Chemical Physics Letters 559 (2013) 26–29

Contents lists available at SciVerse ScienceDirect

Chemical Physics Letters

journal homepage: www.elsevier .com/locate /cplet t

Monitoring LED-induced carotenoid increase in grapes by TransmissionResonance Raman spectroscopy

Alicia G. Gonzálvez, Nerea L. Martínez, Helmut H. Telle 1, Ángel González Ureña ⇑Unidad de Láseres y Haces Moleculares, Instituto Pluridisciplinar, Universidad Complutense de Madrid, Paseo Juan XXIII-1�, 28040 Madrid, Spain

a r t i c l e i n f o

Article history:Received 19 November 2012In final form 28 December 2012Available online 9 January 2013

0009-2614/$ - see front matter � 2013 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.cplett.2012.12.054

⇑ Corresponding author.E-mail address: laseres@pluri.ucm.es (Á.G. Ureña).

1 Present address: Department of Physics, SwanseSwansea SA2 8PP, United Kingdom.

a b s t r a c t

Transmission Resonance Raman (TRR) spectroscopy combines increased signal-to-noise ratio withenhanced analytical sensibility. TRR was applied to directly monitor, without any sample preparation,the enhancement of b-carotene content in table grapes when they are irradiated by low power UV-LEDs.It was shown that, with respect to control samples, the carotenoid content in the grapes increased aboutfive-fold, using UV-LED irradiation doses being two orders of magnitude lower than the maximum limitallowed by United States Food and Drug Administration. These promising results may pave the way forthe development of easy, non-invasive techniques to improve food quality.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Raman Scattering is a widely used technique in AnalyticalChemistry due to various reasons, one of the most important beingthat in many instances no special sample preparation is required[1–7]. This is contrasted by a severe limitation, namely that Ramansignals are normally very weak (typically only one in every 106–108 incident photons is Raman scattered). This drawback has dri-ven the development of distinct enhancement techniques as, forexample, Resonance Raman (RR) or Surface Enhanced Raman Spec-troscopy (SERS). The achievements in Resonance Raman spectros-copy and the impact on aspects of its analytical potential havebeen reviewed recently [8]. Although Resonance Raman scatteringis associated with greatly improved sensitivity the technique alsocomes with some fundamental catches, namely that of inherentlyintense fluorescence emission produced by chromophores in thebiological material, and non-negligible diffuse scattering losses inhigh-density samples.

An elegant method to circumvent the aforementioned disturb-ing fluorescence, produced by the resonant excitation of the illumi-nated sample, is to exploit the benefits of a variant technique,Transmission Raman spectroscopy (TRS). While known since themid-1960s his particular technique was re-discovered for practicaluse in 2006. Then researchers demonstrated its capability in ana-lyzing samples in tablet or powder form, up to several millimetersof thickness. Among the many advantages of this approach it isworth mentioning its ability to probe bulk content of powdersand tissues, rejecting Raman and fluorescence components

ll rights reserved.

a University, Singleton Park,

produced at the sample surface as well as the absence of justsub-surface sampling. Because of these advantages TRS hasevolved into a successful technique for the analysis of pharmaceu-tical specimen [9,10].

In recent work from our laboratory a Transmission ResonanceRaman (TRR) spectrometer was described, with the particularapplication to carotenoid detection (specifically b-carotene in car-rots); a detection limit of sub-nanograms for b-carotene was re-ported [11]. That particular study may be placed within theframework of research which increasingly focuses on the nutri-tional relevance of a series of phytochemicals, such as flavonoids,carotenoids and glucosinates [12–14]. This is not surprising sincenowadays plant secondary metabolites are widely used in humanhealth and disease prevention.

The aforementioned carotenoids may be viewed as a represen-tative example of plant secondary metabolites. They are organicpigments mostly found in red-, orange- and yellow-colored fruitsand vegetables. Their biosynthesis is attracting a growing interestfor a series of reasons, the major one being the fact that vertebratesdo not synthesize these pigments. Human beings need the intakeof carotenoids to convert them into retinoids, like retinal (the mainvisual pigment). In addition, all carotenoids containing a b-ring canbe converted to retinol, i.e. the precursor of vitamin A. The lack ofthis vitamin is widespread in developing countries but is equallyencountered in the poorest part of the population of developedcountries.

The metabolic necessities sketched above raised the interest inthe scientific community to search for breakthroughs for the pro-duction of secondary metabolites in general, and for metabolicengineering of plant carotenoids in particular.

