Content uploaded by Douglas Borchman
Author content
All content in this area was uploaded by Douglas Borchman on Apr 28, 2015
Content may be subject to copyright.
Short Communication
Ophthalmk
Ophthalmic
Res
2015;53:131
-
135
Received:
August
31,
2014
Research Accepted after revision: January 1,201 S
001: 10.1159/000371852
PUblished
online: February 14,2015
Pilot Study
of
the Influence
of
Eyeliner
Cosmetics on the Molecular Structure
of
Human Meibum
Morgan Huntera
Rahul
Bhola
b Marta
C.
Yapperta
Douglas
Borchman b
Dylan
Gerlach b
Departments
of
'Chemistry
and
bOphthalmology
and
Visual
Sciences,
University of Louisville, Louisville,
Ky.,
USA
KeyWords
Dry
eye·
Infrared spectroscopy' Lipids· Makeup· Meibum
Abstract
Background/Aims:
It
has been suggested that eye makeup
could interact with human meibum causing a decrease
in
the
stability of
the
tear
film.
The
aim
of this pilot study was
to
measure makeup-human meibum interactions
in
vitro.
Methods: Human meibum-makeup interactions were quan-
tified by measuring order-to-disorder
lipid
phase transitions
using infrared spectroscopy. Results: Makeup products
ex-
hibited lipid phase transition temperatures
that
were much
higher
than
those
for
meibum. One product increased the
lipid phase transition temperature
by
4.2Q( when combined
with human meibum causing a large increase (from 30 to
49%)
in
the
order of
the
meibum-lipid hydrocarbon chains
and Significantly decreased
the
minimum frequency, enthal-
py and entropy of the phase transition of human meibum.
Another eyeliner caused
no
significant
(p
< 0.05) change
in
the
phase transition parameters of human meibum. Conclu-
sion: Infrared spectroscopy
may
be used to measure interac-
tions between human meibum and makeup. One makeup
product increased
the
lipid
order (viscosity) which could
have adverse effects
on
tear
film
stability. Modern cosmetics
©
2015
S.
Karger
AG.
Basel
KAR..G
ER..
1~
0030-3747/15/0533-0131$39.50/0
are highly regulated
and
relatively safe
to
use; however,
it
could be beneficial
to
design makeup products
that
do
not
interact with meibum, especially since women have a higher
prevalence of dry eye symptoms. ©
2015
S.
Karger
AG,
Basel
Introduction
Women
and
men
have
painted
their
eyes since
prehis-
toric Paleolithic
and
neolithic times [lJ. Today,
mascara
is applied to the eyelashes to color
them.
Mineral,
linseed,
castor, eucalyptus, lanolin,
turpentine,
sesame oils
and
waxes such
as
paraffin,
carnauba
wax
or
beeswax are
commonly
mixed
with
iron
oxides for
color
[2J.
Reports
of
poisoning
in
Arabic cultures
from
the use
oflead-con-
taining black eye paste have
been
reported
[3J.
Modern
cosmetics are
highly
regulated
and
relatively safe
to
use
[2J.
A specular reflection device
indicated
that
moistur-
izers used
to
remove
cosmetics
appeared
as
oily floaters
or
oily plaques
on
the
surface
of
the
eye,
and
it has
been
suggested
that
eyeliners, eye
shadow
and
mascara
could
block
meibomian
glands,
contaminate
the
meibum
and
increase the debris levels
on
the superficial
lid
layer [4J.
Indeed, lipophilic substances applied to
the
lower
eyelid
RahuJ
Bhola.
MD
Kentucky
Lions
Eye
Center
301
E Muhammad Ali Blvd
E·Mail
karger@karger.com
Louisville,
KY
40202
(USA)
www.karger.com/ore E·Mail rObhoIOI@exchange.louisville.edu
skin are able to reach the inferior tear meniscus supracu-
taneously
and
mix
with the tear film lipid layer
[5,
6].
Maintenance
of
meibum
structure may be critical to nor-
mal tear film function [7].
Dry
eye
is
a multifactorial disease that
is
associated
with elevated tear osmolarity and ocular irritation.
It
is
one
of
the
most
common
causes for patients to consult
with an eye care professional
[8].
