A Colored Avocado Seed Extract as a Potential Natural Colorant

Abstract

There is an increasing consumer demand for and scientific interest in new natural colorants. Avocado (Persea americana) seed when crushed with water develops an orange color (λ_{max_visible} = 480 nm) in a time-dependent manner. Heat treatment of the seed prevented color development, whereas the addition of exogenous polyphenol oxidase (PPO), but not peroxidase restored color development. Color development was also inhibited by the addition of tropolone, an inhibitor of PPO. Color formation resulted in a decrease in the concentration of polyphenols indicating utilization for color formation. The orange color intensified as the pH was adjusted from 2.0 to 11.0, and these changes were only partially reversible when pH was adjusted from 7.5 to 11.0 in the presence of oxygen, but completely reversible when the pH was changed in the absence of oxygen. The color was found to be stable in solution at −18 °C for 2 mo. These results suggest that the avocado seed may be a potential source of natural colorant, and that color development is PPO-dependent.

Introduction

Color plays a key role in determining the expectations and perceptions of consumers with respect to food (Delgado-Vargas and Paredes-Lopez 2003a, 2003b). It is one of the most obvious characteristics of a food and, if not appealing, negatively impacts consumer acceptance. Color is interrelated with flavor intensity and sweetness and salinity sensations and may also indicate the safety of a food. Although artificial colorants have a long history of use, and are easy to produce, stable, less expensive, and have better coloring properties than natural colorants, consumers have increasingly begun to consider synthetic colorants undesirable. Consequently, there has been increased effort to discover new natural alternatives (Socaciu 2008).

By definition, a natural pigment is one that is synthesized by and accumulated and/or excreted by living cells. It may also be synthesized by cells under stress or by dying cells (Hendry and Houghton 1996). Historically, natural pigments (e.g., saffron) were used to color food products. Currently, 26 natural colorants including anthocyanins, curcumin, carminic acid, lycopene, betanin, paprika, and saffron are permitted for use as exempt colorants in the United States. Like their synthetic counterparts, natural pigments can be formulated as dyes (hydrophilic powders or lipophilic oleoresins) or lakes (water insoluble forms, formulated by adsorbing dyes on salts) (Socaciu 2008; Amiot-Carlin and others 2008). The growth in the color segment will largely come from natural colorants, with the market expected to reach $773 million by 2014 (Rice 2010). The market for natural colorants was $250 million in 2006 and had an estimated annual growth rate of 2.9% (Trim 2006).

Whereas the synthetic colorants generally have similar properties, natural colorants, even if they are of the same shade, may differ in terms of chemistry and functionalities. For example annatto and turmeric differ in their functionalities. Annatto, as norbixin, will precipitate at pH below 5.0 whereas turmeric will not (foodproductdesign.com 2011). Thus it is crucial to understand the properties of any prospective natural colorant to decide about its applications. The factors that deteriorate colorant stability and efficacy include the presence of temperature, light, oxidizing and reducing agents, acids or time of storage (Socaciu 2008). There is still a lack of natural colors to completely replace synthetic colorants, and hence, efforts to search more options are needed (Hazen 2011).

Avocado (Persea americana Lauraceae) is an important tropical crop which is rich in unsaturated fatty acids, fiber, vitamins B and E, and other nutrients (Gomez Lopez 1998). The Hass variety is the most commonly grown. Total U.S. avocado production during the 2004/05 season was 162721 tons with 90% originating in California (World Horticultural Trade 2006). Mexico, the world’s largest producer, supplied 1.14 million tons in 2007 (McLeod 2008). The seed accounts for 16% of total avocado weight, and is an under-utilized resource (Ramos-Jerz 2007). Ethnopharmacological studies of the Aztec and Maya cultures have reported the use of decoctions of avocado seeds for the treatment of mycotic and parasitic infections (Kunow 2003). Seeds have also been reported for use against diabetes, inflammation, and gastrointestinal irregularity (Kunow 2003; Ramos-Jerz 2007). The powdered form has been used for skin eruptions and to cure dandruff (Morton and Dowling 1987).