It is well-known that plant abiotic stress, like UV-irradiation,triggers a plant defense mechanism enhancing the contents ofsome secondary metabolites, the so-called defense molecules. An

A.G. Gonzálvez et al. / Chemical Physics Letters 559 (2013) 26–29 27

example for such a metabolite is trans-resveratrol (e.g. found ingrapes), a compound widely studied because to its beneficialproperties for human health. However, despite the many studiesof phenolic elicitation in plants induced by UV irradiation, thereare few investigations related with this abiotic method to elucidatecarotenoids in fruits and plants. In this present work we report onan easy approach to enhance the carotenoid content in fruit using anon-invasive optical method. Specifically, it is shown how a lowdose (for its definition see further below) of UV-B irradiation ofgrapes significantly enhances its carotenoid content. The presenceof carotenoids in grapes is well documented [15–18], withb–carotene being the most abundant within this class of pigment[16].

Probably the key aspect of the present investigation is the use ofTRR spectroscopy to monitor the b-carotene content in grapes,without having to resort to any sample preparation or extractionmethod. This is afforded by the enhanced sensibility of the TRRmethod to specific molecular compounds, because of the resonantcharacter of the laser excitation (the wavelength is chosen to giveaccess to an electronic transition band). A further, particularly rel-evant aspect of the present investigation is the use of UV-light froma low power LED. This opens the possibility to implement a simpletechnique and methodology using low-cost constituent compo-nents, which make the method commercially viable.

Figure 1. Schematic layout of the experimental Transmission Resonant Ramansetup; LED irradiation of the grape is at 90� to the b-carotene TRR excitation/observation axis. Laser – cw Ar+ laser, operating at k = 514.5 nm; D – diaphragm; S –(grape) sample mounted in a XY micrometer sample manipulator; L – cylindricallens with f = 25 mm; F – razor-edged filter (Semrock LP03-514RU-25); SP/CCD –spectrometer with CCD-array detector. Insert: typical emission spectra of the twoUV-LEDs used to stimulate secondary metabolites in grapes; (red) solid line – LEDkpeak = 295 nm; (blue) dashed line – LED kpeak = 300 nm. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version ofthis article.)

2. Methods

In the experiments carried out in this study five assortments ofgrapes (muscatel variety) were used. These were divided into threesub-groups. Two groups of grapes were irradiated for 30 min withlight from UV-LEDs, with peak emission at 295 and 300 nm,respectively. The particular devices had radiative power of20 lW and FWHM of 3 nm (the emission spectra of these two LEDsare shown in the insert in Figure 1). The third group of grapes wasused as a control group, i.e. they were not irradiated. After treat-ment the grapes underwent analysis within our Transmission Res-onant Raman spectrometer, whose schematic layout is depicted inFigure 1; its main details can be found elsewhere [11] The TRRexcitation wavelength in the experiments was 514.5 nm, whichsatisfies the resonance conditions for the b-carotene absorptionband rather well at its tail end. Consequently, the TRR signal wassubstantially enhanced, and at the same time contributions fromfluorescence were minimized [1,11].

The other conditions for the experiments reported here were asfollowing. The laser beam diameter at the (grape) sample locationwas of the order 1.0–1.5 mm, with the Ar ion laser power for the514.5 nm line set to typically 70 mW. The temperature of theCCD detector was kept at below �60 �C to minimize dark currentnoise. The Raman spectra (as the examples shown in Figure 2)were recorded over a period of 660 s, averaging 66 scans of 10 seach. This was required in order not to saturate the detector andto make it easier to eliminate cosmic ray events from the spectra;subsequent to the acquisition and prior to data interpretation spec-tral background subtraction was performed by advanced rollingcircle filtering [19].

Individual measurement runs constituted a sequence of firstirradiating the grapes for 30 min with the UV-LED light and, subse-quently, Raman spectra were recorded as a function of the timeelapsed after having irradiated the grape, in regular intervals from0 h up to 53 h. Note that the Raman measurements were carriedout without any sample treatment or preparation, i.e. a directand online sample spectral monitoring was implemented. Thismeasurement protocol was repeated for all grapes in a set, i.e. fiveeach irradiated by 295 and 300 nm UV-LED light, respectively, andfive control grapes.