It
is
more
prevalent in
women [9,
10]
and, as suggested
but
not
tested (4], could
be exacerbated by eye makeup.
Materials and Methods
The work described has been carried out in accordance with the
Code of Ethics
of
the World
Medical
Association (Declaration of
Helsinki) for experiments involving humans. In this study,
two
pop-
ular cosmetics were mixed with meibum
to
test
if
they altered the
molecular structure
of
human melbum. Meibum
was
collected and
pooled
[11]
from 7 donors present in Professor Borchman's labora-
tory at the time
of
the
study:
MA2l, FC2l,
FB21,
MCS9,
MC22,
FA42
and
FA19,
where M
is
male, F
is
female,
A
is
Asian, C
is
Cau-
casian, B
is
black and the number
is
the
age
of the donor in
years.
The donors did not complain
of
dry
eye
symptoms or have any ocu-
lar
diseases.
Written informed consent
was
obtained from
all
do-
nors. Protocols and procedures
were
reviewed
by
the institutional
review boards
of
the University
of
Louisville.
Stock
(1
mg/mI) CHCl3
solutions
of
meibum alone, and meibum mixed with
25
weight per-
cent makeup were dried onto a
AgCI
window and lyophilized
for
infrared spectroscopic
analysis.
Two
popular brands of eyeliner
makeup were studied: L'Oreal
(Clichy,
Hauts-de-Seine, France) In-
fallible
pencil eyeliner and water-based
Revlon
(New
York,
N.Y.,
USA)
ColorStay liquid eyeliner.
The
L'Oreal product contains over
20
ingredients including cyclopentasiloxane, hydrogenated olive oil
esters, candelilla
wax,
microcristailine
wax,
sodium hyaluronate
as
well
as
iron oxides, titanium dioxide and ferric ferrocyanide. The
Revlon
product
also
contains over
20
ingredients such
as
water,
oc-
tylacrylamide copolymer,
Aloe
barbadensis
leaf juice, imidazolidinyl
urea, methylparaben, ethylparaben, iron oxides and Black
2.
Infrared spectroscopy
was
used to measure and quantify the
phase transitions
of
lipids
[11,
12].
The lipid hydrocarbon chain
conformational order may
be
evaluated in terms of the amount of
CH2 trans- and gauche rotamers. Conformation
is
how molecules
are arranged in space. The methylene moieties may arrange
as
ga
uche rotamers, prevalent in disordered lipids like olive
oil,
which
is
fluid at room temperature. The methylene moieties may also
ar-
range
as
trans-rotamers. more abundant in ordered lipids like but-
ter, which
is
solid at room temperature.
Meiburn and meibum-eyeliner samples were mixed in deute-
rium oxide and deuterated chloroform solvents for nuclear
mag-
netic resonance
(NlvIR)
analysis in a Varian VNMR 700-MHz
NMR spectrometer (Varian, Lexington, Mass.,
USA)
eqUipped
with a S-mm
lH[
13
C/1
SN] 13C-enhanced cold probe (Palo
Alto,
Calif.,
USA).
Data are presented
as
the
average
± the standard error
of
the
mean unless indicated otherwise. Differences were compared using
Student's t test, and a p <
0.05
was
considered statistically significant.
OphthahnlcRes 2015;53:131-135
DOl: 10.1159/000371852
132
Results
The
amount
of
makeup applied to one eyelid was esti-
mated graVimetrically by quantifying
the
amount
of
makeup applied
onto
a
30
x 1
mm
area
of
paper. The av-
erage and standard deviation
of
the
amount
of
L'Oreal
and Revlon products applied
was
0.0010 ± 0.0002
and
0.00305 ± 0.0005
g.
respectively.
The phase transitions for meibum were reproducible,
and the phase transition parameters were similar
to
those
obtained previously for normal
human
meibum
(fig.
la)
[11]. The L'Oreal product Significantly increased
(p
<
0.05) the phase transition temperature
of
meibum
lipid
by 4.2°C (fig. lb, table 1). Because the lipid phase transi-
tion temperature
of
human meibum
is
near the physio-
logical temperature
of
the surface
of
the eye, the small
increase
in
lipid phase transition temperature caused a
large increase, from
30
to
49%,
in the order
or
stiffness
of
the meibum lipid hydrocarbon chains.