Avocado seeds have more antioxidant activity and polyphenol content than the pulp (Soong and Barlow 2004; Wang and others 2010). Phytochemical studies on avocado seeds have identified various classes of natural compounds such as phytosterols, triterpenes, fatty acids, furanoic acids, abscisic acid, proanthocyanidins, and polyphenols (Ding and others 2007; Leite and others 2009). Wang and others (2010) have reported the presence of catechin, epicatechin, and A- and B-type procyanidin dimers and trimers, tetramers, pentamers, and hexamers in the seed.

Our laboratory has observed that avocado seed when ground with water and incubated in the presence of air produces a bright orange color (previously unreported). The aim of the present work was to characterize the color formation, to study the mechanism of color formation and the color properties of the extract.

Material and Methods

Reagents

Ripened avocado (Persea americana, Hass variety) were sourced locally and stored at 4 °C until used. No change in seed color was observed during storage. Mushroom polyphenol oxidase (PPO) (tyrosinase), horseradish peroxidase, hydrogen peroxide, and tropolone were purchased from Sigma Chemical Co. (St. Louis, Mo., U.S.A.). DMSO (Dimethyl sulfoxide) was purchased from EMD Chemicals (Gibbstown, N.J., U.S.A.). Folin Ciocalteu reagent was purchased from Fluka (Buchs, Switzerland). Gallic acid was purchased from Alfa Aeser (Lancaster, Pa., U.S.A.). All other reagents were of the best commercial grade available.

Preparation of colored extract

Avocado seeds were separated from fruit, washed and peeled. Seeds were weighed, cut with a knife, and ground in 0.7 vol of deionized (DI) water in a Waring blender for 60 s at high speed. The resulting paste (pH 6.4) was spread to 1 cm thickness, on a flat surface (to enable uniform exposure to air) and incubated at 24 °C for 35 min. The paste was mixed with a spatula 2 to 3 times during this time. The colored paste was transferred to a beaker, an equal volume of methanol was added, and the mixture was sonicated for 20 min in a Branson 3510 sonicating water bath (Danbury, Conn., U.S.A.). An additional 2 vol of methanol was added, and the mixture centrifuged at 1200 × g for 10 min. The supernatant was collected and dried under vacuum to remove the methanol. Residual water was then removed by freeze drying, and the resultant powder was stored in a desiccator at −20 °C.

Kinetics of color formation

The kinetics of color formation was studied using 2 methods: paste was incubated for different time intervals, the color extracted and absorbance of the extract measured or paste was incubated and CIE (International Commission on Illumination) $L^{\mathrm{*}}{a}^{\mathrm{*}}{b}^{\mathrm{*}}$ values were read at different time points.

For the absorbance measurement, pastes were prepared as described above with the modification that 10 g samples were incubated for 0.5, 2, 5, 10, 15, 20, 25, 30, and 35 min after which they were combined with methanol (3 vol), and then sonicated and centrifuged as above. Visible absorbance spectra were recorded immediately (λ = 380 nm to 700 nm) using an Agilent 8453 spectrophotometer (Agilent Technologies, Santa Clara, Calif., U.S.A.) by placing samples in disposable 1.5 mL cuvettes (Plastibrand, Wertheim, Germany).

To measure $L^{\mathrm{*}}{a}^{\mathrm{*}}{b}^{\mathrm{*}}$ values, seeds were weighed, cut, and ground in 1 vol deionized (DI) water in a Waring blender for 60 s at high speed. The ground paste was spread in a Petridish (CM-A128, Minolta) and $L^{\mathrm{*}}{a}^{\mathrm{*}}{b}^{\mathrm{*}}$ values recorded at specified time intervals using a Konica Minolta chromameter CM-3500 d (Konica Minolta, Ramsey, N.J., U.S.A.) in reflectance mode. The paste was mixed thoroughly just before taking measurements. The conditions were a ‘D’ illuminant with an 8° observer. Color difference ($\mathrm{\Delta }E$) was calculated using the following formula:

$$\mathrm{\Delta }E=\sqrt{{\left(L-L_0\right)}^2+{\left(a-a_0\right)}^2+{\left(b-b_0\right)}^2},$$

where $L_0,a_0$, and $b_0$ are $L,a$, and $b$ values, respectively, for the first time point measured.

