of Ang-
mar et al. [1963] used to determine mineral content (vol% min).
For each test and control side of a lesion six line profiles were
taken either side of the interface (fig. 1).
Assuming that sound enamel is 86.2 vol% min, the area be-
tween the control (⌬Zc) and test (⌬Zt) enamel profiles (absolute
change) was calculated by trapezoidal integration to the point of
convergence for all lesions. Percentage mineralisation increase
(%R) of the treated hypomineralised enamel was calculated using
[(⌬Zc – ⌬Zt)/⌬Zc] ! 100 and results were also expressed as ab-
solute mineral content change. Predominantly the observed
changes occurred within the first 300 ␮m so the mineral change
at 50-␮m increments from the surface to a depth of 300 ␮m was
determined.
Following microradiography samples were examined under a
polarised light microscope using air, water and Thoulet’s solu-
tions with refractive indices of 1.41 and 1.47 as imbibing media.
A Student t test comparing the mineral content of test and con-
trol sides was undertaken for each sample.
Results
The initial lesions, as represented by the control halves,
were found to be variable in terms of both depth and min-
eral content (table 1) and with porosity varying between
10 and 125%. Lesions in the surface removal + NaOCl
group had nearly twice the initial integrated mineral con-
tent deficit (⌬Zc) compared to those for which only the
surface layer had been removed. No area of increased
mineral content and decreased porosity could be seen
over the test or control sides of the lesion confirming that
the surface layer had been completely removed. Porosity
for the test half of the lesion was reduced by at least 5%
and up to 25%, with a final porosity of ^5% achieved for
all but part of one lesion. Changes to the test side were
more sharply delineated under polarised light than trans-
verse microradiograph analysis, but all twelve scan lines
did eventually converge, indicating the unexposed lesion
half was a valid control (fig. 1).
Mineral content changes across each lesion are sum-
marised in table 1. Due to the differences in ⌬Zc the %R
values for the NaOCl group could be comparable or low-
er despite often larger absolute mineral gains. There was
a trend for NaOCl treatment to more consistently sustain
mineral augmentation levels in the outer 200 ␮m, but one
sample from the surface layer removal only group main-
tained higher levels beyond 200 ␮m. On average over the
full depth of change the greatest absolute mineral gains
Crombie/Cochrane/Manton/Palamara/
Reynolds
 80

Mineral content (vol% min)

70

60

50
Control lines
40
Test lines
30
0 50 100 150 200 250 300 350
Distance (μm)

80

Mineral content (vol% min)

Control lines
70
Test lines
60

50

40

30
0 50 100 150 200 250 300 350
Distance (μm)

Fig. 1. Polarised light images of MIH lesions after exposure to
CPP-ACFP solutions in vitro viewed in water (refractive index
1.33), demonstrating the test area (black bracket) where the poros-
ity of much of the lesion has reduced to ^5%, and the control area
(white bracket) where porosity exceeds 5%. Transverse microra-
diograph images of MIH lesions after exposure to CPP-ACFP so-
lutions in vitro, with lines marking scans of test (black) and con-
trol (white) sides for mineral content analysis, and corresponding
graph with arrows indicating the area of altered mineral content
used to calculate %R and absolute changes (white) and the conver-
gence of the mineral content profiles demonstrating equivalence
of test and control sides (black). The top row represents a polished
lesion without NaOCl pre-treatment, the bottom row represents
a polished lesion with NaOCl pre-treatment.

were observed in the NaOCl group. The mineral content
of the test side was significantly higher than that of the
control side for all lesions (p ! 0.001).
Discussion
The effective management of MIH-affected teeth is an
ongoing issue for the majority of clinicians and many of
the published guidelines and recommendations for treat-
ment include the application of remineralising agents
such as high concentration fluorides and CPP-ACP/CPP-
ACFP [William et al., 2006; Crombie et al., 2008]. The
present study establishes proof-of-concept evidence that
developmentally hypomineralised enamel can be im-
proved both in terms of increasing the mineral content
and reducing the porosity of the lesion using gold stan-
dard techniques.
In this study the lesion surface layer was physically re-
moved as preliminary work found lesions failed to re-
spond to mineralisation treatments when it was present.
This may be similar to the natural carious lesion surface
layer that can restrict access to deeper tissue layers for in-
ward diffusion of ions or materials [Larsen and Pearce,
1992; Meyer-Lueckel et al., 2007; Cochrane et al., 2012].
As the physical removal of the surface layer may be too
destructive for clinical application minimally invasive al-
ternatives for altering this layer need to be explored or
earliest possible intervention advocated when feasible be-
fore the layer has formed. Additionally, protein in MIH is
likely to be a problem, and so a deproteination treatment
was trialled, which appeared to show some additional
benefit. This was despite the teeth having undergone for-
malin fixation, during which protein cross-linking is ex-
pected which is likely to reduce the subsequent effective-
ness of deproteination agents such as NaOCl. The use of
alternative disinfection/sterilisation procedures that do
not cross-link proteins may even lead to a better outcome.
On the other hand the lesions in this group had lower
mineral content and enamel permeability may be a con-
founding factor in facilitating improved mineral uptake.
The depth to which the effects of deproteination agents
extend should also be examined as they may prove to be
limited to the surface.

