Archives of Biochemistry and Biophysics 504 (2010) 4049

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics

journal homepage: www.elsevier.com/locate/yabbi

Validation model for Raman based skin carotenoid detection

Igor V. Ermakov, Werner Gellermann *
Department of Physics and Astronomy, University of Utah, Salt Lake City, UT 84112, United States

a r t i c l e

i n f o

Article history:
Available online 1 August 2010
Keywords:
Human skin
Antioxidants
Carotenoids
Raman spectroscopy
HPLC
Validation

a b s t r a c t
Raman spectroscopy holds promise as a rapid objective non-invasive optical method for the detection of
carotenoid compounds in human tissue in vivo. Carotenoids are of interest due to their functions as antioxidants and/or optical absorbers of phototoxic light at deep blue and near UV wavelengths. In the macular region of the human retina, carotenoids may prevent or delay the onset of age-related tissue
degeneration. In human skin, they may help prevent premature skin aging, and are possibly involved
in the prevention of certain skin cancers. Furthermore, since carotenoids exist in high concentrations
in a wide variety of fruits and vegetables, and are routinely taken up by the human body through the diet,
skin carotenoid levels may serve as an objective biomarker for fruit and vegetable intake. Before the
Raman method can be accepted as a widespread optical alternative for carotenoid measurements, direct
validation studies are needed to compare it with the gold standard of high performance liquid chromatography. This is because the tissue Raman response is in general accompanied by a host of other optical
processes which have to be taken into account. In skin, the most prominent is strongly diffusive, nonRaman scattering, leading to relatively shallow light penetration of the blue/green excitation light
required for resonant Raman detection of carotenoids. Also, sizable light attenuation exists due to the
combined absorption from collagen, porphyrin, hemoglobin, and melanin chromophores, and additional
uorescence is generated by collagen and porphyrins. In this study, we investigate for the rst time the
direct correlation of in vivo skin tissue carotenoid Raman measurements with subsequent chromatography derived carotenoid concentrations. As tissue site we use heel skin, in which the stratum corneum
layer thickness exceeds the light penetration depth, which is free of optically confounding chromophores,
which can be easily optically accessed for in vivo RRS measurement, and which can be easily removed for
subsequent biochemical measurements. Excellent correlation (coefcient R = 0.95) is obtained for this tissue site which could serve as a model site for scaled up future validation studies of large populations. The
obtained results provide proof that resonance Raman spectroscopy is a valid non-invasive objective
methodology for the quantitative assessment of carotenoid antioxidants in human skin in vivo.
2010 Elsevier Inc. All rights reserved.

Introduction
Carotenoid molecules play an important protective role in the
skins antioxidant defense system [1]. The eight most concentrated
carotenoid antioxidants in human skin are lycopene, a-carotene,
b-carotene, lutein, zeaxanthin, cryptoxanthin, phytoene and phytouene, with lycopene and the carotenes accounting for about
6070% of total carotenoid content [2]. They are thought to act as
scavengers for free radicals [3], singlet oxygen [4], and other harmful
reactive oxygen species [5,6] formed by metabolic processes or by
excessive exposure of skin to the UV components of sunlight.
If unbalanced due to a lack of antioxidants, the destructive effects
of reactive oxygen species and free radicals can lead to skin malignancies and disease. In animal models, carotenoids have been shown
to inhibit carcinoma formation in the skin [7]. In humans, it has been
Corresponding author. Fax: +1 801 581 4801.
E-mail address: werner@physics.utah.edu (W. Gellermann).
0003-9861/$ - see front matter 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.abb.2010.07.023

shown that skin carotenoid levels are strongly and signicantly correlated with carotenoid levels in plasma [8]. As is found in plasma,
skin carotenoid levels are lower in smokers than in nonsmokers. Carotene levels in skin are known to increase with supplementation [9],
and supplemental b-carotene is used to treat patients with erythropoietic protoporphyria, a photosensitive disorder [10]. Supplemental carotenoids have also been shown to delay erythema in normal
healthy subjects exposed to UV light [11,12]. There is limited evidence that they may be protective against skin and other malignancies [13], but more research is required to conrm these ndings.
Since carotenoids are lipophilic molecules, they are well placed
in the skin to act as chain-breaking antioxidants protecting epidermal polyunsaturated fatty acids from oxygen peroxidation [14].
Other dermal antioxidants such as superoxide dismutase, glutathione peroxidase, alpha-tocopherol, ascorbic acid, and melanins
work in collaboration with carotenoids to provide skin with a
defensive mechanism against free radical attack and oxidative
stress [15]. Because these molecules work as a network, denitive