3. Results and discussion

3.1. Spectroscopy and temporal evolution of b-carotene after UVirradiation

Typical Raman spectra of the irradiated grapes as well as a non-irradiated control sample are shown in Figure 2. The particularspectra shown here were taken 24 h after the irradiation. Allspectra exhibit three main peak features, corresponding to thewell-known vibrational bands of beta-carotene. The two mostprominent features at about 1160 and 1525 cm�1 can be associatedwith its –C–C– (m2) and –C@C– (m1) stretch vibrations [20,21]. Here,the latter peak was used to follow the time evolution of theb-carotene content in the grapes. Note that the ‘free’ b-carotenem1-mode, in the literature tabulated at 1516 cm�1, appears slightlyshifted to 1526 cm�1 in the grapes, due to matrix effects [11]. Quitenoticeable is the apparent, significant enhancement of these peaksfor the UV-irradiated samples.

The temporal evolution of b-carotene content in the grapes,after their irradiation with UV light, is plotted in Figure 3. The datapoints in the figure represent the average of the TRR-signal associ-ated with the –C@C– m1-stretch mode (peak at �1526 nm). For theirradiated grapes – (blue) square symbols for 300 nm LED, (red)circular symbol for 295 nm LED – the b-carotene content increasesdramatically, reaching a maximum after about 24 h and thendecreasing again, down to about its initial value. For the controlgroup – (blue) triangular symbols – the b-carotene content re-mains mostly constant. In the maximum the b-carotene contentin the irradiated grapes is up to five times higher than in the con-trol (untreated) grapes.

Figure 2. TRR spectra of b-carotene in grapes, using 514.5 nm laser stimulation; allspectra were recorded 24 h after commencing the individual measurement run. Toptrace (blue line) – grape irradiated with LED peaking at 300 nm. Middle trace (redline) – grape irradiated with LED peaking at 295 nm. Bottom trace (black line) –control sample, grape not irradiated. The spectra are offset to each other for clarity.The annotation of b-carotene Raman peaks is according to Tschirner et al. [21]. (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this article.)

Figure 3. Normalized TRR signal for the Raman peak of the C@C stretch mode of b-carotene (1526 cm�1) in grapes, as a function of time elapsed after irradiation withUV LED; (blue) squares – 300 nm LED irradiation; (red) circles – 295 nm LEDirradiation; (black) diamonds – control samples, no LED irradiation. The error barsindicate the standard deviation of five repeat measurements for each group of grapesamples. The bell-shaped curve (dashed line) serves only to guide the eye,indicating the evolution of b-carotene over time, which for the LED-irradiatedsamples peaks after about 24 h, exhibiting a five-fold b-carotene increase withrespect to the control samples. (For interpretation of the references to color in thisfigure legend, the reader is referred to the web version of this article.)

28 A.G. Gonzálvez et al. / Chemical Physics Letters 559 (2013) 26–29

The overall conclusion from the Raman spectral analysis is thata significant enhancement of the b-carotene content is observedsubsequent to UV-B LED irradiation. While this interpretationlooks persuasive, one still may raise the question of whether theTRR-spectra on their own are sufficiently exclusive to make theassignment, or if other, non-laser spectroscopic methods may berequired to substantiate our results. Two findings may be helpfulin this context. Firstly, Bhosale et al. [7] carried out a study inwhich they quantified the nutritionally important carotenoids infruits. When comparing of their data from Resonance Raman (RR)spectroscopy and High-Pressure Liquid Chromatography (HPLC)analysis a very strong correlation between the results from thetwo methods was found. Secondly, in recent work from our labora-tory [11] it was shown that using the very method of TransmissionResonance Raman (TRR) spectroscopy we were able to quantify thecarotenoid content in calibrated standards and carrots root sam-ples. It should be noted that the latter carrot sample has a signifi-cantly higher optical density than the muscatel grapes used here.Thus one may conclude that under the experimental conditions,at which the present investigation was carried out, quantificationshould be equally valid.

Thus, the answer to the above question is affirmative: the mea-sured TRR signal is indeed directly proportional to the fruit’s carot-enoid content.

An interesting aspect to be considered concerns the distributionof the induced carotenoid concentration within the volume of thegrape. In general, carotenoids in grapes exhibit a distinct profile be-tween skin and pulp [16]. Specifically, it has been found that skincontributes roughly three times as much to the b-carotene concen-tration than does the pulp. In this context, one should bear in mindthat the UV-B LED radiation will just penetrate beyond the grapeskin. Therefore, one should expect that the carotenoid enhance-ment stems predominantly from the skin layer even though ourTRR measurements probe the entire grape.

