ECOLOGY and ECONOMIC POTENTIAL of SECONDARY METABOLITES PRODUCED by YAUPON HOLLY, Ilex Vomitoria

ECOLOGY AND ECONOMIC POTENTIAL OF SECONDARY METABOLITES PRODUCED BY YAUPON HOLLY, Ilex vomitoria

MATTHEW J. PALUMBO

A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2006

Matthew J. Palumbo

To my mother who has supported my every interest …and to the memory of my father who instilled in me a deep respect for Nature.

ACKNOWLEDGMENTS

I would like to thank my advisor Jack Putz who not only handed me a great master’s project and helped me get into graduate school but who also kept on my ass and stuck with me throughout this process despite my persistent bitching and moaning and tortoise-like work pace.

Jack and his family, Claudia, Juliana, and Antonio have graciously welcomed me (and my dog!) into their home and on their beautiful land on many occasions, which made this an especially enjoyable graduate school experience. I would also like to thank Steve Talcott, who although is not now an official member of my committee, has nevertheless been a great chemistry teacher and an invaluable source of knowledge that has greatly enhanced my Master’s project. I would also like to thank Michelle Mack who sat on my master's committee for providing useful comments on my research and manuscript drafts and Rick Stepp for his willingness to join my committee at such a late stage of my master’s project.

I also thank other faculty members in the Department of Botany who enhanced my education. In particular, I thank Walter Judd whose taxonomy classes were amazing whirlwinds through the incredible floral diversity on this planet; and Steve Mulkey whose plant ecophysiology course gave me a broader perspective of the physiological and evolutionary mechanisms inherent to the study of ecology. My fellow graduate students were also a great help and inspiration. I particularly thank Morgan Varner for helping me to understand the dynamics of the first ecosystem that I ever really came to (somewhat) understand. I also thank

Eddie Watkins for teaching me how to measure light with a fish-eye lens and computer software, which has become a critical component of my research. I thank Kim and Jorge (from Steve

Talcott’s lab) for helping me get through my last round of caffeine assays.

André Khuri from the Department of Statistics provided invaluable assistance interpreting data for chapter 1. I also thank Jenny Schafer for “teaching” me how to use Sigma Plot, which

was used to plot the graphs in chapter 1. And I thank Paulo Brando, the new ‘R’ guru in Jack

Putz’s lab, for last-minute help with statistics and graphing. I also thank Ann Wagner who is the

botany lab manager for her help and patience with my technological ineptitude. Also thanks to

Isabel Meister for help with setting up the campus-plant study. I also thank all my friends in

Gainesville from the Department of Botany, other departments at the University of Florida, and

elsewhere. Without the sharing of experience that comes with friendship, this degree would have been impossible. I particularly want to thank Jason and Nicole Frederick for all their support and for providing me with a home (and not just a place to lay my head) here in Gainesville.

I would like to thank those students in my laboratory sections and lectures who made my

TA-ship a rewarding way to pay the bills. My gratitude goes to these students for their forgiveness, willingness to put up with my idiosyncrasies, laughter at my pathetic attempt at jokes, and the general challenge they represented. I only hope they learned as much from me as I did from them.

Finally, I thank my family. I thank my brother who encouraged me to return to school when I was only considering it 6 years ago. I thank my grandfather who never hesitated with

financial assistance for my education when it was needed and who has allowed me to share my

passion for plants and ecology with him. And I thank my mother who always supported me and

encouraged me to stay with the program on days when it felt overwhelming. I can never thank

her enough.

Finally, I thank Vito, this man’s best friend, who always greets me at the end of the day with his smiling enthusiasm and zest for life whether I can return it or not.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES...... 7

LIST OF FIGURES ...... 8

CHAPTER

1 NITROGEN FERTILIZER AND GENDER EFFECTS ON THE SECONDARY METABOLISM OF YAUPON, A CAFFEINE-CONTAINING NORTH AMERICAN HOLLY...... 11

Introduction...... 11 Materials and Methods ...... 14 Plant Material and Nitrogen Fertilization...... 14 Chemical Analyses ...... 15 Statistical Analyses...... 16 Results...... 16 Methylxanthine Alkaloids ...... 16 Phenolics...... 16 Nitrogen Content ...... 17 Antioxidant Capacity...... 18 Discussion...... 18 Conclusions...... 24

2 Ilex vomitoria: AN OVERLOOKED NORTH AMERICAN CAFFEINE SOURCE...... 28

Introduction...... 28 Caffeine and Conservation ...... 28 History of Yaupon Tea ...... 28 Yaupon Chemistry and Production ...... 31 Materials and Methods ...... 33 Experimental Design ...... 33 Chemical Analyses ...... 33 Measurement of Light Availability ...... 35 Statistical Analysis ...... 35 Results...... 35 Conclusions...... 38

LIST OF REFERENCES...... 47

BIOGRAPHICAL SKETCH ...... 54

LIST OF TABLES

Table page

1-1 ANOVA results for foliar concentrations of methylxanthine (MX) alkaloids in Ilex vomitoria...... 25

1-2 ANOVA results for foliar concentrations of phenolic compounds in Ilex vomitoria...... 25

2-1 Mean ± 1 SE ORAC (oxygen radical absorbance capacity) values for yaupon holly (Ilex vomitoria, variety ‘Nana’), wild –type yaupon (I. vomitoria), yerba maté (I. paraguariensis), and green tea (Camellia sinensis)...... 46

LIST OF FIGURES

Figure page

1-1 Mean (+1 SE) foliar concentrations of alkaloids and nitrogen by treatment and gender in Ilex vomitoria (note difference in scales)...... 26

1-2 Mean (+1 SE) foliar concentrations of phenolics and antioxidant capacity by treatment and gender in Ilex vomitoria (note difference in scales)...... 27

2-1 Cultivated hedge of yaupon (Ilex vomitoria) variety ‘Nana’ on the University of Florida campus, Gainesville, Florida...... 40

2-2 Mean (+/- SE) A) leaf mass, B) caffeine concentration, and C) total caffeine yield in control and fertilized Ilex vomitoria cultivar ‘Nana’ individuals obtained from the second harvest of this study...... 41

2-3 Mean (+/- SE) A)antioxidant capacity and B) total antioxidant capacity in control and fertilized Ilex vomitoria cultivar ‘Nana’ individuals obtained from the second harvest of this study ...... 42

2-4 Relationship between % full sun and caffeine concentration (% dry weight of foliage) for yaupon holly (Ilex vomitoria) cultivar ‘Nana.’ Line indicates linear regression...... 43

2-5 Relationship between % full sun and antioxidant capacity, as measured by the ORAC method, for yaupon holly (Ilex vomitoria) cultivar ‘Nana.’ Line indicates linear regression...... 44

2-6 Mean (+1 SE) foliar caffeine concentrations in four varieties of Ilex vomitoria. Lower case letters specify groupings as indicated by Tukey’s Tests for multiple comparisons...... 45

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science

ECOLOGY AND ECONOMIC POTENTIAL OF SECONDARY METABOLITES PRODUCED BY YAUPON HOLLY, Ilex vomitoria

Matthew J. Palumbo

December 2006

Chair: Francis E. Putz Major Department: Botany

Leaves of yaupon (Ilex vomitoria, Aquifoliaceae), a dioecious shrub native to the coastal

plain of the southeastern United States, were traditionally brewed into a caffeinated beverage

widely consumed by indigenous peoples and European colonists. Despite its importance during

pre-Colombian and colonial times, yaupon holly is no longer widely exploited for its stimulating chemical properties. Therefore, to foster interest in yaupon, I assessed its commercially harvestable leaf productivity while investigating factors that influence its production of caffeine

and related alkaloids as well as phenolic compounds that partially function as anti-oxidants.