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

measurement of a subset of these antioxidants provides an indication of the relative strength of the whole system.
The effectiveness of this protective network can be diminished
either by excessive generation of free radicals or by insufcient
antioxidant molecules being supplied to the skin. The result is a
state of oxidative stress where important skin constituents are exposed to free radical damage and associated structural and chemical degenerative effects. If an individual is measured and found to
have a lower than normal skin carotenoid levels, that persons antioxidant defense system would likely be relatively ill-equipped to
balance oxidative processes compared to an individual having
higher levels of antioxidants. Skin antioxidant measurements provide an opportunity for intervention strategies such as increasing
the dietary intake of fruits and vegetables, smoking cessation,
and/or prescribing dietary antioxidant supplements.
The gold standard technique for measuring carotenoids is the
biochemical method of high performance liquid chromatography
(HPLC).1 Requiring chemical decomposition of the sample of interest, HPLC works well for the measurement of carotenoids in serum,
where it has been used to assess carotenoid antioxidant status following the collection of blood samples. Serum measurements, however, are more indicative of short-term dietary intakes of carotenoids
rather than steady state accumulations in skin tissue sites or drops in
concentration due to the inuence of external oxidative stress factors such as smoking and UV light exposure. Skin carotenoid HPLC
measurement have been carried out, too, but it requires highly invasive tissue biopsies. Optical detection approaches for skin carotenoids could potentially overcome these limitations by rapidly and
objectively measuring carotenoid content directly in the skin tissue
sites of interest in a completely non-invasive fashion. The methods
could be used to assess microscopically small tissue volumes if
needed, could be used to track carotenoid status over time, and could
allow for inter-subject comparisons of skin carotenoid levels in vivo.
It is clear, however, that any optical detection method would have to
be correlated with HPLC results of excised skin tissue, a rather nontrivial task, in order to gain acceptance as a viable detection
alternative.
A rst optical approach for skin carotenoid detection used
reectance spectroscopy [11,16]. Dermal carotenoid absorption
and related carotenoid concentrations could be derived from the
measured reectance spectra, correlations with plasma carotenoid
levels could be demonstrated, and skin uptake of b-carotene could
be tracked in supplementation experiments. A difculty of the
reectance approach lies in the analytical derivation of concentration data since the latter cannot be simply derived with Beers law
due to the unknown path length of the reected light in the tissue.
To overcome this problem, a non-linear mapping model was introduced that provides a one-to-one mapping relation between reectance and absorption spectra and that takes into account tissue
inhomogeneity [16]. In a later publication, the spectral reectance
method was compared with objective skin color measurements via
determination of tri-stimulus chromaticity values, with a reasonably good correlation, and it could be demonstrated that carotenoids reduce photosensitivity in Caucasian populations [17].
Another optical approach for the detection of tissue carotenoids
is based on resonance Raman scattering (RRS) spectroscopy. Initially,
we used this method for the detection of carotenoids in the human
retina (see Ref. [18] and references therein). In healthy subjects,
carotenoids are typically very highly concentrated in the macular region of the retinal area, and are thought to protect this critical tissue
region via optical ltering and antioxidant action. The macular
carotenoids are located just below the transparent outer nerve ber

1
Abbreviations used: HPLC, high performance liquid chromatography; RRS, resonance Raman scattering.

41

layer of the retinal layer structure. Their high concentration and the
absence of potentially confounding absorbers in the excitation light
path prior to the carotenoid containing layer provide a highly favorable RRS excitation/detection scenario, as illustrated in Fig. 1 (upper

Fig. 1. Comparison of light excitation/detection paths in RRS based detection of

carotenoids in the human retina (top panel) and skin (bottom panel). In the ocular
case, the excitation light has to traverse only a transparent nerve ber layer, NFL,
prior to excitation of the carotenoid containing layer, MP. RRS responses from the
MP layer, indicated as black arrows, can be obtained in simple 180, single-path,
backscattering geometry free of confounding chromophores. Additional chromophores (circles) exist, but they are concentrated in a posterior layer, the retinal
pigment epithelium, RPE, and their uorescence contributions (white arrows) to the
total optical response can be simply subtracted. In the dermal case, several
potentially confounding chromophore species (open squares and circles) exist
simultaneously besides the carotenoids in stratum corneum (SC), epidermis (EP)
and dermis (D). Varying in general between tissue sites and subjects in strength and
composition, their inuence on the RRS responses has to be minimized. In this
study this is achieved by limiting the light excitation to the stratum corneum layer.

42

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

panel). Since the blue excitation light can proceed without attenuation to the carotenoid containing tissue layer of interest, the RRS response, obtained in 180 backscattering geometry, can be taken as a
direct measure of the macular carotenoid level. The portion of the
excitation light that is not absorbed by the carotenoids, traverses
into the deeper retinal layers, where it generates uorescence from
other chromophores (like lipofuscin in the retinal pigment epithelium), but this uorescence contribution to the overall detected light
response can be treated as a simple superposition and therefore be
subtracted. In fact, our validation studies with excised ocular tissue
structures and eye cups demonstrated a very high correlation between Raman and HPLC methods [19].
Subsequently we suggested RRS as a feasible method also for
the non-invasive quantitative detection of carotenoids in human
skin [20,21]. However, this tissue poses a much more challenging
optical excitation/detection scenario, as sketched in Fig. 1 (lower
panel). First difculties arise since on average skin carotenoid concentrations are about two orders of magnitude lower relative to
the human macula, and since skin tissue is highly heterogeneous.
Furthermore, a variety of other chromophores (like collagen, porphyrin, hemoglobin, and melanin) are simultaneously excited with
the carotenoids of interest, thus producing competing absorption
and associated uorescence (autouorescence) events. Additionally, the outer skin layer, the stratum corneum, produces highly
diffusive scattering events for all light components, including excitation light as well as all uorescence and Raman light components
traveling back towards the detector from within the excited tissue
volume. The overall spectral response in skin is therefore a complex superposition of a weak carotenoid RRS response with strong
absorptions from other chromophores, strong uorescence (autouorescence) contributions, and strong scattering. Since all nonRaman contributions to the total optical response are generated
in the same tissue layers/volume as the carotenoid RRS response,
they cannot be simply subtracted. Furthermore, their respective
concentrations and compositions can be expected to vary between
subjects and tissue sites. In principle, one would therefore have to
measure their combined absorption strength separately with other
spectroscopy methods in order to derive a suitable correction factor for the RRS carotenoid measurement of the tissue site of
interest.
One strategy to avoid the confounding inuences is to limit the
dermal RRS measurements to the palm of the hand or the heel of
the foot, tissue sites that have a relatively thick stratum corneum
layer and therefore prevent the excitation light from penetrating
into deeper layers. Furthermore, these tissue sites are relatively
free of melanin, independent of ethnicity. Using this approach,
we could demonstrate that RRS is able to track carotenoid concentrations over time and to monitor concentration changes occurring
as a result of dietary modications and/or carotenoid supplementation [22]. Also, using sequential excitation with two lasers at
green and blue excitation wavelengths, which differ signicantly
in respective lycopene excitation efciencies, we could demonstrate the possibility to selectively deduce skin lycopene levels
[23]. Furthermore, we developed portable dermal carotenoid
instruments for rst use in a clinical setting [2], and as platforms
for mass produced devices suitable to track carotenoid uptake in
the nutritional supplement industry [22,24]. The main correlation
results between RRS derived skin carotenoid levels and carotenoid
supplements and vegetable consumption have been conrmed in
independent studies [25,26].
Recently it was demonstrated that dermal carotenoid RRS measurements can be carried out with LEDs, instead of lasers, as excitation sources, resulting in the development of a more robust
instrument conguration with high thermal tolerance [27]. On
the clinical side, we have begun to extend dermal carotenoid measurements to the eld of neonatology, where the method avoids