3.2. Biosynthesis of b-carotene

The main steps of the biosynthetic pathway of carotenoids arewell described in the literature, [22] and therefore only a few com-ments are provided here to facilitate the understanding of the re-sults. Carotenoids – like all isoprenoids – are synthesized inplastids (organelles responsible for photosynthesis). One of themost relevant b-carotene precursors is phytoene, the first linearunsaturated chain of 40 carbon atoms, which by introduction offour double-bonds converts to lycopene, whose cyclization leadsto b-carotene. It is also well established that the last pigment isnot the end products but, subsequently, it transforms into severalderivatives, a process that may explain the observed b-caroteneconcentration decline after 24 h.

These complex molecules participate in light-harvesting andphoto-protection from excess light energy, thus quenching tis-sue-damaging free radicals as singlet oxygen species [23].Although light is well recognized as an important factor that influ-ences fruit carotenegenesis, the role of UV-B radiation on carote-noids is poorly understood and was rarely investigated, exceptfor tomatoes for which several studies have been carried out. Forexample, Giuntini et al. [24] found that in some tomato genotypescarotenoid was promoted by UV-B irradiation (280–315 nm). Laz-zeri et al. [25] carried out a study of carotenoid profiling in theflesh and peel of tomatoes under UV-B depletion; the particularspecimen under investigation were a wild-type tomato and a highpigment (hp-1) tomato, a mutant characterized by increased fruitpigmentation. In this latter study it was found that while UV lightexerted a negative modulation effect mostly on lycopene synthesisin wild-type tomatoes, it hardly affected carotenoid accumulationin hp-1.

A.G. Gonzálvez et al. / Chemical Physics Letters 559 (2013) 26–29 29

The observed enhancement of the b-carotene content in grapes,induced by low doses of UV-B radiation peaking around 300 nm,suggests the possibility of the photo-activation of distinct stepsin the outlined biosynthetic pathway, most likely by resonantabsorption in some precursor compound. This speculation seemsto be reasonable; in a comprehensive guide to carotenoid analysisin food Rodriguez-Amaya [26] asserts that the carotenoid that pre-cedes f-carotene in the de-saturation biosynthetic pathway, phyto-ene (with three conjugated double bonds), is the only compoundwhose main absorption band lies within the UV-B range (see Fig-ure 9 in Rodriguez-Amaya [26]).

Following these arguments on may postulate a mechanismwhich encompasses electronic excitation by resonant UV-Babsorption of this b-carotene precursor, which triggers the subse-quent addition of double bonds required for highly conjugatedcarotenoids, such as b-carotene and its derivatives. Although thisinference should be taken as a working hypothesis only, requiringfurther investigation to be fully confirmed, it is worth noting thatexploiting electronic energy excitation is a common practice in la-ser chemistry if one wishes to enhance chemical reaction yields[5,27].

4. Conclusion

There are three aspects of the present Letter that we consider tobe of major relevance. In the first place, one finds a remarkableboost of b-carotene in grapes, applying only very low doses ofUV-B light. This may become of great important in ‘functional’ foodresearch (such as processed foods or foods fortified with health-promoting additives), supporting the needs of an increasinglyhealth-conscientious society. Furthermore, increased amounts ofcarotenoids may help to extend the shelf life of agricultural pro-duce, thus reducing the substantial losses incurred by excessiveripening or aging.

Secondly, the use of LED technology is quite novel and promisesgreat potential. In particular, it opens the way to potentially imple-ment very low-cost commercial protocols for post-harvestimprovement of fruit, food and vegetable quality.

Last but not least, the third aspect worth noting is the use ofTransmission Resonance Raman spectroscopy to monitor the b-carotene content in grapes, without having to resort to any samplepreparation or extraction method. This is afforded by the highselectivity and sensitivity of the TRR technique (resonant characterof the laser excitation wavelength), but that at the same time inter-fering fluorescence emission is substantially suppressed.

Clearly, the combined recourse of these factors may well assistin resolving various obstacles in an efficiency-conscious foodindustry, and in food quality control, irrespective of other, morefundamental-oriented research objectives.

In this context it should be pointed out that performing Ramanspectroscopy on whole grapes (or any other fruit) is a non-trivialtask, specifically if one endeavors to measure trace compounds dis-tributed in the entire volume, and not only on the surface, and if

one does not wish to compromise the integrity of the fruit itself.To best of our knowledge this is the first time that TransmissionResonant Raman spectroscopy of a whole fruit (a grape in this case)has been successfully implemented in a non-invasive manner.

Finally, it should be remarked that the use of the low dose UV-BLED irradiation in combination with the Transmission ResonanceRaman spectroscopy could be viewed as a promising techniqueto investigate the biochemistry and physiology of complex sys-tems, topics that currently attract increasing scientific attention.