Mineralisation of Hypomineralised Caries Res 2013;47:259–263 261

Enamel in vitro
Table 1. Baseline characteristics of the MIH lesions studied and their changes in mineral content after exposure to CPP-ACFP solu-
tions in vitro across their full depth and across 50-␮m increments to 300 ␮m

Sample Lesion depth (depth Measurement Depth increment, ␮m

No. of mineral change)
0–50 50–100 100–150 150–200 200–250 250–300 0–depth change

Tooth preparation: surface layer removed only

1 1,124 (580) ⌬Zc 1,140 1,183 1,010 1,231 1,203 1,192 11,220
⌬Zc – ⌬Zt 647 309 179 4 198 64 1,791
%R 56.7 26.1 17.7 0.3 16.5 5.4 16.0
2 863 (190) ⌬Zc 1,140 1,280 1,325 4,814
⌬Zc – ⌬Zt 305 444 231 – – – 1,057
%R 26.7 34.7 17.5 22.0
3 921 (340) ⌬Zc 1,101 1,071 969 1,160 993 1,118 7,834
⌬Zc – ⌬Zt 667 48 338 177 459 236 2,038
%R 60.6 4.4 34.9 15.2 46.2 21.1 26.0
Average 969 (357) ⌬Zc 1,127 1,178 1,101 1,196 1,098 1,155 7,956
⌬Zc – ⌬Zt 540 267 249 91 329 150 1,629
%R 48.0 21.7 23.4 7.8 31.4 13.3 21.3

Tooth preparation: surface layer removal + NaOCl

4 981 (315) ⌬Zc 2,328 1,862 2,216 1,961 1,951 1,980 12,749
⌬Zc – ⌬Zt 412 298 521 373 56 45 1,642
%R 17.7 16.0 23.5 19.0 2.8 2.3 12.9
5 867 (380) ⌬Zc 1,675 1,666 1,768 1,746 1,785 1,756 13,024
⌬Zc – ⌬Zt 545 744 413 494 37 189 2,417
%R 32.5 44.6 23.4 28.3 2.1 10.8 18.6
6 781 (465) ⌬Zc 2,167 1,954 2,167 1,998 2,027 2,200 18,783
⌬Zc – ⌬Zt 448 570 315 292 –6 124 2,026
%R 20.7 29.2 14.6 14.6 –0.3 5.6 10.8
Average 876 (387) ⌬Zc 2,057 1,827 2,050 1,902 1,921 1,979 14,852
⌬Zc – ⌬Zt 468 537 416 386 29 119 2,028
%R 23.6 29.9 20.5 20.6 1.5 6.2 14.1

All 923 (372) ⌬Zc 1,592 1,503 1,576 1,619 1,592 1,649 11,404
⌬Zc – ⌬Zt 504 402 333 268 149 132 1,828
%R 35.8 25.8 21.9 15.5 13.5 9.0 17.7

CPP-ACFP was chosen as the mineralisation agent as
this has been shown in previous studies to promote high
levels of mineral return. In a similar in vitro study Mayne
et al. [2011] found 43% mineral return to artificial carious
lesions that were approximately 400 ␮m deep after a 30-
day exposure. The degree of mineral uptake in the cur-
rent study did not reach these levels, however, natural
carious lesions are less amenable to repair than artificial
carious lesions and developmental hypomineralisations
may be even more difficult to mineralise. While improv-
ing the quantity of mineral is important, so is the type of
mineral deposited: agents that encourage deposition of
the more stable fluoridated apatites would be most ad-
vantageous in increasing resistance to future acid disso-
lution, but considering the high carbonate content of
MIH-affected enamel covering the existing crystals even
with hydroxyapatite is likely to be beneficial. The ob-
served reduction in porosity may explain the anecdotal
improvement in sensitivity after use of commercially
available crèmes with CPP-ACP, by decreasing thermal/
tactile stimulation as well as reducing pathways for bacte-
rial ingress and associated pulpal inflammatory changes