the drawing of blood samples in infants. This will make it possible

to investigate correlations between infant tissue carotenoid levels
and oxidative stress related degenerative disorders, as well as to
develop effective antioxidant containing infant formula [28].
However, before RRS can be accepted as a reliable biomarker for
human research, the method needs to be scrutinized more thoroughly in view of the caveats discussed above. In particular, for
any chosen set of instrumentation parameters and tissue type involved in RRS based carotenoid detection, data is critically needed
on validity as compared to chemical analysis of excised tissue.
Resonance Raman method, linearity of response, inuence of
excitation wavelength on skin carotenoid spectra, and heel skin
measurements
RRS based carotenoid detection takes advantage of the strong
electronic absorption bands of carotenoids in the blue/green spectral region (peak at 450 nm, 80 nm width). This absorption is
caused by strong electric dipole-allowed vibronic transitions of
the carotenoid molecules conjugated p-electron from the 1Ag
ground state to the 1Bu excited state. Optical excitation into this
absorption leads to resonantly enhanced Raman scattering with
more than 1000-fold increase of the Raman scattering cross section
relative to non-resonant excitation. The Raman spectrum of carotenoid molecules is characterized by two prominent Stokes lines at
1159 and 1524 cm1, shown in Fig. 2(a) for a solution of beta carotene in acetone. The lines originate, respectively, from the CC
single-bond and the C@C double-bond stretch vibrations of the
molecules conjugated carbon backbone [29]. A weaker peak at
1008 cm1 is attributed to rocking motion of the molecules methyl
side groups [29].
Due to the unique ordering of the excited electronic states,
carotenoids exhibit an unusually weak intrinsic uorescence. This
allows one to detect the RRS response relatively free of interfering
intrinsic carotenoid uorescence signals. In RRS spectra of pure

Fig. 2. (a) RRS spectrum of b-carotene in a solution (acetone); (b) optical response
obtained from living human skin; and (c) RRS carotenoid spectrum obtained after
subtraction of background from spectrum (b). All spectra were obtained with
excitation at 488 nm. Note that RRS spectra are virtually undistinguishable at room
temperature at the used spectral resolution (1.3 nm).

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

carotenoid solutions, the intensity ratio between Raman signal and

background uorescence is as high as 1:1 at the spectral position
of the C@C Raman peak, and the intensity of each Stokes line varies
proportionally with the concentration of the molecules. Therefore,
in principle, the Raman peak height can be used as a measure of
carotenoid content. Quantitatively, the relation between RRS light
intensity, IR, and excitation light intensity, Iexc, is given by

IR k Iexc rkNEi :

Here, N(Ei) is the population density (concentration) of the molecules, and rk is the Raman scattering cross section, a constant
whose magnitude depends on the excitation and collection geometry, and excitation wavelength. In optically thick media, as in the
skin, a deviation from the linear Raman response of IR versus concentration N can be expected if strong absorbers are present. In this
case one would have an excitation/detection scenario with impeded
light propagation and self-absorption of the Stokes Raman lines.
A cross section of human skin is shown schematically in Fig. 3.
The outermost layer of the skin, the stratum corneum, is relatively
uniform and bloodless. It is entirely composed of cells with missing
nuclei, and features low melanin content irrespective of ethnicity.
The stratum corneum is followed by deeper epidermal and dermal
layers. In contrast to the stratum corneum, these two deeper layers
are rather heterogeneous due to the presence of blood capillaries,
hair follicles, sebaceous glands, and sweat glands. Furthermore,
dermis and epidermis are rich in confounding blood chromophores
and melanin, with blood components varying rather rapidly. As a
consequence, time varying attenuation effects can be expected
for both excitation light and Raman light paths in these layers.
The simplest light propagation scenario for RRS based skin
carotenoid detection can be realized by limiting the RRS skin carotenoid detection method to the stratum corneum layer, as sketched
in Fig. 3, with its smaller subset of potentially confounding chromophores due to the absence of melanin and blood capillaries.
Since the light penetration depth in the visible wavelength region
is typically only 400 lm, this can be realized by choosing the
palm of the hand or the heel of the foot as tissue probing sites; sites

Fig. 3. Illustration of skin morphology and excited tissue volume within the
stratum corneum layer.