Acknowledgments

Financial support from the Ministerio de Ciencia y Tecnología ofSpain (Grants CTQ2007-61749 and CTQ2011-23218), and the Gov-ernment of the Madrid Region is gratefully acknowledged. A.G.Gonzalvez acknowledges a (FPI) pre-doctoral fellowship from theMinisterio de Ciencia e Innovación of Spain.

References

[1] P. Matousek, Chem. Soc. Rev. 36 (2007) 1292.[2] M.J. Pelletier, Analytical Applications of Raman Spectroscopy, Blackwell

Science, Oxford, U.K., 1999.[3] E. Smith, G. Dent, Modern Raman Spectroscopy: A Practical Approach, John

Wiley & Sons, Chichester, U.K., 2005.[4] B. Schrader (Ed.), Infrared and Raman Spectroscopy, VCH, Weinheim, Germany,

1995.[5] H.H. Telle, A. González Ureña, R.J. Donovan, Laser Chemistry: Spectroscopy,

Dynamics, and Applications, John Wiley & Sons, Chichester, U.K., 2007.[6] M.E. Darwin, I. Gersonde, M. Meinke, W. Sterry, J. Lademann, J. Phys. D: Appl.

Phys. 38 (2005) 2696.[7] P. Bhosale, I.V. Ermakov, M.R. Ermakova, W. Gellermann, P.S. Bernstein, J. Agric.

Food Chem. 52 (2004) 3281.[8] E.V. Efremov, F. Ariese, C. Gooijer, Anal. Chim. Acta 606 (2008) 119.[9] K. Buckley, P. Matousek, J. Pharm. Biomed. Anal. 55 (2011) 645.

[10] M. Boiret, Y.-M. Ginot, Spectrosc. Eur. 23 (6) (2011) 6.[11] A.G. Gonzálvez, A. González-Ureña, Appl. Spectrosc. 66 (10) (2012) 1163.[12] F. Bourgaud, A. Gravot, S. Milesi, E. Gontier, Plant Sci. 161 (2001) 839.[13] Y. Nakagawa, S. Fujiwara-Fukuta, T. Yorimitsu, S. Tanaka, R. Minami, L.

Shimooka, H. Nakagoshi, Mech. Dev. 128 (2011) 258.[14] W. Russell, G. Duthie, Proc. Nutr. Soc. 70 (2011) 389.[15] P. Guedes de Pinho, A.C. Silva Ferreira, M. Mendes Pinto, J. Gómez Benítez, T.A.

Hogg, J. Agric. Food Chem. 49 (2001) 5484.[16] J. Marais, C.J. van Wyk, A. Rapp, S. Afr. J. Enol. Vitic. 2 (1991) 64.[17] C. Oliveira, A.C. Silva Ferreira, M. Mendes Pinto, T.A. Hogg, F. Alves, J. Agric.

Food Chem. 51 (2003) 5967.[18] A. Razungles, Z. Gunata, S. Pinatel, R. Baumes, C. Bayonove, Sci. Des Aliments

13 (1993) 59.[19] A.G. Gonzálvez, A. González Ureña, R.J. Lewis, G.J. van der Zwan, Phys. Chem. B.

116 (2012) 2553.[20] N. Tschirner, M. Schenderlein, K. Brose, E. Schlodder, M.A. Mroginski, P.

Hildebrandt, C. Thomsen, Phys. Stat. Sol. B 245 (2008) 2225.[21] N. Tschirner, M. Schenderlein, K. Brose, E. Schlodder, M.A. Mroginski, C.

Thomsen, P. Hildebrandt, Phys. Chem. Chem. Phys. 11 (2009) 11471.[22] J. Hirschberg, Curr. Opin. Plant Biol. 4 (2001) 210.[23] G. Giuliano, R. Aquilani, S. Dharmapuri, Trends Plant Sci. 5 (2000) 406.[24] D. Giuntini, G. Graziani, B. Lercari, V. Fogliano, G.F. Soldatini, A. Ranieri, J. Agric.

Food Chem. 53 (2005) 3174.[25] V. Lazzeri, V. Calvenzani, K. Petroni, Ch. Tonelli, A. Castagna, A. Ranieri, J. Agric.

Food Chem. 60 (2012) 4960.[26] D. Rodriguez-Amaya, A Guide to Carotenoid Analysis in Foods, ILSI Press,

Washington, DC, USA, 2001.[27] R.D. Levine, Molecular Reaction Dynamics, Cambridge University Press,

Cambridge, U.K., 2005 (Chapter 7).