First, I tested predictions of the carbon-nutrient balance (CNB) hypothesis by determining

whether nitrogen availability and plant gender influence production of caffeine and related

alkaloids as well as phenolic compounds in leaves of pot-grown yaupon plants fertilized with

ammonium nitrate. The CNB hypothesis predicts that additional nitrogen should result in increased alkaloid concentrations and decreased phenolic concentrations. An extension of the

CNB hypothesis to dioecious plants predicts that females have higher C/N ratios and therefore

higher phenolic concentrations and lower alkaloid concentrations than male conspecifics. I found that caffeine and total alkaloid concentrations were 5-10 times higher in fertilized than control plants but did not vary by gender. An interaction between gender and fertilization,

however, suggests that fertilization elicits a greater response in caffeine production in females than in males. Total phenolic concentrations were higher in control females than control males as predicted by the CNB hypothesis, but did not vary by treatment; nor were there differences by gender among fertilized plants. In addition, high correlations between antioxidant capacity and both classes of phenolic compounds detected (cinnamic acid derivatives and flavonoids) indicate that in addition to their putative defensive function against herbivores, phenolics protect yaupon from oxidative stress. Results of this study indicate a need to re-appraise the physiological

mechanisms by which resource availability affects secondary metabolism and a need to consider

the selective pressures to which secondary metabolism responds.

Next, I investigated the suitability of one yaupon cultivar (‘Nana’) for commercial

production by determining the effects of nitrogen fertilization on leaf yield, caffeine content, and

antioxidant content. Fertilized plants produced 21.9% more leaf tissue than control plants and

246.9% higher caffeine concentrations, which resulted in 350.9% higher total caffeine yield per

unit area from fertilized plants. Although antioxidant capacity on a concentration basis did not differ by treatment, total antioxidant capacity per plant was higher in fertilized plants due to a

greater leaf mass yield from fertilized plants. In addition, increases in caffeine concentrations

and antioxidant capacity with increasing canopy cover suggests that yaupon should be open- grown. Low caffeine concentrations in ‘Nana’, however, prompted me to assay caffeine

concentrations among different yaupon varieties including ‘Nana’. I found that of the commonly

cultivated yaupon cultivars in North Florida, ‘Pendula’ produces highest and ‘Nana’ the lowest

caffeine concentrations. Wild-type varieties, however, also produced ample caffeine

concentrations and may be a better choice for sources of cultivars as they provide a wider range

of chemical traits, which can be selected according to the desires of consumers.

CHAPTER 1 NITROGEN FERTILIZER AND GENDER EFFECTS ON THE SECONDARY METABOLISM OF YAUPON, A CAFFEINE-CONTAINING NORTH AMERICAN HOLLY

Introduction

Leaves of yaupon (Ilex vomitoria, Aquifoliaceae), a dioecious shrub native to the coastal

plain of the southeastern United States, were traditionally brewed into a caffeinated beverage

widely consumed by indigenous peoples and European colonists (Hudson 1979). Although

yaupon was the principal ingredient of a beverage imbibed by Timucuan and Creek Indians

during a ceremony that included ritual vomiting (Merrill 1979), qualitative chemical analyses

detected no emetic properties (Fuller et al. 2002). Partially because of its repellant and

misleading specific epithet, yaupon’s virtue as a caffeine source is now mostly ignored, while

markets for its close South American relative, Ilex paraguariensis (yerba maté) grow rapidly

(Graham 1998). To revive interest in yaupon, we investigated whether nitrogen fertilization increases foliar concentrations of caffeine and other methylxanthine alkaloids. In addition, we

determined the effects of nitrogen fertilization on concentrations of phenolic compounds, which

confer antioxidant properties on yerba maté (Carini et al. 1998). Finally, we used this dioecious

and easily cultivated alkaloid and phenolic-producing species to test a well-known theory that

addresses nutrient allocation to plant secondary metabolites.

The carbon/nutrient balance (CNB) hypothesis (Bryant et al. 1983; Bryant et al. 1988;

Tuomi et al. 1988; Reichardt et al. 1991; Herms and Mattson 1992) is used to explain the effects of nutrient availability on the concentration of plant secondary metabolites (e.g., alkaloids and phenolics). The CNB hypothesis asserts that plants allocate carbon and nitrogen to secondary metabolism only after growth requirements are met and that growth is constrained more by nutrient limitations than by photosynthesis. According to this theory, the excess carbohydrates that accumulate in nutrient-limited plants when photosynthesis outpaces growth are diverted to

the production of carbon-based secondary compounds (e.g., phenolics). Removal of nutrient

limitation by fertilization increases tissue nutrient concentrations and allows growth to outpace

photosynthesis. Therefore, according to the CNB hypothesis, production of nitrogen-based secondary metabolites (e.g., alkaloids) should increase as nitrogen is acquired in excess of growth requirements. Conversely, concentrations of carbon-based phenolic compounds are predicted to decrease with fertilization due to decreased rates of photosynthesis relative to growth.

Although some studies on the effects of nitrogen fertilization on secondary metabolites support the CNB hypothesis, others contradict it (Herms and Mattson 1992; Koricheva et al.

1998). Critics of the CNB hypothesis explain that its failures can be attributed to its assumption that plants respond to varying levels of resource availability in a physiologically passive manner driven by simple mass-action (Hamilton et al. 2001; Nitao et al. 2002). These critics contend that although phenotypic plasticity in secondary metabolite production is influenced by resource availability, it is also subject to selective pressures and is therefore genetically regulated and possibly adaptive. For example, plasticity in nicotine production maximizes seed production in

Nicotiana attenuata by allowing individuals to regulate resource allocation in response to herbivore pressure (Baldwin 1998; Baldwin 1999).