43

which have the thickest stratum corneum layer (between 500 and
1000 lm) among skin tissue. A further important advantage of
palm and heel tissue sites lies in the high carotenoid concentrations of the respective stratum corneum layers relative to other
skin sites. This is due to the high lipid-to-protein ratio found in
those tissue sites and to the fact that carotenoids are lipophilic
by nature.
Skin carotenoid measurements based on simple reection spectroscopy appears to support these assumptions. Following dietary
supplementation, statistically signicant changes could be observed in stratum corneum carotenoid content. Also, it could be
demonstrated that dermal carotenoid levels measured at various
tissue sites are highly correlated with serum carotenoid levels [11].
To validate the RRS carotenoid detection method, we compared
in this study in vivo RRS results for heel skin tissue with HPLC results obtained after removal of a thin sliver of tissue. Heel tissue
sites are ideally suited for this type of experiment since the stratum corneum skin layer is extremely thick, typically ranging between 1 and 2 mm. Also, the stratum corneum is essentially
bloodless, so it is easy for participants to self-excise a tissue sample
of sufcient weight (1050 mg) for HPLC analysis without any bodily damage.
The Raman instrumentation and associated light delivery/collection probe used in this study is shown schematically in Fig. 4.
It consists of a portable, ber-based, computer-interfaced instrument with high light throughput. The excitation light, which originates from a 488 nm argon laser, is routed via ber, band-pass
laser line lter, and dichroic beam splitter to the tissue site of interest, as illustrated in Fig. 4 for a heel skin site. A typical measurement uses a 2 mm spot size and 10 mW power for 10 s exposure
time. The Raman-scattered light is collected in a 180 backscattering geometry, then routed through the same beam splitter into a
separate detection path which contains a holographic Rayleigh
light rejection lter, a ber bundle, and a small high-throughput
spectrograph coupled to a cooled two dimensional silicon array
camera.
Typical heel skin spectra are obtained in near real time using
specially developed software and displayed on a computer monitor, as shown in plots (b) and (c) in Fig. 2. The RRS carotenoid responses ride as relatively sharp spectral features on an intense,
spectrally broad autouorescence background (Fig. 2, spectrum
b), the latter clearly demonstrating the presence of chromophores
other than carotenoids in the light excitation path. Since the intensity of this background is nearly 100 times higher than the carotenoid RRS signal, it is necessary to use a detector with high dynamic
range to retrieve the carotenoid RRS response with sufcient accuracy. For data retrieval, the autouorescence background is modeled with a fourth order polynomial and subtracted from the raw
spectrum [20]. As a result one obtains the isolated RRS spectrum
of the tissue sites carotenoids featuring the three RRS lines characteristic for carotenoids (Fig. 2, spectrum c). Their spectral positions
are undistinguishable from pure b-carotene in liquid organic solvents (Fig. 2, spectrum a).
The RRS peaks result from contributions of all skin carotenoid
species absorbing at the excitation wavelength. Since their absolute positions and bandwidth are indistinguishable at room (human body) temperature, it allows us to use the absolute peak
height of the C@C signal at 1524 cm1 as a measure of the overall
carotenoid content at the tissue site. Results can be obtained with
safe light excitation intensities [30], even though the RRS effect is
relatively weak.
In order to investigate the optical measurement conditions for
highest repeatability of in vivo results we rst developed a skin
phantom. Consisting of a mixture of glycerol, ne aluminum oxide
powder, Coumarin 540A dye, and b-carotene, it simulates the
spectral response of skin in terms of light scattering, carotenoid

44

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

Fig. 4. Left: Schematic diagram of RRS instrument and ber-based light delivery/detection probe module used in this study (left panel, not to scale). Right: Illustration of RRS
measurement of heel skin tissue site.

RRS response, and spectral shape as well as strength of the overlapping autouorescence. Measurements of the phantom carotenoids
revealed an extremely high repeatability with a standard deviation
below 1% for 10 consecutive measurements. Next we compared the
result with in vivo repeatabilities. Standard deviations between
human subjects obtained in vivo for 10 consecutive RRS scans per
subject were seen to fall into a wide range between 0.5% and
15%, with an average around 4%. Since instrumentation aspects
can be excluded, the cause for the strong inuence of the sampling
site on the reproducibility has to lie in a spatially non-uniform skin
carotenoid distribution in those subjects. However, most subjects
appear to have a relatively uniform carotenoid distribution, with
a resulting standard deviation of RRS intensities better than 4%.
In each case, the repeatability gures can be held to a minimum
by tightly controlling the positioning of the probe on the tissue site
of interest.
We checked the linearity of the RRS response by measuring
adjacent skin tissue sites under variation of exposure times and laser power levels. The results of these experiments are summarized
in Fig. 5. In one experiment, we measured a selected heel skin site
ve times with xed exposure time, then repeated the measurements ve times while increasing the exposure time by 5 s in each
step. In a second experiment, we measured a selected heel skin site
ve times again with xed exposure time, then repeated the measurements six times while increasing the excitation laser power by
2 mW in each step. The plots show the average of each measurement set along with its error bar (standard deviation), and a
straight line through the points as guide to the eye. As can be seen
from the plots, both relationships demonstrate nearly perfect linear behavior, and are therefore in agreement with Eq. (1). This
proves that under the used instrument and measurement conditions and limitation of the optical light paths to the outer stratum
corneum, the absorption and uorescence effects of the residual
chromophores are sufciently small.
A further important parameter for validity considerations of the
Raman method is the spectral position of the laser excitation wavelength within the carotenoid absorption band. It is desirable to use
instead of the argon laser a solid state laser for excitation. Current
solid state laser lines useful for carotenoid excitation exist at 488
and 473.2 nm. Because solid state lasers are far superior in terms
of wall plug efciency, are much more compact and generate less

Fig. 5. RRS intensity of C@C carotenoid peak, measured under excitation at 488 nm
versus exposure time (a) and excitation power (b), respectively. Both plots demonstrate linearity of RRS response under the chosen experimental conditions.

heat as well as noise, we explored their RRS performance relative

to the argon laser. The bandwidths of the solid state laser lines
are larger than that of the argon laser line. As a consequence, the
resulting RRS carotenoid line widths are broader as well, and