The
L'Oreal prod-
uct also Significantly
(p
< 0.05) decreased the
minimum
frequency, enthalpy and entropy
of
the phase transition
of
human
meibum (table
1).
The Revlon
product
caused
no
Significant
(p
> 0.05) change
in
the phase transition
parameters
of
human
meibum
(fig.
Ie, table
1).
Because
of
the high optical opacity
of
the Revlon prod-
uct, very little infrared light
was
transmitted making it tech-
nically difficult to measure
an
exact phase transition of
the
Revlon product alone. Regardless
of
the technical difficulty,
we
observed that both the L'Oreal and Revlon products
alone exhibited phase transitions that were
at
much
higher
phase transition temperatures than for
meibum
(fig.
Id,
table
1).
Broadening
of
the
-CR
r methylene resonances
near
1.24
ppm
[13]
confirmed the interaction
of
the L'Oreal
eyeliner with meibum
(fig.
2).
The Revlon eyeliner showed
little interaction with meibum supporting the infrared
spectroscopy phase transition data.
We
also studied the
in-
teraction
of
the wax paImityl oleate with
the
two makeup
products. PaImityl oleate
was
chosen because it has a phase
transition at a temperature which could be conveniently
measured,
13.1
±0.1
0c,
The phase transition increasedsig-
nificantly (p < 0.0001) to
14.08
±0.07 and 15.32 ± 0.07"C
when the paImityl oleate
was
mixed at
25%
by
weight with
the Revlon and L'Oreal products, respectively.
Discussion
We quantified meibum-makeup interactions
in
a 25%
by weight mixture. Concentrations
of
makeup
much
higher than this could overwhelm the infrared bands
of
Hunter/BholalYappert/Borchmanl
Gerlach
2,855
c
C\
2,854
:.2
u~
2,853
~
"'E
1;;
u
.!!
;:
2,852
~
u
- C
E
~
2,851
E
0"
a-~
2,850
I
'"
2,849
U
2,848
a
2,855
C\
c
2,854
:.2
u
~
ail
2,853
~1
.l:!
C
2.852
...
c
E
'"
2,851
Eg.
2,850
-2,850
a-~
a-~
N
:£'
2,
849
1 I 2,849
u u
2,848
I I I I
2.,848
0
20
40
60
80
10
20
30
40
50
60
70
80
90
c
Temperature
(CCl
d
Temperature
(0C)
2,855
en
-_
....
_--
...
-
0
c:.
2.854
:.2
u~
2.853
~t-E
iii~
2..852
.~
Cj'
Qj
~
2.851
E ::l
E
0"
2.850
»'"
"'-l=
I
'"
2,849
u 2,848
0
20
40
60
80
0
20
40
60
80
Temperature
(CCl
b
Temperature
(0C)
2.855
en
--------
c 2.854
:.2
l::l""
2.853
""
•
~
§ 2.852
.~
a:
OJ
~
2,851
E ::l
E
0"
Fig.
1.
Lipid
phase
transition
curves
measured using infrared
spec-
troscopy. The lower the
CH
2 symmetric stretching frequency, the
more ordered
the
wax
hydrocarbon chains.
Lines
are
the
fit
of the
data
to
equation 1 in
Borchman
et
al.
[12].
Symbols
are
for
exper-
iments conducted on different
days.
a Human meibum. b Human
meibum plus L'Oreal
Infallible
pencil eyeliner
(25%
by weight).
Table
1.
Lipid phase transition parameters
- - - = Curve
fit
of meibum-only data from
a.
c Human meibum
plus water-based Revlon ColorStay liquid eyeliner
(25%
byweight).
- - -
:::
Curve
fit
of meibum-only data from a. d Comparison of
L'Oreal Infallible pencil eyeliner
(.)
and water-based Revlon
ColorStay liquid eyeliner
(_).
makeup alone.
Phase transition
Meibutl1
Meiburn+
Significance,
Meibum +
Significan~e,
L'Oreal
parameter L'Oreal p
Revlon
p
Minimum frequency,
cm-!