Role of PPO in color formation

Avocado seeds were prepared and ground as above. An additional 1.5 vol of DI water was added and mixed, and the paste was kept at 100 °C for 30 min to destroy endogenous enzyme activity. The paste was cooled to 24 °C in a water bath. To 25 g of cooled paste 1000, 2500, or 5000 U of mushroom PPO (prepared in phosphate buffered saline:NaCl, 136 mM; KCl, 2.7 mM, Na₂HPO₄, 10 mM; KH₂PO₄, 1.7 mM at pH 6.8) was added and the mixture spread on a flat surface as described above. After 35 min, methanol (3 vol) was added to the samples and they were sonicated and centrifuged as above. Absorbance spectra were obtained as described above. Positive control samples were prepared similarly but were not subjected to heat treatment. Negative control samples were subjected to heat treatment but exogenous PPO was not added.

Involvement of PPO was examined by use of tropolone, an inhibitor of PPO. Tropolone (58 μM) was added during grinding of the seed, after which the paste was spread on a flat surface for 35 min and processed as above. For studies with exogenous PPO, tropolone were added to the inactivated paste prior to the addition of PPO and the reaction was carried as outlined above.

Role of peroxidase in color formation

Avocado seeds were prepared, ground and heat inactivated as above. To 25 g of cooled paste, 1000, 2500, or 5000 U horseradish peroxidase (in phosphate buffered saline, pH 6.8) and 1.9 g H₂O₂ per 1000 U were added. The paste was incubated and processed as described for exogenous PPO experiments.

Phenolic content of the colored extract

To the freeze-dried extract, 50 vol hexane was added and the mixture was shaken for 30 min in an orbital shaker at 600 rpm to remove residual lipids. Hexane was removed by centrifugation at 1200 × g for 10 min. The marc was then extracted with 50 vol of methanol:ethyl acetate (1:1, v/v) for 1 h in the orbital shaker at 600 rpm. The supernatant was separated by centrifugation, and dried under vacuum at 24 °C. The marc was then extracted with 50 vol of DI water at 600 rpm for 1 h, centrifuged and the aqueous supernatants dried under vacuum at 24 °C for 12 h. Both the methanol:ethyl acetate and water fractions were used for determination of phenolic content by the modified method of Singleton and Rossi (1965). Aliquots of the samples of appropriate dilutions were prepared in methanol and were combined with 790 vol of DI water and 5 vol of Folin–Ciocalteu reagent. The solution was mixed and 15 vol of 15% sodium carbonate solution was added to the sample. Samples were mixed again and the absorbance determined at 765 nm after 2 h. Gallic acid was prepared in methanol and was used as the standard. The total amount of phenolic compounds was expressed as gallic acid equivalent, GAE (mg/g). Uncolored extract, prepared in a method analogous to the colored except that methanol was added immediately after grinding the seeds, was also tested for phenolic content for comparison.

Effect of pH on color of extract

Phosphate buffered saline at pH 2.0, 4.0, 6.0, 7.5, 9.0, and 11.0 was prepared in 20% aqueous methanol and combined with freeze dried extract at a final concentration of 2 mg/mL. The samples were vortexed and then centrifuged at 1200 × g for 10 min. Visible absorbance spectra were recorded as described above. Hunter Lab values were determined using a Konica Minolta chromameter CR400 by placing the sample in a 1 mm path length rectangular quartz cuvette (Fisherbrand, Pittsburgh, Pa., U.S.A.) against a white background. A ‘C’ illuminant and an 2° observer were used. For the determination of ΔE at different pH values, L₀, a₀, and b₀ values were that of a blank measured by placing DI water in the cuvette.