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

may not be as clearly separable from the autouorescence background. Furthermore, the autouorescence intensity increases
strongly when changing excitation from 488 to 473.2 nm. The
question arises therefore, as to whether it is necessary to validate
the RRS method separately for each excitation condition.
To nd out, we compared in a rst step a compact solid state laser emitting at 488 nm with an argon laser emitting at the same
wavelength. We found that the solid state laser RRS results are
fully interchangeable with its gas counterpart, i.e. the carotenoid
excitation efciency and spectral quality were essentially identical
[I.V. Ermakov, W. Gellermann, unpublished results, 2004]. This
shows that laser line width differences can be ignored between
the two sources. Next, we investigated changes in the carotenoid
RRS response regarding spectral quality and repeatability when
using a solid state laser at 473 nm as excitation source.
In Fig. 6 we compare the raw RRS carotenoid spectra obtained for
the palm of the same human subject with 473 nm and 488 nm excitation wavelength. All other excitation parameters, i.e. power density at the skin, exposure time, etc., were kept unchanged. The
spectra were corrected for the spectral throughput of probe and
spectrograph. As seen from Fig. 6, the C@C Raman peak components
of the skin spectra have almost the same amplitude, as can be expected from Raman theory, i.e. the resonance Raman excitation efciency follows the absorption prole. The absorption coefcients of

45

all skin carotenoids change only by several percent when moving

from 488 to 473 nm since all individual carotenoid absorption bands
are very wide and the change in excitation wavelengths occurs near
the spectrally smooth absorption maxima. However, contrary to the
Raman peak intensity, the intensity of the skin uorescence background increases by a factor of three for 473-nm excitation, and
therefore reduces the already small ratio between the useful C@C Raman signal to the autouorescence background even further, to
about 0.3%. In comparison, at 488 nm excitation, the average ratio
reaches 1%. The lowered Raman to uorescence ratio deteriorates
the signal-to-noise ratio forp
the
C@C RRS signal due to increased shot
noise, which now is about 3 times higher compared to excitation
with 488 nm. To compensate for this reduction, we increased the
exposure (integration) time accordingly to determine whether it is
possible, in this way to retain the same signal-to-noise ratio achievable with 488 nm excitation. Indeed, the results showed that comparable performance gures can be obtained in either case, and that the
increased exposure time at 473 nm is a small sacrice, outweighing
all drawbacks associated with air-cooled argon laser or the currently
more expensive solid state alternative at 488 nm.
For RRS/HPLC comparisons we recruited eight volunteer subjects (3668 year age range) including both genders (4 males, 4 females). All subjects were Caucasians with light to medium skin
pigmentation, thus excluding any attenuation effects on RRS derived skin carotenoid levels which would occur if signicant melanin concentration levels were to exist in the probed tissue volume.
Also, all subjects were free of any skin disorders. The back of the left
heel of each subject was brought in contact with the optical window of the Raman probe, marked, and measured three times. The
RRS results are listed in Table 1. Within 15 min of the RRS measurements, a sliver of tissue was self-excised at the center of the marked
area, using a sterile single-use regular blade. The skin samples were
immediately weighed, deposited in an individual air tight vial, and
stored at dry ice temperature for later extraction and HPLC analysis.
For RRS/HPLC correlation we used Microsoft Excels LINEST function
and Sigmaplot (Systat Software, Inc.) graphics.
Biochemical carotenoid detection method

Fig. 6. Fluorescence spectrum (top) and Raman spectrum (bottom) of skin tissue
site measured with excitation at 473 nm (curves a) and 488 nm (curves b). The
exposure times and excitation intensities in both cases were identical and equal to
30 s and 10 mW, respectively. Strong uorescence components occurring under
473 nm and 488 nm excitation mask carotenoid RRS lines. Using a detector with
high dynamic range and subtraction of background uorescence, RRS signals can be
retrieved. Raman spectrum excited with 473 nm appears to be much noisier
compared to that excited with 488 nm due to higher quantum noise associated with
higher uorescence background.

The gold standard HPLC technique for skin carotenoid analysis

includes effective transfer of the analyte from the tissue to a suitable solvent (extraction), spatial separation of different analytes
from the extract mixture, and an objective, usually optics-based
detection of the spatially separated chemical components. When
dealing with a complex mixture of several analytes, one sometimes
has to perform the second step several times using different HPLC
techniques to isolate a specic carotenoid, such as different column types, direct/reverse phase HPLC, etc. Each step of HPLC is
subject to a rigorous calibration where the known concentrations
of all chemicals in question are measured to nd the effective system response for each. Thus, HPLC is a complex multi-step,
destructive, and time consuming technique that needs relatively
large tissue volume (10 mg) to satisfy detection accuracy.
The collected skin samples were weighed after excision, immediately frozen and shipped at dry ice temperature for later HPLC
analysis [Craft Technologies, Inc., Wilson, NC]. The analysis protocol involved the following procedures [31]. The skin samples were
weighed again and transferred into tubes with 430 ll PBS and
70 lL collagenase (70 mg/mL PBS). They were incubated for 1 h
at 37 C. Glass beads, 2 mL of 50% methanol/50% THF and 100 lL
of tocol (internal standard) were added. The samples were mixed
to macerate the tissue and extract the carotenoids. Samples were
saponied overnight at 4 C by addition of 100 lL 10% pyrogallol
and 1 mL of 40% potassium hydroxide. Samples were extracted
two times with 5 mL of hexane and dried over sodium sulfate.
The combined hexane extract was dried in a SpeedVac. The

46

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

Table 1
RRS carotenoid measurements of human heel skin sites: statistics and comparison with HPLC derived results.
Subject ID