Although variation in production of caffeine and phenolics has apparently not yet been shown to be adaptive, there is evidence that their production is genetically regulated. For example, high narrow-sense heritability for caffeine concentration (h2 = 0.80) in seeds of robusta

coffee (Caffea canephora) indicates that most phenotypic variation in its caffeine output is

attributed to genetic and not environmental factors (Montagnon et al. 1998). In other species,

phenylalanine ammonia-lyase (PAL) activity, which catalyzes the first step of phenolic

biosynthesis, is finely regulated both by a multi-gene family and by negative feedback inhibition

(Dong et al. 1991; Kervinen et al. 1998). This fine genetic regulation contrasts with the CNB hypothesis that presumes mass flow, as directed by environmental factors, exerts the strongest

influence on the production of secondary metabolites (Reichardt et al. 1991).

The CNB hypothesis was extended by Herms and Mattson (1992) to predict gender

differences in secondary metabolism. The basis for these predictions is the assumption that due

to the nutrient requirements of fruit and seed maturation, female plants allocate higher

proportions of nutrients to reproductive structures than males. If this assumption is correct,

female plants should have higher C/N ratios in vegetative tissue and higher concentrations of

carbon-based secondary compounds than males. The logic of this argument notwithstanding,

studies examining differences in nitrogen concentration and carbon-based secondary compounds

by gender have yielded mixed results (Ågren et al. 1999; Doormann and Skarpe 2002; Bañuelos

et al. 2004) indicating that the mechanisms underlying gender differences in secondary

metabolism are not yet fully understood.

Given the inconsistent evidence for the CNB hypothesis, I sought to contribute to the

debate by determining how nitrogen fertilization and gender influence the production of

methylxanthine alkaloids and phenolics in yaupon holly. In addition, I measured total foliar

nitrogen concentrations and antioxidant capacities to further explore the effects of gender on

responses to increased nitrogen availability. I hope that while exploring the mechanisms that

govern resource allocation to secondary metabolism in yaupon, I can stimulate interest in this

long-used but recently forgotten native plant.

Materials and Methods

Plant Material and Nitrogen Fertilization

Yaupon holly (Ilex vomitoria) grows from North Carolina south to central Florida and west to east Texas in coastal scrub and in the understories of coastal and inland hardwood forests and pinelands. For this pot study, a rooted leaf-bearing stem was collected randomly from each of 26 male and 26 female Ilex vomitoria genets growing in the understory of an oak savanna near

Gainesville, Florida (29°39´N, 82°19´W). Genets were defined as clusters of stems separated by at least 5 m from the nearest conspecific of the same gender (as determined during the flowering season). Each rooted ramet was planted in a 1 L pot filled with surface soil (0-20 cm) from the collection site. Potted plants were grown outdoors in partial shade on the University of Florida campus. Percentage full sun for both the source area ( x = .23, s = .16, N = 10) and pot study site

( x = .29, s = .16, N = 4) were similar as estimated by Gap Light Analyzer (GAP) software

Ecosystem Studies, New York) from photographic images taken with a 180° hemispheric lens (t

= .62, P = .55). Although light availability can influence secondary metabolite production

(Herms and Mattson 1992, Koricheva et al. 1998), this experiment only allowed assessment of nitrogen fertilization and gender effects.

After 8 weeks and continuing for the next 12 weeks, half of the plants of each gender were fertilized with ammonium nitrate (NH4NO3) at a rate of 250 mg N per week. Four plants died during cultivation resulting in a final sample size of 13 control males, 10 fertilized males, 12 control females, and 13 fertilized females. All leaves produced after the commencement of the fertilization treatment were harvested for chemical analysis.

Chemical Analyses

Collected leaves were dried at 105ºC to a constant weight and total nitrogen content was determined using the Kjeldahl method (Bradstreet 1965). Dried leaf tissue for determination of

methylaxanthine alkaloid and phenolic concentrations was extracted in a 100% methanol for 1

min and filtered through Whatman #4 filter paper. Isolates were then diluted with an equal

volume of water, centrifuged for 5 min at ca. 2000 rpm, and filtered through a 0.45 mm filter for

HPLC analysis. Separation was conducted with a Waters 2695 Alliance HPLC system using a

Supelcosil LC-18 column (250 X 4.6 mm) and detected at 280 nm with a Waters 996 PDA

detector scanned from 200-400 nm.

Following the protocol of Talcott et al. (2000), a gradient mobile phase was run consisting

of a 2% aqueous solution of acetic acid in phase A and a 30% acetonitrile plus 2% acetic acid solution in phase B. The gradient ran phase B from 0-30% for 20 min, 30-50% for 10 min, 50-

70% for 20 min, and 70-100% for 5 min at 0.8 mL/min. Phase B ran for an additional 15 min to

elute remaining non-polar compounds followed by equilibration of the column with 100% phase

A prior to injection of the next sample. Compounds were identified by UV/VIS spectral

interpretation and compared to authentic standards and retention times when available. Alkaloid

and phenolic concentrations were calculated on percent dry weight basis.

Isolates extracted in methanol were also used to quantify total reducing compounds by the

Folin-Ciocalteu method (Swain and Hillis 1959), which provides an estimate of total soluble

phenolics. Data were expressed in chlorogenic acid equivalents. The same isolates were used to

determine antioxidant capacity by the oxygen radical absorbance capacity (ORAC) method as

modified by Ou et al. (2001). Fractions were diluted 400-fold in pH 7.2 phosphate buffer prior

to pipetting into a 96 well microplate. Peroxyl radicals were generated by 2, 2’-Azobis (2-

amidinopropane) dihydrochloride (AAPH) with fluorescein as the fluorescent probe.

Fluorescence loss was measured on a Molecular Devices fmax Microplate Reader (485 nm

excitation, 538 nm emission, 37°C) every 2 min for 70 min. Areas under decay curves generated

by these measurements were compared to areas for blank preparations and a standard curve of

Trolox, a water-soluble analogue of tocopherol (vitamin E) and expressed as µΜ Trolox

equivalents/g.

Statistical Analyses

Differences in chemical concentrations between treatments were analyzed with a 2-way full factorial ANOVA followed by Tukey’s HSD tests for post-hoc pair-wise contrasts.

Correlations were determined by Pearson product-moment correlation.

Results

Methylxanthine Alkaloids

Concentrations of caffeine, theobromine, and total methylxanthine alkaloids (Figure1-1) were substantially higher in leaves of fertilized than control plants (Table 1-1). Although concentrations of these compounds did not vary by gender, there was a significant interaction between fertilization and gender in the caffeine response (Table 1-1). This interaction may be

explained by a greater response to fertilization in females (10.4-fold increase) than males (5.6-

fold increase). In contrast to the caffeine effect, there were no significant interactions in either

the theobromine or total methylxanthine alkaloid responses.

Phenolics

Concentrations of total soluble phenolics in yaupon foliage did not differ by treatment or

by gender but were influenced by an interaction between the two effects (Table 1-2). Post-hoc

analysis (Figure 1-2) indicates that the interaction is at least partially attributed to slightly higher

total phenolic concentrations in control females than control males (P < .05).