To determine if the effect of pH on the color of the extract was reversible, extract at a concentration of 2 mg/mL in phosphate buffered saline containing 20% aqueous methanol (pH 7.5) was prepared as described above. The visible absorbance spectra were determined. The pH of the solution was adjusted to 11.0 by the addition of NaOH and the spectra determined again. The pH was readjusted to 7.5 by the addition of HCl and the visible absorbance spectra were recorded a final time. A similar experiment was conducted by adjusting the pH from 7.5 to 2.0 and back to 7.5. Both sets of experiments were also conducted while purging nitrogen through the samples at a rate of 5 mL/s. This was done to assess the role of oxygen on the effects of pH on color. The amount of dissolved oxygen after N₂ flush was determined using a Thermo-scientific portable DO meter, Orion 087003 (Orion, Barrington, Ill., U.S.A.) and found to be reduced to 0.15 ± 0.01 mg/L. The addition of acid and base to change the pH changed the final volume of the solutions by less than 2%.

Color stability

Freeze-dried extract was dissolved in phosphate buffered saline in 20% aqueous methanol at pH 7.5 at a concentration of 2 mg/mL. The samples were centrifuged at 1200 × g for 10 min to remove any undissolved solids. The supernatant was transferred to new tubes, sealed and kept at −18 °C, 4 °C, 24 °C, and 40 °C. Aliquots of samples were periodically removed and the absorbance at 480 nm and ΔE values measured.

Statistical analysis

Results are shown as mean of 3 independent determinations. Error bars represent the standard error of mean (SEM) or the standard deviation (SD) as indicated in the figure legends. Differences between means were tested for significance by one-way analysis of variance (ANOVA) with Dunnett’s posttest significance or two-way ANOVA with Bonferroni posttest as appropriate. Significance was achieved at P < 0.05. All statistical analyses were performed using Graphpad Prism software (Graphpad Software, San Diego, Calif., U.S.A.).

Results

Kinetics of color formation

The kinetics of color formation in avocado seed were followed by measuring changes in visible absorbance intensity and by determining $L^{\mathrm{*}}{a}^{\mathrm{*}}{b}^{\mathrm{*}}$ values and color difference (ΔE, Figure 1). Figure 1A shows visually how the orange color intensity increased with time. The λ_max of the color was 480 nm (Figure 1B) and absorbance increased over the time course of the experiment in an exponential fashion approaching an asymptote by 20 min (Figure 1C). A similar time-dependent increase in ΔE values was observed (Figure 1D). Within 0.5 min there was noticeable difference in color compared to water.

Figure 1–

Kinetics of color formation in avocado seed paste. (A) A visual change in color of the paste was observed with the time of incubation. Representative results of 3 independent experiments. (B) Spectrophotometric analysis at 35 min showed a λ_max at 480 nm. (C) Absorption at 480 nm (mean ± SD of 3 replicates) and (D) ΔE increased with the time of incubation (mean of 2 independent experiments).

Role of PPO in color development

We hypothesized that the color development in crushed avocado seed was enzyme dependent and observed that heat treatment prevented color development (negative control, Figure 2A) compared to untreated seed (positive control, Figure 2A). Since PPO is involved in color production in number of fruits and vegetables (Munoz and others 2007), and is present in avocado (Gómez-López 2002), we investigated whether addition of exogenous PPO would restore color formation. Addition of increasing amounts of mushroom PPO to heat-inactivated crushed seed resulted in restoration of color development (Figure 2A). To further demonstrate the role of PPO in color formation, reactions were also carried out in the presence of tropolone. The addition of tropolone prevented the formation of color when added during grinding and likewise inhibited color formation in PPO-supplemented heat-inactivated paste formation of color (Figure 2A).

Figure 2–

Role of PPO and peroxidase in the development of color in avocado seed paste. (A) The absorbance at 480 was determined in heat-inactivated avocado paste following addition of increasing amount of exogenous PPO. Tropolone (marked ‘T’ in the figure) was added to determine if selective inhibition of PPO prevented color formation. (B) The absorbance at 480 nm was determined after the addition of exogenous peroxidase and H₂O₂ to heat-inactivated paste. The data represent the mean ± SD of 2 independent experiments.