Raman measurements, counts

HPLC, ng/mg

Individual measurements
S1
S2
S3
S4
S5
S6
S7
S8

38,600
31,000
13,500
12,598
17,377
25,907
13,064
24,264

34,668
31,152
14,500
12,288
18,788
26,228
15,311
25,241

33,600
31,331
14,500
12,068
18,070
23,290
11,464
24,838

sample for astaxanthin analysis was extracted with acetone rather

than CH3OH/THF and was hydrolyzed using cholesterol esterase. It
was separated by HPLC using a Diol column.
The HPLC system consisted of a computer data system, an autosampler maintaining samples at 20 C, a column heater at 31 C,
and a diode array detector (ThermoSeparation Products, Fremont,
CA). The separation was performed isocratically on a Spherisorb
ODS2 column (3 lm, 4.0  250 mm with titanium frits, ES Industries, West Berlin, NJ) protected by a Javelin guard column containing a similar stationary phase (Thermo Electron Corp,
Bellefonte, PA). The mobile phase consisted of acetonitrile/dioxane/isopropanol/triethylamine (80/15/5/0.1) at a ow rate of
1.0 mL/min. The alcohol component contained 100 mM ammonium acetate. The diode array detector with light pipe ow cell
was programmed to monitor tocol and phytoene at 290 nm, phytouene at 325 nm, and carotenoids at 450 nm.
Linear calibration curves were prepared for three concentrations of analytes which spanned the physiological levels of micronutrients in serum. The calibrants included astaxanthin, lutein,
zeaxanthin, a-cryptoxanthin, b-cryptoxanthin, lycopene, a-carotene, b-carotene, phytoene, and phytouene.
A typical HPLC chromatogram is presented in Fig. 7, with the
absorption of the carotenoid extract at the carotenoid maximum
absorption wavelength, shown as a function of time. The area under
each peak of the chromatogram is associated with the concentration
of a specic carotenoid or group of carotenoids. Carotenoid species
are separated spatially in HPLC instrumentation due to the specic
and different retardation times as they traverse the length of the column. The degree of retardation depends on the chemical structure
and mass of the carotenoid species and leads to the specic retention
time (time when a specic analyte elutes) at the chromatogram.
Good signal-to-noise ratios for the carotenoid peaks allow us to
determine their concentration in skin with high accuracy.

Average

STD

STD, %

35,623
31,161
14,167
12,318
18,079
25,142
13,280
24,781

2633
166
577
266
706
1612
1933
491

7.4
0.5
4.1
2.2
3.9
6.4
14.6
2.0

1.427
1.137
0.748
0.748
0.676
1.263
0.561
1.064

Results and discussion

The HPLC results for all eight subjects are reported in Table 2 as
amounts in ng/1 mg tissue for each carotenoid species found in the
sample. The particular HPLC analyses concentrated on 13 major
carotenoid species, namely lutein, zeaxanthin, cis-lutein/zeaxanthin, a-cryptoxanthin, b-cryptoxanthin, trans-lycopene, cis-lycopene, a-carotene, trans-b-carotene, cis-b-carotene, canthaxanthin,
phytoene, and phytouene.
Phytoene and phytouene were excluded from further consideration since they do not contribute to the RRS signal under blue light
excitation. These shorter-chain carotenoids absorb in the UV range
such that for effective Raman excitation one has to use a narrow linewidth UV source [30]. It is worthwhile to note that the UV active
carotenoids are found in signicant amounts in the skin [2]. This
study conrms the nding, with the tissue sample of one subject
(S4) containing as much as 48% phytoene relative to the total carotenoid content. The column farthest to the right in Table 2 lists the total concentration of skin carotenoids absorbing in the visible. Since
these contribute to the Raman signal, they are compared with the
RRS data of Table 1.
Comparing the two sets of data, one needs to be aware of the
fact that either set is prone to a number of systematic errors. For
instance, any HPLC analysis is set up to see a denite and limited subset of the species, thus not including minor carotenoids
and a potentially large class of oxidized carotenoids. This is in
contrast to the RRS effect which would pick up any C@C signal
from those unaccountable molecules as well. On the other hand,
under single wavelength excitation, RRS measurements give a value, R, comprising the contribution of a number of carotenoid
species,

R Iexc

ri N i ;

Fig. 7. Typical HPLC chromatogram of carotenoids extracted from heel skin tissue site.

47

1.427
1.137
0.748
0.748
0.676
1.263
0.561
1.064

where the contribution of each species is not strictly proportional to

its concentration Ni but, in general would have to be corrected for
slightly differing Raman excitation cross sections ri. In the current
model, we assume that all carotenoid species in the skin contribute
to the Raman signal with the same strength,

0.040
0.055
0.035
nd
nd
nd
nd
nd

R Iexc r

Ni ;

0.314
0.351
0.116
0.088
0.090
0.157
0.084
0.132

0.124
0.147
0.063
0.051
0.021
0.052
0.027
0.049

0.262
0.130
0.141
0.683
0.013
0.059
0.015
0.042

nd not detectable.
Canthaxanthin not detectable in all samples.
a
Total concentration of carotenoid molecules in skin samples excluding UV-absorbing carotenoids phytoene and phytouene.

0.085
0.082
0.053
0.050
0.025
0.042
0.033
0.024
0.110
0.129
0.119
0.140
0.085
0.262
0.084
0.232
0.200
0.174
0.190
0.243
0.137
0.350
0.115
0.276
0.233
0.111
0.052
0.041
0.101
0.169
0.052
0.153
0.045
0.025
0.021
0.018
0.032
0.033
0.02
0.037
0.008
0.007
0.007
0.003
0.006
0.017
0.006
0.009
0.082
0.039
0.051
0.050
0.052
0.039
0.031
0.046
0.198
0.072
0.076
0.064
0.127
0.142
0.109
0.106
38.2
61.3
45.9
23.2
45.6
14.2
47.6
8
S1
S2
S3
S4
S5
S6
S7
S8

Subject ID Tissue weight, mg Lutein Zeaxanthin cis-Lutein/ a-Cryptoxanthin b-Cryptoxanthin trans-Lycopene cis-Lycopene a-Carotene trans-b-Carotene cis-b-Carotene Phytoene Phytouene Totala, ng/mg
zeaxanthin

Table 2
Carotenoid composition of human skin as measured with HPLC.