2,850.13±0.16
2,849.59±0.06
0.002*
2,849.84±0.05
0.09
2,848.3±0.1
2,849.55±0.05
Maximum frequency,
cm-!
2,854.08±0.21
2,B53.8±0.1
0.23
2,B53.84±0.OB
0.29
2,854.B±0.2
Cooperativity
B.l±1.3
6.3±1.3
0.33
7.5±0.5
0.67
B.7±0.6
Phase transition
temperature,
°C
2B.9±0.6
33.l±0.5
<0.0001*
27.7±0.3
0.08
57.7±0.6
>44.3
Lipid
order at
33.4°C,
% trans-rotamers
30.0±O.7
49.3±0.7
<0.0001*
31.4±0.3
0.07
96±1
7B.9±0.7
Enthalpy, kcallmol
196±2
120±2
<0.0001*
175±2
<0.0001'"
19l±6
Entropy, kca1/mol/degree
0.650±0.008
0.394±0.005
>0.0001*
0.5B4±0.00B
>0.0001'"
0.58±0.02
Eyeliners: L'Oreal
Infallible
pencil eyeliner and water-based
Revlon
ColorStay liquid eyeliner.
'"
p <
0,05:
statistically significant, when
makeup
was
mixed with meibum
(25%
by
weight). Data
are
the
average
± the standard deviation of the mean. p values were determined
using Student's t test.
Makeup-Meibum Interactions Ophthalmic Res 2015;53:131-135
133
DOl:
10.1159/000371852
I I I I I I I
1.26
1.2.5
1.24
1.23
1.22
1.21
1.20
Parts
per million
Fig.
2.
lH-NMR spectra
of
human meibum (bottom) and human
meibum
with
L'Oreal Infallible pencil eyeliner (toPi
25%
by
weight).
meibum, interfering with the quantification of meibum
lipid conformation. Concentrations
of
makeup much
lower
than
the
3:
1 (meibum:makeup) ratio we used (25%)
would
not
be physiologically relevant since
we
calculate
that the makeup:meibum ratio on the eyelid surface
is
much
higher, 1:10 (91%) to 1:70 (99%), based
on
the
amount
of
casual meibum on the surface
of
the eyelid
[14].
Infrared spectroscopy
was
used in the current study to
measure lipid-makeup interactions with the hydrocarbon
chains. The conformation
of
hydrocarbon chains
of
hu-
man
meibum lipids is in between that
of
fluid-like lipids
such
as
olive oil and/or solid-like lipids such
as
butter
[15]. Fluid lipids are generally
not
completely liquid, but
interact with one another and are said to be
in
a gel phase.
Solid lipids are
not
completely solid and contain some
mobility so they are said to be in a liquid crystalline phase.
We found that one makeup product caused the hydrocar-
bon chains
of
meibum to become more 'solid' and in a
gel
phase where the hydrocarbons pack tightly together and
van der Waals' interactions between hydrocarbon chains
are maximal.
When
the lipids were more fluid
and
in a
liquid crystalline phase, gauche rotamers caused the hy-
drocarbon chains to bend. The surface area
of
fluid lipids
was larger
than
that
of
ordered lipids, and
van
der Waals'
interaction between chains was minimal.
In
simpler
Ophthalmic
Res
2015;53:131-135
134 001: 10.1159/000371852
terms, one makeup product caused the meibum to be-
come more viscous.
Meibum lipid hydrocarbon chains interact with each
other viilhydrophobic van der Waal's interactions
[11,
12].
Cations, which interact with lipids ionically, do not
cause a change in human meibum packing [16]. There-
fore,
we
speculate that the reason why the Revlon-based
product did not cause a change in meibum conformation
is
that the Revlon product
is
water based and insoluble in
solvents such
as
chloroform. Therefore, the hydrophilic
ingredients did not interact with the hydrophobic chains
of the meibum. The
L'Ore<tl
product contains hydropho-
bic moieties such
as
cydopentasiloxane, hydrogenated
olive oil esters, candelilla wax and microcristalline wax.
The acyl chain esters are
all
saturated
and
very ordered.