HPLC analysis of yaupon leaf issue detected two classes of phenolic compounds, cinnamic

acid derivatives and flavonoids. Based on retention times, maximum wavelength of absorbance,

and UV spectroscopic analysis by PDA, the cinnamic acid derivatives were composed of

chlorogenic acid, chlorogenic acid isomers, and chlorogenic acid di-isomers (dicaffeoyl-quinic

acids). Based on the above indicators as well as a PDA-library of flavonoid samples, the

flavonoids are most likely rutin and luteolin.

Concentrations of cinnamic acid derivatives, total flavonoids, and total phenolics as

detected by HPLC did not vary by gender or in response to fertilization (Table 1-2) but these

treatments interacted to affect the production of cinnamic acids and total phenolics detected by

HPLC (Table 1-2). As found with total soluble phenolics, post-hoc pair-wise tests suggest that

these interactions are attributable to greater concentrations of both cinnamic acid derivatives and

total HPLC phenolics in control females than control males (P < .05). Fertilized samples, in

contrast, differed neither by gender nor from control plants in concentrations of cinnamic acid derivatives or total HPLC phenolics and there was no gender-by-fertilization interaction in the flavonoid response.

Nitrogen Content

Although foliar nitrogen concentrations (Figure 1-1) were much higher in fertilized than control plants (F = 233.2, P < .0001), nitrogen concentration did not vary by gender nor were there significant interactions. In all plants, nitrogen concentrations were closely correlated with concentrations of caffeine (r = .98, P < .0001), theobromine (r = .67, P < .0001), and total methylxanthine alkaloids (r = .98, P < .0001). Furthermore, a far greater proportion of nitrogen was invested in alkaloid production by fertilized plants. For example, female and male control plants invested 7.4 and 12.8 % of their total nitrogen in the production of methylxanthine

alkaloids, respectively, whereas in fertilized females and males, the respective proportions were

29.8 and 28.3%.

Antioxidant Capacity

Measures of oxygen radical absorption capacity (ORAC) were extremely high in all plants

and did not vary by treatment or gender (Figure 1-2). ORAC values were highly correlated with

total phenolics (r = .94, P < .0001), total cinnamic acid derivatives (r = .93, P < 0.0001), total

flavonoids (r = .58, P < .0001), and total phenolics as measured by HPLC (r = .93, P < .0001).

Discussion

The results of this study both support and contradict predictions of the CNB hypothesis.

Yaupon responded to nitrogen fertilization with large increases in concentrations of caffeine and

total methylxanthine alkaloids but not with decreases in concentrations of cinnamic acid

derivatives, flavonoids, or total phenolics. Furthermore, neither alkaloid nor total nitrogen

concentrations differed by gender, although an interaction between gender and fertilization

influenced caffeine production. As predicted by the CNB hypothesis, cinnamic acid derivative

and total phenolic concentrations were higher in females than males among the control group but

did not differ by gender among fertilized plants nor did flavonoid concentrations differ by gender

in either control or fertilized groups.

The lack of phenolic responses by yaupon to nitrogen addition in yaupon indicates the

need for alternatives to the mechanism set forth by the CNB hypothesis for interpretation of

these results. Although the CNB hypothesis predicts that carbon allocation to carbon-based secondary metabolism decreases with nitrogen fertilization due to faster rates of growth relative to photosynthesis (Bryant et al. 1983), nitrogen fertilization does not necessarily favor growth over photosynthesis (Kannan and Paliwal 1997). Furthermore, even when nitrogen additions stimulate biomass accumulation but not photosynthetic rates, phenolic concentrations do not

always decrease (Reichardt et al. 1991; Bezemer et al. 2000). Similarly, Donaldson et al. (2005) found no correlation between relative growth rates and the relative foliar mass of phenolic glycosides among nitrogen-fertilized Populus tremuloides.

The CNB hypothesis places an importance on relative increases of growth over photosynthesis in response to nitrogen fertilization because it posits that secondary metabolite production is driven by carbon and nitrogen overflow beyond allocation to growth (Bryant et al.

1983; Reichardt et al. 1991). By this mechanism, relatively lower photosynthetic rates with additional nitrogen constrain carbohydrate availability for carbon-based secondary metabolism while nitrogen supplied in excess of growth requirements is allocated to nitrogen-based secondary metabolism. Studies have revealed, however, that nitrogen fertilization does not necessarily decrease starch or total sugar concentrations even when it decreases phenolic concentrations (Balsberg Påhlsson 1992) and that nitrogen additions may even stimulate increases in concentrations of total nonstructural carbohydrates (Kaakeh et al. 1992).

Closer examination of the influence that carbohydrate metabolism exerts on all secondary metabolic pathways may resolve patterns of resource allocation to secondary metabolite production. In particular, biosynthesis of nitrogen-based secondary compounds is dependent on carbon skeletons generated by photosynthesis (Dennis and Blakeley 2000) and therefore, like the biosynthesis of carbon-based secondary compounds, should also be limited by carbon availability. For example, biosynthesis of phenylalanine, the amino acid precursor of phenolic compounds, is dependent on organic acids derived from carbohydrate pools generated by glycolysis and the pentose phosphate pathway (Weaver and Herrmann 1997). Similarly, biosynthesis of purine, the nucleotide precursor of caffeine and related methylaxanthine alkaloids

(Suzuki et al. 1992), also begins with an organic acid derived from the pentose phosphate

pathway and derives nitrogen from amino acids with carbon skeletons also supplied by organic

acids (van der Graaff et al. 2004). Carbon partitioning obviously warrants further investigation

in yaupon and other plant species that employ numerous secondary metabolic pathways.

A refined understanding of the mechanisms by which nitrogen availability influences

carbohydrate, nitrogen, and secondary metabolism can clarify its influence on carbon and nitrogen allocation for the production of both alkaloids and phenolics. For example, recent

studies have shown that nitrate, and not merely downstream nitrogen metabolites, signals the

induction of organic acid, amino acid, and nucleotide metabolism but represses starch

biosynthesis (Scheible et al. 1997; Scheible et al. 2004). By a similar mechanism, nitrate also induces nicotine but inhibits phenylpropanoid and flavonoid biosynthesis in Nicotiana tabacum

(Fritz et al. 2006). These studies indicate that carbohydrate, nitrogen, and secondary metabolism are, in fact, coordinated according to nitrogen availability. C/N ratios alone, however, may be insufficient for predicting resource allocation to secondary metabolites because their production is influenced by nitrate-induced signals and not merely by the mass flow of carbon and nitrogen through various metabolic pathways (Stitt and Krapp 1999).

Root-shoot ratios are also known to reflect C/N ratios (e.g., Ågren and Ingstead 1987) and therefore may shape the effect of nitrogen additions on carbohydrate, nitrogen, and secondary metabolic responses. For example, studies that showed a decrease in root/shoot ratios with additional nitrogen also showed concomitant increases in soluble sugar and amino acid concentrations but decreases in starch concentrations (Green et al. 1994) as well as decreased concentrations of condensed tannins but no changes in benzoic acid derivatives (Donaldson et al.