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

where r is effective or average Raman excitation cross section/efciency. To justify this approach, we note that the RRS section ri (k)
usually follows the absorption prole ei (k), and that the specic
absorption coefcients emax
and emax
differ from one another by
i
i
max
and ri  rj = r.
no more than 10% [32], so ei  emax
j
On the optical side, we observed that human skin is not always
uniform in carotenoid content, even though this in not detectable
by the naked eye. This is demonstrated by Fig. 8 and Table 3, which
contain the results of 10 Raman measurements for three of the volunteer subjects described above. Five RRS measurements were taken at the same skin site, with the optical probe not moving
between measurements; subsequently, ve measurements were
performed on adjacent skin sites. The optical probe was intentionally removed after each measurement, and replaced prior to completing the following measurement. For each series of
measurements, the standard deviation was calculated. Similar data
on the skin phantom is also reported in Table 3 for reference purposes. The skin phantom can be considered as ideal sample in
terms of homogeneity, as it delivers the best repeatability with a
relative standard deviation better than 0.5%. For living skin, the
repeatability is still excellent (about 1%) as long as the optical
probe remains on the selected site of the skin. However, when
removing the probe and placing it back onto the site, we effectively
change the sampling site due to the positioning error. This error, in
our estimate did not exceed the diameter of the sampling spot
(2 mm), but causes a signicant and reproducible span in standard deviation of the RRS measurements, varying between 1%
and 4%. The consistent and relatively high standard deviation observed in subject B serves as indirect evidence for a non-uniform
carotenoid distribution in human skin in some subjects. If this
inhomogeneity is on the same scale as the sampling mismatch between RRS measurements, lower repeatability gures result in
these cases.
In spite of the presence of these unavoidable errors and
assumptions, we obtained excellent correlations between Raman
and HPLC data, as seen from Fig. 9, with R as high as 0.95 or
R2 = 0.91. Note that the regression line even passes through the origin, proof that the inuence of the additional chromophores to the
RRS method can be ignored under the used instrument conguration and measurement parameters. Over a wide range of physiological concentrations of skin carotenoids, an excellent linearity
therefore exists between skin carotenoid levels measured with
the RRS method versus HPLC. This result validates RRS spectroscopy as an objective and accurate method for skin carotenoid
measurements.
We can use the HPLCRaman correlation results to directly calibrate our RRS instrument in carotenoid concentration units
According to the regression analysis, the cumulative skin carotenoid content, C, measured in lg/g of skin tissue is linked to the
height of the C@C carotenoid Raman peak, I, in the instrument via

C lg=g 4:3  105  I counts:

Integrating Eq. (4) with our data acquisition software, it is possible

to obtain skin carotenoid content in convenient lg/g units in real
time.

48

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049

Fig. 8. Examples of standard deviations for ve consecutive carotenoid RRS

measurements obtained for a tissue phantom and for three subjects, obtained with
optical probe xed on the tissue site (closed circles) and optical probe removed
between measurements (open circles). Testretest RRS results suggest different
degrees of carotenoid homogeneity in living human skin. With xed probe,
standard deviations for all subjects are small and very comparable. With reapplied
probe, values of standard deviations are higher and differ signicantly.

Table 3
Testretest skin carotenoid RRS measurements obtained for tissue phantom and three
subjects in ve consecutive measurements per sample site. The data suggest different
degrees of carotenoid homogeneity in human skin.
Meas.
conditions

Probe xed on tissue site

Probe removed between

meas.

Measurements,
counts

STD,
%

Measurements,
counts

STD,
%

Phantom

25,701
26,005
25,946
26,373
25,296

0.31

25,117
25,934
25,355
26,398
26,094

0.41

Subject A

52,619
48,075
52,397
49,218
47,326

0.98

47,633
52,749
53,215
48,325
50,219

1.0

Subject B

27,938
28,777
31,304
29,840
30,614

0.92

25,441
34,715
29,262
37,426
24,014

3.84

Subject C

29,474
27,461
31,105
28,823
27,539

1.05

31,184
31,828
25,003
26,050
27,755

2.14

Fig. 9. Heel skin carotenoid levels of eight subjects measured with RRS method
in vivo, versus HPLC results of subsequently excised tissue samples. Dotted line
indicates corresponding linear regression crossing the origin, with resulting high
correlation coefcient R = 0.95.

tion algorithm for the RRS carotenoid levels under these measurement conditions. This holds for two tested excitation wavelengths
of 488 nm and 473.2 nm, and by inference, for any wavelength
lying on the long-wavelength shoulder of the carotenoid absorption range.
Along with previously reported, more indirect evidence for a
reasonably good correlation between RRS skin carotenoid levels
and levels in serum, these correlation experiments validate the
RRS method and establish it as a rapid, non-invasive optical methodology for measurement of carotenoid antioxidant status in vivo.
It is possible to use the method for rapid quantitative assessment
of dermal carotenoid levels in clinical and eld settings in large
populations under the specied experimental conditions. This
should make it possible to apply the method as biomarker for fruit
and vegetable intake in large-scale dietary intervention studies as
well as in clinical studies investigating correlations between dermal carotenoid levels and exposure to oxidative stress.
The observed very high correlation for tissue sites with thick
stratum corneum indicates that it may be possible to extend dermal RRS carotenoid detection also to skin tissue sites with higher
contributions of additional chromophores, such as abdominal skin.
This would come with a penalty of lower accuracy, and would have
to be validated in a similar RRS/HPLC correlation study for that
more complex tissue. However, this might well be acceptable in
clinical studies looking only for relatively large changes in dermal
carotenoid content between subjects, or in the same subjects over
time in response to oxidative stress exposure. Lastly, the obtained
validation results will make it possible to cross-correlate other
optical carotenoid detection methods, such as reectance spectroscopy, with the Raman method to evaluate their accuracy in the
measurement of skin carotenoids.