Since the enzyme peroxidase can also be involved in oxidative reactions in fruits and vegetables (Chisari and others 2007), its role was explored. The addition of exogenous peroxidase and H₂O₂ to the inactivated paste did not result in significant increases in absorbance at 480 nm compared to heat-inactivated control samples (Figure 2B). Visually, the peroxidase paste had a slight brown color, but did not develop the orange color characteristic of PPO-treated samples.

Phenolic content

In order to assess the impact of color production on seed phytochemistry, the total polyphenol content of the extracts were determined. The phenolic content of the colored extract and the uncolored extract was found to be 219.4 ± 4.5 mg/g GAE and 283.2 ± 5.8 mg/g GAE, respectively (Figure 3). A higher concentration of phenolics was found in the methanol:ethyl acetate fraction than in the water fraction for both the colored and uncolored extracts (Figure 3).

Figure 3–

Comparison of the phenolic content of the colored and the uncolored avocado seed extract. Higher total phenolic concentrations were observed in the uncolored extract. This difference was statistically significant (P < 0.05). Mean ± SEM of 3 independent experiments.

Effect of pH on color

The intensity and the hue of the colored avocado seed extract in solution were increased by increasing the pH of the solution. Increases were observed across the visible spectrum, and at pH 11.0, a second visible absorbance maxima (λ = 440 nm) and a maxima in the near UV portion of the spectra developed (Figure 4A). Absorbance at 480 nm increased slightly between pH 2.0 and 9.0, reaching a point of inflection, followed by a rapid increase to pH 11.0 (Figure 4B).

Figure 4–

Effect of pH on the color of the avocado seed extract. (A) Absorption spectra at different pH values show a λ_max at 480 nm at pH 7.5, 9.0, and 11.0; at pH 11.0 an additional λ_max at 440 nm was observed. Representative results of 3 independent experiments. (B) Absorption at 480 nm and ΔE values were seen to increase as a function of pH. Mean ± SD of 3 replicates.

ΔE also increased as a function of increasing pH (Figure 4B). The change in ΔE values was largely due to decreased lightness, L (from 89.1 to 80.0), and increased b (yellowness) (from 2.0 to 24.8) from pH 2.0 to 11.0. The value of a (redness) also increased but to a lesser extent (from −0.1 to 1.8). ΔE less than 1.5 does not represent a significant change in color compared to baseline whereas ΔE greater than 5 represents a readily observable change (Obón and others 2009).

To determine if the effect of pH on color was reversible, the pH of the colored avocado seed extract in solution was increased from 7.5 to 11.0 and then reduced back to 7.5. As the pH was increased in the presence of air, an increase in absorbance intensity in the visible region was observed as was the appearance of a new maxima (λ = 393 nm, Figure 5A). When the pH was reduced back to 7.5, absorbance intensity in the near UV range was reversed, but larger differences remained in the visible portion of the spectrum (Figure 5A). When similar experiments were conducted under a nitrogen atmosphere, the shape of absorbance spectra at pH 11.0 was different from the spectrum at the same pH under air, and the only changes occurred in the near UV. Changes in the spectrum induced by increasing the pH to 11.0 under nitrogen were also completely reversed when the pH was reduced back to 7.5. Much smaller changes in absorbance spectra were observed when pH was reduced from 7.5 to 2.0 (Figure 5B). A complete reversal of the changes was observed both in the presence and absence of oxygen when pH was increased back to 7.5.

Figure 5–

Effect of pH change and the presence or absence of air on the rehydrated avocado seed extract. (A) When the pH was increased from 7.5 to 11.0, in the presence of oxygen, the absorption intensity increased and 2 more absorbance maxima emerged at pH 11.0. When pH was decreased back to 7.5, only partial restoration of color character was observed. When pH was increased to 11.0 under a nitrogen atmosphere, an increase was observed in near UV range. On reducing the pH back to 7.5, complete reversal of absorbance intensities was observed. Representative results of 3 independent experiments. (B) When pH was reduced from 7.5 to 2.0 and adjusted back to 7.5, similar effects were observed under both air and nitrogen atmospheres. Representative results of 3 independent experiments.