2006). Because our study on yaupon involved the use of rooted vegetative offshoots with

presumably low root/shoot ratios (e.g., Ritchie et al. 1992), the potential influence of root/shoot

ratios on secondary metabolic output should be considered.

Determining the influence of selective agents may also inform our understanding of the

physiological mechanisms that drive variation in secondary metabolic output (Berenbaum 1995;

Hamilton et al. 2001; Nitao et al. 2002). For example, given that many plant secondary

metabolites are putative defenses against pathogens and herbivores, predictions about allocation

patterns for their production should reflect the selective pressures to which they are a response

(Berenbaum 1995). Induction of caffeine and/or theobromine synthesis upon attack by fungal

pathogens in tea (Camellia sinensis; Kumar et al. 1995; Punyasiri et al. 2005) and cacao

(Theobroma cacao; Aneja and Gianfagna 2001), for example, provides evidence that selective

agents may shape the production patterns of these alkaloids. Although plant nutrient

concentrations have long been correlated with herbivore preferences (e.g., Mattson 1980), the

effect of nitrogen availability on chemical responses to herbivores and pathogens has not been

extensively explored. Lou and Baldwin (2004) found, however, that plants with low nitrogen

availability maintain higher constitutive levels of nicotine than plants with high nitrogen

availability but that only plants of the latter group increased nicotine production in response to application with larval oral secretions in Nicotiana attenuata.

Secondary metabolites may serve adaptive functions in addition to defense, which may also help explain their responses to additional nitrogen (Herms and Mattson 1992; Nitao et al;

2002). For example, our results indicating that fertilized yaupon plants allocated a higher proportion of nitrogen to methylxanthine alkaloid production than control plants suggest that alkaloid biosynthesis is not only actively regulated in yaupon holly but that caffeine may also serve multiple functions. Caffeine concentrations in fertilized female and male plants (3.9% and

3.4%, respectively) far exceed those necessary to deter or even kill herbivores (Nathanson 1984;

Slansky and Wheeler 1992, Hollingsworth et al. 2003). Their high allocation of nitrogen to

caffeine after nitrogen fertilization suggests that caffeine also serves as a storage metabolite for

nitrogen accumulated in excess of immediate plant needs (Chapin et al. 1990). The suitability of

caffeine for nitrogen storage, however, requires verification by studies addressing both the metabolic and ecological costs of storing caffeine.

The multiple functions of phenolic compounds may render it difficult to predict their production in response to nitrogen additions as well. For example, chlorogenic acid and flavonoids are known to inhibit herbivores (e.g., Ikonen et al. 2001; Ikonen et al. 2002:

Onyilagha et al. 2004) but both also protect plants from light stress by scavenging free radicals and absorbing potentially harmful UV radiation (Grace and Logan 2000; Harborne and Williams

2000). Correlations between antioxidant capacity and all measures of phenolic concentrations indicate that both cinnamic acid derivatives and flavonoids provide yaupon with protection from oxidative stress. That a particular secondary metabolite serves numerous functions supports the notion that its production is under genetic control and most likely regulated by numerous stress factors (Nitao et al. 2002), making it difficult to ascertain the mechanisms driving its phenotypic variation without first isolating the effects of each factor.

Our results also largely contradict the CNB hypothesis prediction that female plants have higher foliar C/N ratios, lower levels of nitrogen-based defensive compounds, and higher levels of carbon-based defenses than male conspecifics (Herms and Mattson 1992; Ågren et al 1999).

In particular, concentrations of neither alkaloids nor total nitrogen differed by gender in yaupon.

The observed interactive effect between gender and fertilization on caffeine production, however, suggests that female plants respond more to nitrogen fertilization than males, and

therefore that resource allocation to secondary metabolism may differ by gender. Furthermore, higher concentrations of total phenolics and total cinnamic acid derivatives in control females than control males also suggest gender differences in resource allocation to secondary metabolism. Measurements of other traits by gender such as growth rates, herbivory, and investment in structural defenses (Jing and Coley 1990; Ågren et al 1999) may resolve the mechanism and selective forces that shape variation in secondary metabolic output. If gender- based phenotypic differences in life history traits are observed, however, they should be appraised in terms of their effects on reproductive success or some other measure of their adaptive significance (Berenbaum 1995).

The dramatic increases in caffeine and theobromine concentrations in response to increased nitrogen availability observed in yaupon may relate to its success across a broad environmental range (Sultan 2000). For example, yaupon occurs both in coastal scrub communities, where it is adapted to high light, salt spray, and drought (Johnson and Barbour

1990), as well as in the understories of mesic hardwood forests, which are typically moister and contain higher levels of soil organic matter (Platt and Schwartz 1990). Because foliar nitrogen concentrations affect plant susceptibility to herbivory (Mattson 1980) and because the native habitats of yaupon vary in soil nutrient availability, an ability to mediate between herbivore pressure and defensive metabolite production in response to nitrogen availability may indicate a plastic response that is possibly adaptive. Such an adaptive strategy may be suggested by our finding that 2.4- and 2.3-fold increases in foliar nitrogen concentration resulted in 10.4- and 5.6- fold increases in the caffeine concentrations of female and male plants, respectively. Future studies that include the identification of yaupon’s herbivores and pathogens may reveal the significance of these differences in allocation to caffeine in relative proportion to total nitrogen

concentration. Common garden and reciprocal transplant experiments could also help to clarify

the role that phenotypic plasticity in secondary metabolic output plays in determining the success

of yaupon across its large geographical and ecological range.

Conclusions

Because yaupon holly (Ilex vomitoria) is dioecious, easily cultivated and produces large amounts of both nitrogen-based and carbon-based secondary compounds, it is an excellent species on which to test theories about resource allocation to secondary metabolism. For example, although an increase in methylxanthine alkaloid concentrations in response to nitrogen

fertilization is predicted by the CNB hypothesis, the lack of phenolic responses indicates that a

re-evaluation of the mechanism by which nitrogen availability influences secondary metabolism is necessary. In addition, the lack of differences in concentrations of both total foliar nitrogen and methylxanthine alkaloids between genders contradicts an extension of the CNB hypothesis to dioecious species. An interactive effect between gender and fertilization on caffeine production as well as higher phenolic concentrations in female conspecifics, however, indicates that resource allocation to secondary metabolite production may differ by gender in yaupon. The precise mechanisms underlying these gender differences, however, warrant further investigation.

Finally, ample production of both caffeine and antioxidant-rich phenolics indicate the potential for renewed commercial use of yaupon holly.

Table 1-1. ANOVA results for foliar concentrations of methylxanthine (MX) alkaloids in Ilex vomitoria. Total Source df Caffeine Theobromine MX Alkaloids

Treatment 1 329.04*** 88.29*** 348.89*** Gender 1 0.47 0.79 0.25 Treatment*Gender 1 4.26* 0.18 3.55† Error 44

F values presented for main effects and their interaction. Levels of significance indicated as follows: *P < 0.05, ***P < 0.0001, †P < 0.1.