We found that saturation
is
the major factor that contrib-
utes to hydrocarbon chain order
[11,
12]
and that when a
saturated
wax
was
mixed with an unsaturated one, the
saturated
wax
disproportionately increased the phase
transition of the mixture by apprOximately
30
° C com-
pared with the saturated wax alone [11]. The mixing
of
the saturated waxes with meibum could easily account for
the small 3°C change in the phase transition temperature
observed when the L'Oreal eyeliner was mixed with
meibum
or
synthetic
wax.
Itis doubtful
that
the CH3 moi-
eties
of
cyclopentasiloxane interact with the CH2 hydro-
carbon groups
of
meibum.
We
used the synthetic
wax
palmityl oleate to study the
interaction with makeup. Waxes are the predominant lip-
id species
of
meibum
[7].
At
present
it
is
difficult
to
make
a mixture
of
synthetic lipids
as
a good model for meibum
because meibum contains over '30,000 molecular species'
[17]
of
lipid and the composition
is
complicated by
branched chains, unsaturation and hydroxylation
and
products such
as
(O-acyl)-w-hydroxy fatty acids that are
not available commercially
[7].
The wax palmityl oleate
may be used to screen products
as
we did. We found that
the L'Oreal product increased the phase transition tem-
perature
of
palmityl oleate more effectively than the Rev-
Ion product Similarly to
the
effect it had
on
meibum. Ul-
timately, because of the complexity
of
the
human
meibum
lipidome, it
is
important
to
use human meibum to test
makeup.
As
a pilot study, one must be careful to draw conclu-
sions about the potential functional results of altering the
phase transition parameters
of
human
meibum. The
phase transition temperature
of
meibum from donors
with dry
eye
due to meibomian gland dysfunction
is
high-
er than that of meibum from normal donors [11], simi-
larly to the lipid phase transition-rising effect
of
the
Hunter/Bhola/Yappert/Borchman/
Gerlach
L'Oreal product. Maintenance
of
the lipid phase transi- toms
[9,
10].
Eye
makeup could even be beneficial
if
in-
tion temperature close to the physiological temperature gredients that enhance tear film stability were included in
may be critical to the maintenance
of
a low evaporation the formulations.
rate [18]. Lipid order (viscosity
or
stiffness) has been re-
lated to the release
of
meibum from the meibomian glands
[19]. Makeup carried withmeibum onto the surfaceofthe
Acknowledgments
eye could also affect the surface properties
of
the lipid
This
work
was
supported
by
the
Kentucky
Lions
Eye
Founda-
layer
of
the tear film. There
is
about
33
times more tion and an unrestricted
grant
from
Research
to
Prevent
Blindness
meibum
lipid
on
the eyelid surface compared with the
Inc.
Morgan Hunter's
fellowship
with the Department of Chemis-
layer
of
meibum lipid on the surface
of
tears,
so
contam- try at the University of
Louisville
was
funded
by
the National
Sci-
ination
of
the tear film could be more critical than con-
ence
Foundation
Research
Experience
for
Undergraduates.
Dylan
Gerlach's
fellowship
was
funded
by
the
Nation Institute of Health,
tamination
of
the eyelid surface [14].
Eye
makeup
is
de- Kentucky
Biomedical
Research
Infrastructure Network grant.
signed to stick to the skin and is naturally hydrophobic.
Modern cosmetics are highly regulated and relatively safe
to use; however, it could
be
beneficial to design makeup
Disclosure
Statement
products that do
not
interact with meibum, especially
since women have a higher prevalence
of
dry
eye
symp- None of the authors had a conflict of interest.
References
Murube
I:
Ocular cosmetics
in
ancient times. 8 Schaumberg
DA,
Sullivan
DA,
Dana
MR:
Ep-
13
Borchman
D,
Yappert
MC,
Milliner
S,
Duran
Ocul Surf2013;1l:2-7. idemiology of
dry
eye
syndrome.
Adv
Exp
D,
Cox
GW,
Smith
RJ.
Bhola
R:
13C
and
IH
2 Murube
I:
Ocular cosmetics
in
modern
times.
Med
Bioi
2002;506:989-998.
NMRester
region
resonance
aSSignments
and
Ocul 5urf2013:11:60-64. 9
Pflugfelder
SC:
Prevalence,
burden,
and
phar-
the
composition
of
human infant and
chUd
3
De
Caluwe
IP:
Saturnine intoxication
from
macoeconomics of
dry
eye
disease.