Stability of the colored extract

At pH 7.5, the color of the extract in 20% aqueous methanol was unstable at 40 °C and a sharp increase in absorbance at 480 nm was observed during the first 12 h of storage after which the intensity decreased (Figure 6A). Samples stored at 24 °C showed similar changes although the kinetics were slower. An increase in absorbance occurred until day 10 followed by a subsequent decrease. At both 4 °C and −18 °C, the absorbance of the samples increased at much slower rates than at the higher temperatures. No significant change was observed until day 60 in the samples stored at −18 °C. The ΔE values of the extract stored at pH 7.5 showed similar trends to those observed spectrophotometrically (Figure 6B). For samples stored at −18 °C at pH 7.5, the ΔE value was 1.65 and no visible change in color was discerned. However, samples stored at higher temperatures (4 °C, 24 °C, and 40 °C) had ΔE values greater than 5 over the time period of storage study and had noticeable changes in color.

Figure 6–

Effect of storage on color of the rehydrated avocado seed extract. (A) Absorbance intensity at 480 nm and (B) ΔE were determined during storage for 60 d at pH 7.5 at −18 °C (■), 4 °C (●), 24 °C (▲), and 40 °C (▼). Mean ± SD of 3 replicates.

Discussion

In the present study, we characterized the color production in avocado seed upon maceration and exposure to air, the mechanism of color formation and the color properties of resultant extract. A stable orange color was produced when avocado seeds were ground and incubated at 24 °C. The color production occurred rapidly with a noticeable change produced as early as 0.5 min of incubation. Kinetic analysis showed that intensity of the color formation increased in an exponential fashion and began to approach an asymptote at 20 min. Such kinetics of formation are typical of an enzyme catalyzed reaction. For example, Arias and others found that pear (Pyrus communis) PPO-mediated oxidation of dihydroxyphenylalanine (DOPA) resulted in a similar trend in absorbance at 420 nm (Arias and others 2007). We observed that heat treatment prevented color formation, whereas the addition of exogenous mushroom PPO restored color formation. PPO from avocado pulp has been found to be very active and resistant to treatments including heat, making avocado products very prone to enzymatic browning (Gómez-López 2002), but there is scant information on PPO from avocado seed. The orange color formed in the seed may have resulted from a particular substrate present in the avocado seed or inhibition of PPO by the products formed during an intermediate stage of the reaction.

PPO catalyzes both the oxygen-dependent hydroxylation of monophenols to their corresponding O-diphenols and the oxidation of O-diphenols to their cognate O-quinones (Lin and others 2010). The quinones then may polymerize into red, brown, or black pigments depending on conditions like the nature and amount of endogenous phenolic compounds, the presence of oxygen, reducing substances, or metallic ions, the pH and temperature, and the activity of the PPO (Dogan and others 2006). Browning reactions in fruits and vegetables occur when tissues are damaged and PPO is released. Here, we observed that the orange colored extract was stable during observation window explored and did not apparently produce highly polymerized melanoidins.

The total phenolic content of methanol:ethyl acetate and water extracts in freeze-dried colored extract was calculated to be 219.4 mg/g. Soong and Barlow (2004) used ethanol:water at high temperature for extraction of phenolics of avocado seed and calculated the phenolics as 88.2 mg GAE/g. Wang and others (2010) used acetone:water:acetic acid (70:29.7:0.3, v/v/v) and found the values to be 51.6 mg/g GAE for the seed of Hass variety of avocado, whereas the corresponding values for pulp was only 4.9 mg/g GAE. In the present study, the phenolic content of the colored extract was 22.5% lower than the content in uncolored extract. The phenolic compounds present in the colored extract may contribute to functional attributes. The reduction in phenolic content during color development may be due to oxidation of phenolic compounds by the PPO. During processing of fresh grapes to raisins, most of the 2 major hydroxycinnamic acids and all of the procyanidins and flavan-3-ols were lost. The reasons for this loss were likely to be both enzymatic and nonenzymatic (Karadeniz and others 2000).