Table 1-2. ANOVA results for foliar concentrations of phenolic compounds in Ilex vomitoria. Total Total Total Total Source df Phenolics Cinnamic Flavonoids Phenolics Derivatives by HPLC Treatment 1 1.83 1.67 0.58 1.65 Gender 1 0.19 0.13 3.48† 0.22 Treatment*Gender 1 6.21* 6.10* 2.84† 6.12* Error 44

F values presented for main effects and their interaction. Levels of significance indicated as follows: *P < 0.05, †P < 0.1.

Figure 1-1. Mean (+1 SE) foliar concentrations of alkaloids and nitrogen by treatment and gender in Ilex vomitoria (note difference in scales). A) Caffeine. B) Theobromine. C) Total methylxanthine alkaloids. D) Total nitrogen. Black bars represent female plants; grey bars represent male plants. Lower case letters specify groupings as indicated by Tukey’s Tests for multiple comparisons when interactions were indicated (P < .05).

Figure 1-2. Mean (+1 SE) foliar concentrations of phenolics and antioxidant capacity by treatment and gender in Ilex vomitoria (note difference in scales). A) Total soluble phenolics. B) Cinnamic acid derivatives. C) Flavonoids D) Total HPLC phenolics. E) Antioxidant capacity. Black bars represent female plants; grey bars represent male plants. Lower case letters specify groupings as indicated by Tukey’s Test for multiple comparisons when interactions were indicated (P < .05).

CHAPTER 2 Ilex vomitoria: AN OVERLOOKED NORTH AMERICAN CAFFEINE SOURCE

Introduction

Caffeine and Conservation

Caffeine, the world’s most highly consumed psychoactive substance (Weinberg and Bealer

2001), affects numerous socio-economic issues at a global scale. For example, when production

of sun-grown coffee in Southeast Asia increased in the 1990s while world consumption rates

remained constant, its market value crashed (Vandemeer 2003). This ‘global coffee crisis’

triggered a positive feedback whereby increased areas of forest were cleared for coffee

production to compensate for financial losses, which in turn threatened the conservation of

tropical Asian megafauna including Sumatran tigers, elephants, and rhinoceroses (O’Brien and

Kinnaird 2003). Although solutions to this crisis like improving yields of sun-grown coffee

while discouraging its cultivation in protected wildlife areas have been offered and debated

(O’Brian and Kinnaird 2004, Dietsch et al. 2004), we suggest that local sources of caffeine be

exploited to alleviate pressure on forest conversion in areas of high conservation value. In North

America, for example, consumers might consider replacing coffee with yaupon holly for their

stimulating morning beverage. Yaupon (Ilex vomitoria), a caffeine-containing shrub that was

long used in its native range of the southeastern United States but recently forgotten, is the focus

of the following study, which examines several factors associated with its production as a

caffeine crop.

History of Yaupon Tea

Consumption of a tea made from yaupon holly began with the indigenous peoples of its

native range who were presumably the first to discover its stimulating and healthful properties

(Hudson 1979). Amerindians brewed leaves and twigs of yaupon holly into a caffeine-

containing beverage whose consumption figured prominently in their social and cosmologic

constructs (Hudson 1995). For example, consumption of tea made from yaupon in Creek society

was reportedly reserved for adult males of high social status and was associated with Yahola, a

sky deity revered for his immaculate nature (Merill 1979, Fairbanks 1979). Rituals that included

fasting, consumption of large quantities of yaupon tea, and vomiting reflect the importance that

these peoples placed upon spiritual purity and cleanliness deemed necessary before assembling for important council meetings or military action. The act of vomiting was more likely a manifestation of a symbolic connection between yaupon and purity than a physiological response to ingestion of yaupon tea, which was consumed in settings that did not include vomiting (Merill

1979) and does not seem to possess any chemical properties that would induce emesis (Fuller et al. 2002).

European settlers of the southeastern United States readily adopted the Amerindian practice of brewing and drinking yaupon tea. Spanish settlers in Florida, for example, learned of

its virtues from the Timucua Indians in the 16th Century (Hudson 1979). Yaupon tea became so

popular among Spanish colonists, who called it te del indio or cacina (Hudson 1979, Hudson

1995), that a Spanish priest reported in 1615 that “there is no Spaniard or Indian who does not drink it every day in the morning or evening” (Sturtevant 1979). By the 18th Century English settlers in South Carolina also began drinking yaupon tea, which they called “black drink” or cassina in South Carolina (Hudson 1979, Hudson 1995) and yaupon in North Carolina (Hale

1891). Europeans presumably found the taste of yaupon tea agreeable as it appeared in 18th

Century markets in both England, where it was called Appalachian tea or Carolina tea, and

France, where it was called Appalachina (Hudson 1979). Despite its initial acceptance in

Europe, broad consumption of yaupon tea ceased both in the United States and abroad by the late

19th Century (Hudson 1979, Hudson 1995, Sturtevant 1979).

The disappearance of yaupon tea from common use has long puzzled scholars especially given that caffeine is the world’s most popular drug and caffeinated beverages are in high demand (Hale 1891, Hudson 1979, Sturtevant 1979). Hale (1891) suggested that diminished consumption of yaupon might be attributed to a lack of cultural exchange between the European settlers and indigenous peoples of the Southeastern United States. This lack of social cohesion contrasts with the mixing of races that occurred in South America, home of yerba maté, a caffeinated beverage prepared from a congener of yaupon, Ilex paraguariensis. Yerba maté has maintained its popularity to the present era. Hudson (1979), in contrast, speculated that discontinued use of yaupon is attributed to interference of its trade by tea and coffee merchants, which led to its association with an underclass that could not otherwise afford the more expensive, aforementioned caffeine-containing products. Although differences in several key socio-economic factors may exist in the consumption of yaupon tea and yerba maté, historical uses of these beverages are otherwise very similar.

European consumption of yerba maté began circa 1540 when Spanish colonists adopted it from the Guaraní Indians and spread from its native range of southern Brazil, Paraguay,

Uruguay, and northeastern Argentina to Andean countries as far north as Ecuador by the 18th

Century (Sturtevant 1979, Jamieson 2001). When coffee and tea became the caffeinated- beverages of choice by South America’s upper classes in the 19th century, however, yerba maté was relegated to status as the drink of the lower classes (Jamieson 2001). Nevertheless, consumption of yerba maté in South America managed to persist alongside other caffeinated beverages in the 20th Century perhaps because it maintained regional identity in the face of

increasingly globalized markets (Jamieson 2001). Annual production of yerba maté in

Argentina, Paraguay, and Uruguay currently exceeds 250,000 Mg (Graham 1998). Furthermore,

yerba maté has recently become a global commodity, as evidenced by a nearly 14-fold increase

in revenue generated from Argentinean exports between 2002 and 2004 (United Nations

Commodity Trade Statistics Database). This global commercialization has been bolstered by the

reportedly high antioxidant content and associated human health benefits of yerba maté (Bastos

et al. 2006) as similarly reputed for green tea prepared from Camellia sinensis (Stewart et al.