262 Caries Res 2013;47:259–263 Crombie/Cochrane/Manton/Palamara/

Reynolds
[Fagrell et al., 2008]. Similarly, decreasing the surface po-
rosity, a putative contributor to bacterial adhesion, may,
along with an improvement in the mineral type and con-
tent, reduce caries risk.
This in vitro study proves that hypomineralised enam-
el has a substantial ability to increase its mineral content.
Further work is needed to determine to what extent this
is possible in vivo and what clinical protocols can maxi-
mise the likelihood and extent of this change. These in
vitro results are consistent with those of a clinical trial
which has suggested application of a commercial product
containing 10% CPP-ACP provided some improvement
in MIH enamel [Baroni and Marchionni, 2011]. The abil-
ity to improve the properties of MIH lesions after erup-
tion supports suggestions that the enamel structure is
fundamentally adequate, and that the condition repre-
sents a failure of the maturation processes. The biggest
challenges in accomplishing maximal improvements
clinically are likely to be achieving changes through the
full enamel thickness and altering the surface layer to im-
prove diffusion without sacrificing this relatively hy-
permineralised enamel. If lesions can be ‘matured’ to the
level of healthy enamel it may become possible to manage
MIH without surgical intervention and avoid the high
biological, social, psychological and financial costs cur-
rently associated with the condition.
In summary, significant post-eruptive ‘maturation’ of
MIH-affected enamel was possible in this in vitro study.
Now that proof of concept has been established further
research is needed to determine whether this has clinical
applicability.
Acknowledgements
The assistance of Glenn Walker, Coralie Reynolds, Yi Yuan,
Peiyan Shen and Fan Cai is gratefully acknowledged. This study
was supported by Dentsply Australia Pty Ltd. and Felicity Crom-
bie was a recipient of a University of Melbourne faculty scholar-
ship. Denstply Australia Pty Ltd. had no role in the study design,
data collection and analysis, decision to publish, or preparation of
the manuscript.
Disclosure Statement
The authors declare no conflict of interest.

References
Angmar B, Carlstrom D, Glas JE: Studies on the
ultrastructure of dental enamel. IV. The min-
eralization of normal human enamel. J Ul-
trastruct Res 1963;8:12–23.
Baroni C, Marchionni S: MIH supplementation
strategies: prospective clinical and labora-
tory trial. J Dent Res 2011;90:371–376.
Cochrane NJ, Anderson P, Davis GR, Adams
GG, Stacey MA, Reynolds EC: An X-ray mi-
crotomographic study of natural white-spot
enamel lesions. J Dent Res 2012;91:185–191.
Cochrane NJ, Saranathan S, Cai F, Cross KJ,
Reynolds EC: Enamel subsurface lesion rem-
ineralisation with casein phosphopeptide
stabilised solutions of calcium, phosphate
and fluoride. Caries Res 2008;42:88–97.
Crombie F: An Investigation into Developmen-
tally Hypomineralised Enamel in First Per-
manent Molar Teeth; PhD thesis Melbourne
Dental School, University of Melbourne,
2011.
Crombie FA, Manton DJ, Weerheijm KL, Kilpat-
rick NM: Molar incisor hypomineralization:
a survey of members of the Australian and
New Zealand Society of Paediatric Dentistry.
Aust Dent J 2008;53:160–166.
Fagrell TG, Lingstrom P, Olsson S, Steiniger F,
Noren JG: Bacterial invasion of dentinal tu-
bules beneath apparently intact but hypo-
mineralized enamel in molar teeth with mo-
lar incisor hypomineralization. Int J Paedi-
atr Dent 2008;18:333–340.
Farah RA, Monk BC, Swain MV, Drummond
BK: Protein content of molar-incisor hypo-
mineralisation enamel. J Dent 2010; 38: 591–
596.
Larsen MJ, Fejerskov O: Chemical and structur-
al challenges in remineralization of dental
enamel lesions. Scand J Dent Res 1989; 97:
285–296.
Larsen MJ, Pearce EI: Some notes on the diffu-
sion of acidic and alkaline agents into natu-
ral human caries lesions in vitro. Arch Oral
Biol 1992; 37:411–416.
Leppaniemi A, Lukinmaa PL, Alaluusua S: Non-
fluoride hypomineralizations in the perma-
nent first molars and their impact on the
treatment need. Caries Res 2001;35:36–40.
Mangum JE, Crombie FA, Kilpatrick N, Manton
DJ, Hubbard MJ: Surface integrity governs
the proteome of hypomineralized enamel. J
Dent Res 2010;89:1160–1165.
Mayne RJ, Cochrane NJ, Cai F, Woods MG,
Reynolds EC: In-vitro study of the effect of
casein phosphopeptide amorphous calcium
fluoride phosphate on iatrogenic damage to
enamel during orthodontic adhesive remov-
al. Am J Orthod Dentofacial Orthop 2011;
139:e543–e551.
Meyer-Lueckel H, Paris S, Kielbassa AM: Surface
layer erosion of natural caries lesions with
phosphoric and hydrochloric acid gels in
preparation for resin infiltration. Caries Res
2007;41:223–230.
Robinson C, Hallsworth AS, Shore RC, Kirkham
J: Effect of surface zone deproteinisation on
the access of mineral ions into subsurface
carious lesions of human enamel. Caries Res
1990;24:226–230.
Robinson C, Shore RC, Bonass WA, Brookes SJ,
Boteva E, Kirkham J: Identification of human
serum albumin in human caries lesions of
enamel: the role of putative inhibitors of re-
mineralisation. Caries Res 1998;32:193–199.
ten Cate JM, Arends J: Remineralization of arti-
ficial enamel lesions in vitro. Caries Res
1978;12:213–222.
Weerheijm KL, Jalevik B, Alaluusua S: Molar-
incisor hypomineralisation. Caries Res 2001;
35:390–391.
William V, Messer LB, Burrow MF: Molar inci-
sor hypomineralization: review and recom-
mendations for clinical management. Pedi-
atr Dent 2006;28:224–232.

Mineralisation of Hypomineralised Caries Res 2013;47:259–263 263

Enamel in vitro