Although often associated with undesirable browning reactions, PPO has also shown to produce attractive colors in some systems. PPO-catalyzed conversion of catechins to theaflavins in the production of black tea is responsible for the characteristic orange color of that beverage (Balentine and others 1997). A yellow-colored product was synthesized through oxidation of ferulic acid with laccase, an enzyme functionally similar to PPO. It was found found that in a biphasic system containing ethyl acetate a stable yellow-colored product is formed whereas in an aqueous system browning occurred. This difference was attributed to the incomplete activity of PPO in the medium containing ethyl acetate (Mustafa and others 2005). Catechin was oxidized at pH 7.5 using PPO and it was observed that the absorbance maxima of the product was 430 nm which corresponded to a yellow-colored product (Jiménez-Atiénzar and others 2004). A yellow dye from phlorodzin, a flavonoid specific to apples, was synthesized in the presence of PPO and oxygen (Guyot and others 2007). The product which has a brilliant yellow color with nuances depending on the pH can be incorporated into water based foods such as beverages (juices, syrups) and confectionary creams (Socaciu 2008).

The absorbance at 480 nm and the ΔE values increased as the pH increased, with a rapid increase after pH 8.0. These rapid changes could be occurring due to dissociation, or other reactions like ring opening or reactions occurring due to alkaline conditions Also, these data show that the colorant from avocado seed may be usable at higher pH. Comparatively, anthocyanins, common natural colorant, give a stable hue only at pH values lower than 5 (Paredes-Lopez 2003b; Hazen 2011). The color change with increased pH under oxygen was irreversible. These data suggest that the colored compounds produced as a result of oxidation by the PPO may be further oxidized under high pH and oxygen. These observations imply that if the extract is to be used as a food colorant, the pH of the food and the conditions of storage must be considered.

The color of the extract was stable at pH 7.5 for 2 mo when stored at −18 °C. At higher temperatures significant changes in the color were observed over shorter time periods. These results suggest that the extract could be used in products stored at low temperatures, but a system for stabilizing the color at higher temperatures may be needed. Since the current studies were conducted in the presence of air, it is possible that the extract would be more stable at higher temperatures if storage conditions included an inert atmosphere. However, the practicality of anaerobic packaging will depend on the application of interest. Walkowiak-Tomczak and Czapski (2007) observed that in a red cabbage anthocyanin model system, pigment loss was reduced by greater than 50% when samples were stored in the absence of oxygen.

Crucial to its proposed color applications in foods is the extract’s safety. A preliminary safety profile of the seed has been conducted in rats. An aqueous extract of fresh avocado seed (2 to 10 g/kg) was administered to 6 groups with 4 rats per group by oral gavage. The animals were fed rat pellet ad libitum and exposed to natural light at room temperature. No signs of depression, writhing, diarrhea, hypermotility, and aggression were observed in the treatment groups in the 2 subsequent wk. The same group examined subchronic toxicity by administering 2.5 g/kg for 28 d. No significant changes were observed in the hematological parameters of the treated mice. Similarly, there was no significant change in plasma markers of hepatotoxicity (i.e. alanine aminotransferase, aspartate aminotransferase) (Ozolua and others 2009). Further safety studies of the colored extract used in our studies are needed to assess its potential application as a food colorant.

Conclusions

In conclusion, this study examined a colored extract produced enzymatically in avocado seed. Avocado seeds are not currently commercially useful and represent a large waste stream (Ramos-Jerz 2007). Their application as a source of natural colorants could be of significant commercial value. Because of its high phenolic content; the colored extract may have additional functional attributes which should be explored.

Practical Application:

There is growing public and scientific interest in the development of natural alternatives to synthetic colorants in foods. Extracts of turmeric, paprika, and beets are examples of food-derived natural colorants. Avocado seeds, which represent an under-utilized waste stream, form a stable orange color when crushed in the presence of air. Our data indicate that avocado seed represents a potential source of new natural colorants for use in foods.