2005).

Yaupon Chemistry and Production

Many Americans might be surprised to learn of a native caffeine-producing plant that can

be used to brew a yerba maté-like beverage. A bigger surprise may come when residents of the

Southeastern United States learn that they often encounter this species and may even have it growing it in their backyard. Yaupon is widely cultivated as an ornamental hedge on urban and suburban landscapes within its native range and at least 17 cultivars are currently marketed

(Bijan Dehgan, personal communication). Early endorsements of yaupon cultivation for commercial exploitation of its chemical properties (Hale 1891, Power and Chestnut 1919), however, were mostly ignored.

In my previous study of the chemistry of yaupon (Chapter 1), I found that fertilization of pot-grown plants with nitrogen substantially increases foliar concentrations of caffeine and theobromine. In addition, I found that yaupon foliage possesses a high antioxidant capacity that is tightly correlated with the production of phenolic compounds but that is unaffected by additional nitrogen. That study revealed that appraisal of several other factors critical to the production of yaupon was still necessary. For example, given the availability of numerous cultivars, it seems important to screen them so as to identify those with desired levels of both

caffeine and antioxidants. Similarly, leaf production and thus yields of caffeine and antioxidants from fertilized and unfertilized plants are unknown. Finally, the influence of canopy cover on the production of both alkaloids and antioxidants in yaupon should be considered because one popular plant defense theory, the carbon/nutrient balance (CNB) hypothesis, predicts that exposure to increased light increases production of antioxidant-rich phenolics while decreasing alkaloids like caffeine (Bryant et al. 1983)

Because antioxidants have reputed health benefits, it may be important to the economic value of yaupon that high antioxidant levels be maintained. The health benefits of antioxidants are partially attributed to their ability to prevent the formation, scavenge, or promote the decomposition of free-radicals, which are known to contribute to numerous human health problems such as cancer, cardiovascular disease, diminished immune system function, brain dysfunction, and cataracts (Ames et al. 1993, Young and Woodside 2001). For example, prevention of lipid peroxidation and DNA fragmentation caused by free radical formation, which are known to cause cardiovascular disease and cancer, respectively, has been attributed to antioxidants present in both yerba maté (e.g., Schinella et al. 2000, Bracesco et al. 2003) and green tea (e.g., Terao et al. 1994, Yokozawa et al. 1998).

In the following study, I investigated the suitability of one yaupon cultivar for commercial production of leaves for tea by determining its leaf production rates, foliar caffeine concentrations, and foliar antioxidant capacity with and without nitrogen fertilization. In this cultivar, I also evaluated the influence of light availability on the production of both caffeine and antioxidants. Finally, I compared caffeine concentrations among three cultivars of yaupon and wild-type yaupon in order to determine which yaupon varieties are best suited for commercial production of foliage for brewing tea.

Materials and Methods

Experimental Design

Yaupon shrubs comprising six separate hedges of the ‘Nana’ cultivar (Figure 2-1) were sampled on the University of Florida (UF) campus in Gainesville, Florida (29°39´N, 82°19´W).

Leaves covering an area of 700 cm2 were clipped from the pruned upper surface of two shrubs

per hedge. For 3 months during the growing season, one shrub per hedge was then fertilized

2 monthly with 250 mg ammonium nitrate (NH4NO3) at a rate of 1g fertilizer/m /mo; the other

shrub in the same hedge but separated by >2 m was reserved as an unfertilized control. After 3

months, leaves were clipped from the same area as the original clipping. Leaves from the second

harvest were then dried at 60˚C for 48 hours and weighed after which 0.5 g of dried and ground

leaf material was extracted with water in test tubes that were then placed in boiling water for 10

minutes. Extracts were then analyzed for caffeine concentration and antioxidant capacity using

the methods described in Chapter 1.

To compare caffeine concentrations in the foliage of different yaupon varieties, nine

separate shrubs each from three cultivars of yaupon (‘Nana’, ‘Pendula’, and “Will Fleming’)

were randomly sampled from cultivated hedges on the UF campus. In addition, leaves were

collected from five wild-type yaupon shrubs occurring in natural stands on UF campus and four

wild-type shrubs growing in an oak savanna approximately 4 km away. The youngest, fully

expanded leaves were harvested from each of the four varieties and processed as above.

Chemical Analyses

To determine caffeine concentrations in yaupon foliage, the water extracts were

centrifuged for 5 min and filtered through a 0.45 mm filter for HPLC analysis. Separation was

conducted with a Waters 2695 Alliance HPLC system using a Supelcosil LC-18 column (250 X

4.6 mm) and detected at 280 nm with a Waters 996 PDA detector scanned from 200-400 nm.

Following the protocol of Talcott et al. (2000), a gradient mobile phase was run consisting of a

2% aqueous solution of acetic acid in phase A and a 30% acetonitrile plus 2% acetic acid solution in phase B. The gradient ran phase B from 0-30% for 20 min, 30-50% for 10 min, 50-

70% for 20 min, and 70-100% for 5 min at 0.8 mL/min. Phase B ran for an additional 15 min to

elute remaining non-polar compounds followed by equilibration of the column with 100% phase

A prior to injection of the next sample. Caffeine was identified by UV/VIS spectral

interpretation and compared to a caffeine standard solution. Caffeine concentrations were

calculated on percent dry weight basis.

The isolates used to determine caffeine concentration were also used to measure

antioxidant capacity by the oxygen radical absorbance capacity (ORAC). Fractions were diluted

400-fold in pH 7.2 phosphate buffer prior to pipetting into a 96-well microplate. Peroxyl

radicals were generated by 2, 2’-Azobis (2-amidinopropane) dihydrochloride (AAPH) with fluorescein as the fluorescent probe. Fluorescence loss was measured on a Molecular Devices fmax Microplate Reader (485 nm excitation, 538 nm emission, 37°C) every 2 min for 70 min.

Areas under decay curves generated by these measurements were compared to areas for blank preparations and a standard curve of Trolox, a water-soluble analogue of tocopherol (vitamin E) and expressed as µΜ Trolox equivalents/g.