Am
J
melbum.
Exp
Eye
Res
2013:112:151-159.
the long-term
use
of
kohl:
an
underestimated
Manag
Care
2008;14(suppl):S102-S106.
14
Chew
CKS,
Hykin
PG,
Janswljer
C,
Dikstein
cause
in
French-speaking countries (in
10
Schaumberg
DA,
Nichols
II, Papas
EB,
Tong
S,
Tiffany
1M.
Bron AI: The casual
level
of
French). I
Fr
OphtalmoI2009:32:459-463.
L,
Uchlno
M,
Nichols
KK:
The International
meibomlan
lipids
in
humans. Curr
Res
4 Guillon
IP:
Abnormal lipid
layers.
Observa-
Workshop
on
Meibomian Gland
Dysfunc-
1993:
12:255-259.
tion, differential diagnosis,
and
classification.
tion:
report of
the
subcommittee
on
the
epi-
15
Borchman
D:
From the bench to the
bedside.
Adv
Exp
Med
Bioi
1998;438:309-313.
demiologyof, and associated risk
factors
for,
infrared spectroscopy and
the
diagnosiS
and
5 Tsubota
K,
Monden
y,
Yagi
y,
Goto
E,
MGD.
Invest Ophtha/mol
Vis
Sci
2011;52;
treatment of
dry
eye
and cataracts. Spectros-
Shimmura
S:
New treatment of dry
eye:
the
1994-2005.
copy
2014;29:38-52.
effect of calcium ointment through
eyeUd
11
Borchman
D,
Foulks
GN,
Yappert
MC,
Bell
J,
16
WeiXE,MarkoulliM, MillarTJ, WillcoxMD,
skin delivery. Br JOphthaimoI1999:83:767-
Wells
E,
Neravetla
S,
Greenstone
V:
Human
Zhao
Z:
Divalent cations in
tears,
and their
770. meibum lipid conformation and thermody-
influence
on
tear
fIlm
stability in humans and
6 Goto
E,
Dogru
M,
Fukagawa
K,
Uchino
M,
namic
changes
with
meibomian-gland
dys-
rabbits.
Invest Ophthalmol
Vis
Sel
2012;53:
Matsumoto
Y,
Saiki
M,
Tsubota
K:
Successful
function. Invest Ophthalmol
Vis
Sci
2011;52:
3280-3285.
tear
lipid layer treatment
for
refractory dry 3805-3817.
17
Nicolaldes
N,
SantosEC: The
di-
and
triesters
eye
in
office
workers
by
low-dose
lipid
appli-
12
Borchman
D,
Foulks
GN,
Yappert
MC,
Tang of
the
lipids of steer and human meibomian
cation on the full-length
eyelid
margin.
Am
J
D,
Ho
DV:
Temperature-Induced conforma-
glands.
Lipids
1985;20:454-467.
OphthalmoI2006;142:264-270. tional
changes
in
human tear lipids hydrocar-
18
Rantamiiki
AH,
Wiedmer
SK,
Holopainen
7 Green-Church
KB,
Butovich
I,
Willcox
M,
bon chains.
Biopolymers
Blospectrosc
2007;
1M:
Melting points -the key
to
the anti-evap-
Borchman
D,
Paulsen
F,
Barabino
S,
Glasgow
87:124-133.
orative
effect
of the
tear
film
wax
esters.
Invest
BJ:
The international workshop on meibo- Ophthalmol
Vis
Sci
2013:54:5211-5217.
mian gland dysfunction: report of the sub-
19
Ashraf
Z.
Pasha
U,
Greenstone
V,
Akbar
T,
committee on
tear
film
lipids
and
lipid-pro-
Borchman
D:
Quantification of human
sebum
tein interactions in health and
disease.
Invest
on
sldn
and human
meibum
on the
eyelid
mar-
Ophthalmol
Vis
Sd
2011;52:1979-1993.
gin
using
Sebutape®,
spectroscopy and
chemi-
cal
analysis.
Curr
Eye
Res
2011;36:553-562.
Makeup-Meibum Interactions Ophthalmic
Res
2015:53:131-135
DOl:
10.1159/000371852
135