Conclusion
References
In conclusion, we directly compared RRS based in vivo skin
carotenoid detection with the biochemical gold standard method
of HPLC. Even though the methodologies are totally different, a
remarkably high correlation (R = 0.95) is obtained for skin tissue
sites in which the light penetration is conned to the stratum corneum layer. With the specied, carefully chosen light excitation
and detection conditions, the inuence of intrinsically uncontrollable, potentially confounding factors on the RRS carotenoid response can be held to a negligible subset of chromophores. The
remaining autouorescence background in the total dermal optical
response therefore can be treated as a simple offset in the reduc-

[1] N. Krinsky, S.T. Mayne, H. Sies (Eds.), Carotenoids in Health and Disease,
Marcel Dekker, New York, 2004.
[2] T.R. Hata, T.A. Scholz, I.V. Ermakov, R.W. McClane, F. Khachik, W. Gellermann,
L.K. Pershing, J. Invest. Dermatol. 115 (2000) 441448.
[3] F. Bhm, J.H. Tinkler, T.G. Truscott, Clin. Cancer Res. 1 (1995) 9899.
[4] C.S. Foote, R.W. Denny, J. Am. Chem. Soc. 90 (1968) 62336235.
[5] A. Farmillo, F. Wilkinson, Photochem. Photobiol. 18 (1973) 447450.
[6] P.F. Conn, W. Schalch, T.G. Truscott, J. Photochem. Photobiol. B 11 (1991) 41
47.
[7] L.C. Chen, L. Sly, C.S. Jones, R. De Tarone, L.M. Luca, Carcinogenesis 14 (1993)
713717.
[8] Y.M. Peng, Y.S. Peng, Y. Lin, T. Moon, D.J. Roe, C. Ritenbaugh, Nutr. Cancer 23
(1995) 233246.

I.V. Ermakov, W. Gellermann / Archives of Biochemistry and Biophysics 504 (2010) 4049
[9] M.R. Prince, J.K. Frisoli, Am. J. Clin. Nutr. 57 (1993) 175181.
[10] M.M. Mathews-Roth, Ann. NY Acad. Sci. 691 (1993) 127138.
[11] W. Stahl, U. Heinrich, H. Jungmann, H. Sies, H. Tronnier, Am. J. Clin. Nutr. 71
(2000) 95798.
[12] J. Lee, S. Jiang, N. Levine, R. Watson, Proc. Soc. Exp. Biol. Med. 223 (2000) 170
174.
[13] E. Giovannucci, J. Natl. Cancer Inst. 91 (1999) 317331.
[14] H.I. Krinsky, S.M. Deneke, J. Natl. Cancer Inst. 69 (1982) 205210.
[15] D.P.T. Steenvorden, G.M.J. Beijerbergen van Henegouwen, J. Photochem.
Photobiol. B 41 (1997) 110.
[16] W. Stahl, U. Heinrich, H. Jungmann, H. Tronnier, H. Sies, Methods Enzymol. 319
(2000) 494502.
[17] S. Alaluf, U. Henrich, W. Stahl, H. Tronnier, S. Wiseman, J. Nutr. 132 (2002)
399403.
[18] I.V. Ermakov, M. Sharifzadeh, P.S. Bernstein, W. Gellermann, in: J.T. Landrum
(Ed.), Carotenoids Physical, Chemical and Biological Functions and
Properties, CRC Press, Atlanta, GA, 2009.
[19] P.S. Bernstein, M.D. Yoshida, N.B. Katz, R.W. McClane, W. Gellermann, Invest.
Ophthal. Vis. Sci. 39 (1998) 20032011.
[20] I.V. Ermakov, M.R. Ermakova, R.W. McClane, W. Gellermann, Opt. Lett. 26
(2001) 11791181.
[21] W. Gellermann, R.W. McClane, N.B. Katz, P.S. Bernstein, Method and Apparatus
for Noninvasive Measurement of Carotenoids and Related Chemical
Substances in Biological Tissue, US Patent #6,205,354, 2001.

49

[22] C.R. Smidt, W. Gellermann, J.A. Zidichouski, Fed. Am. Soc. Exp. Biol. J. 18A
(2004) 480.
[23] I.V. Ermakov, M.R. Ermakova, W. Gellermann, J. Lademann, J. Biomed. Opt. 9 (2)
(2004) 332338.
[24] W. Gellermann, J.A. Zidichouski, C.R. Smidt, P.S. Bernstein, in: L. Packer, U.
Obermueller-Jevic, K. Kraemer, H. Sies (Eds.), Carotenoids and Retinoids, AOCS
Press, Champaign, Illinois, 2005 (Chapter 6).
[25] C.R. Smidt, D.S. Burke, Current Topics in Nutraceutical Research 2 (2004) 79
91.
[26] S. Rerksuppaphol, L. Rerksuppaphol, J. Med. Assoc. Thai 89 (2006) 12061212.
[27] S.D. Bergeson, J. B Peatross, N.Y. Eyring, J.F. Fralick, D.N. Stevenson, S.B.
Ferguson, J. Biomed. Opt. 13 (2008) 044026-1044026-6.
[28] G.M. Chan, C. Rau, W. Gellermann, M. Ermakova, Retinopathy of prematurtiy
and carotenoids in human milk fed infants, Abstract, 2006 Meeting of the
American Academy of Pediatrics, Washington, DC, 2006.
[29] Y. Koyama, in: G. Britton, S. Liaaen-Jensen, H. Pfander (Eds.), Carotenoids,
Spectroscopy, vol. 1B, Birkhuser, Basel, 1995, pp. 135146,.
[30] I.V. Ermakov, M. Sharifzadeh, M.R. Ermakova, W. Gellermann, J. Biomed. Opt.
10 (6) (2005) 064028.
[31] The methods are essentially the same as described in N.E. Craft,
Chromatographic Techniques for Carotenoid Separation, Current Protocols in
Food Analytical Chemistry F2.3.1F2.3.15, John Wiley & Sons, Inc., 2001.
[32] G. Britton, S. Liaaen-Jensen, H. Pfander (Eds.), Carotenoids, vol. 1B, Birkhaeuser
Verlag, Basel, Switzerland, 1995, p. 57.