The water extracts derived from the second study comparing caffeine concentrations among yaupon varieties were centrifuged for 20 minutes and filtered through a 0.45 mm filter for

HPLC analysis. Separation was conducted with a Waters 717 Plus Autosampler HPLC system using a Dionex C-18 column (250 X 4.6 mm) and detected at 280 nm with a Waters Dual λ

Absorbance Detector scanned. Following a modified protocol of Talcott et al. (2000), a gradient mobile phase was run consisting of a 2% aqueous solution of acetic acid in phase A and a 30%

acetonitrile plus 2% acetic acid solution in phase B. The gradient ran phase B from 0-30% for

20 min, 30-50% for 10 min, 50-70% for 3 min, and 70-100% for 2 min at 1.0 mL/min. Phase B

ran for an additional 5 min to elute remaining non-polar compounds followed by equilibration of

the column with 100% phase A prior to injection of the next sample. Compounds were

identified by comparison of spectra and retention time with a caffeine standard solution.

Caffeine concentrations were calculated on percent dry weight basis.

Measurement of Light Availability

Percentage full sun for each of the six yaupon hedges sampled in the nitrogen fertilization

Simon Fraser University, British Colombia and Institute of Ecosystem Studies, New York) from photographic images taken with a 180° hemispheric lens.

Statistical Analysis

Caffeine content and antioxidant capacity for each sample of the nitrogen fertilization study were calculated by multiplying by total leaf mass. Differences in total leaf mass, caffeine

concentration, total caffeine content, antioxidant capacity, and total antioxidant capacity content

were then analyzed by paired-sample t-tests. Effects of canopy cover on caffeine concentrations

and antioxidant capacity were determined by linear regressions. Caffeine concentrations in the

foliage of the three cultivars and wild-type yaupon were compared by one-way ANOVA

followed by Tukey’s HSD test for post-hoc pair-wise contrasts.

Results

Fertilized plants of the ‘Nana’ variety of I. vomitoria yielded only 1.2% more leaf tissue

(t = 2.95, N = 6, P < 0.05) than control plants over the 3 month period monitored but the foliage produced contained 3.5% higher caffeine concentrations (t = 3.16, N = 6, P < 0.05), which resulted in 4.5% higher total caffeine yield from fertilized plants (t = 2.69, N = 6, P < 0.05;

Figure 2-2). Although antioxidant capacity on a concentration basis did not differ by treatment

(t = 1.22. N = 6, P = 0.14), total antioxidant capacity per plant was higher in fertilized plants

(t = 2.07, N = 6, P < 0.05) due to their total leaf mass yield (Figure 2-3).

Measurements of canopy cover revealed a positive influence of light availability on the

production of both caffeine and antioxidants. A marginally significant relationship between

caffeine concentration and percentage full sun was detected [caffeine concentration = -0.11 +

2 0.009 (% full sun); SEb = 0.005; P = 0.1; R = 24.7; Figure 2-4] suggesting a positive influence

of light availability on caffeine production. The effect of light on ORAC was stronger;

antioxidant capacity increased rapidly and consistently with increasing light levels [oxygen

2 radical absorbance capacity = 210.4 + 16.0 (percentage full sun); SEb = 3.2; R = 72.1; P <

0.001; Figure 2-5].

Caffeine concentrations varied among the among yaupon varieties sampled in the second part of this study (F = 3.13, P < 0.05). Post-hoc pair-wise comparisons indicate that foliar caffeine concentrations were highest in the ‘Pendula’ variety, lowest in variety ‘Nana ’ and intermediate in “Will Fleming’ and wild-type yaupon (Figure 2-6).

Discussion

Yields of caffeine and antioxidant capacity in yaupon hedges, especially after nitrogen fertilization, suggest the suitability of yaupon for producing a stimulating and healthful beverage.

The caffeine concentrations in both the control and N-fertilized ‘Nana’ cultivar hedges were disappointingly low (Figure 2-2) compared with other studies on wild-type individuals. For example, in a study that assayed leaves harvested from individuals of yaupon growing in maritime forest, caffeine concentrations ranged from 0.35 to 0.94% (Edwards and Bennett 2005).

Similarly, in the pot study described in Chapter 1, control plants ranged 0 to 1.91% caffeine and one fertilized plant had a caffeine concentration of 5.25% caffeine.

Although measures of antioxidant capacity in ‘Nana’ were moderately high, as measured by the ORAC method, they were lower than those found in my previous study of yaupon

(Chapter 1), as well as in Asian green tea and yerba maté (Chandra and Gonzalez de Mejia 2004;

Table 1). Such comparisons need to be made with care because of the strong positive relationship I found between light availability and antioxidant production. This finding also suggests that if antioxidants are favored by consumers, yaupon should be grown at high light intensities. Although there was only a weak positive relationship between light availability and caffeine production, it still appears that high light conditions may increase caffeine concentrations in yaupon, which is contrary to a prediction of the carbon/nutrient balance (CNB) hypothesis (Bryant et al. 1983).

Comparison of caffeine concentrations among yaupon varieties indicates that ‘Pendula’ may be the best candidate for cultivation as a caffeine crop among common cultivars.

Development of cultivars from wild-type varieties with desired caffeine concentrations, however, may be a better option as indicated by these results and those from studies mentioned earlier

(Chapter1, Edwards and Bennett 2005), which showed a wide range of caffeine concentrations.

The extent to which this variation is genetically based has yet to be determined, but it is clear that yaupon cultivar ‘Nana’ is not the best choice for cultivation if caffeine production is to be favored.

The appropriateness of yaupon as a caffeine source is also indicated by its ample caffeine concentrations, which are comparable to those found in unprocessed leaves of yerba maté whose reported caffeine concentrations vary from 0.65 - 0.85% (Mazzaferra 1994, Reginatto et al.

1999). In future studies, the influence of methods employed in both the processing and

preparation of caffeinated beverages should be addressed. For example, more caffeine is

apparently lost in the processing of yerba maté by the barbaqua method than by the supelco

method due to the higher temperatures to which leaves are exposed in the former (Graham 1998).

The relationship between the caffeine concentration of the caffeinated product and amount of

product used in the preparation of beverage also warrants consideration. For example, although

Asian tea leaves from Camellia sinensis contain ~ 3.5% caffeine and coffee beans derived from

Coffea arabica contain ~1.1% caffeine by weight (Spiller 1984), a 177 cm3 cup of tea contains

25 - 100 mg of caffeine as compared to the 130 – 180 mg of caffeine found in the same volume

of Arabica coffee due to the higher mass of product used in its preparation (Weinberg and Bealer

2002). Finally, the method in which a caffeinated beverage is prepared greatly influences its

caffeine concentration. For example, coffee prepared by the drip method contains more caffeine

than that prepared by percolation, which again contains more caffeine than instant coffee (Gilbert

1986).

Conclusions

Yaupon holly (Ilex vomitoria) provides North Americans with the opportunity to secure a native caffeine source and therefore mitigate the socio-economic and biological conservation problems associated with the consumption of other caffeinated beverages. Yaupon is capable of producing ample concentrations and yields of both caffeine and antioxidants, which can be adjusted to desired levels through alteration of soil nutrient and light regimes. The value of yaupon is increased by its high antioxidant capacity, which has bolstered the value of its South

American congener yerba maté (Ilex paraguariensis), due to the reported health benefits it provides. Although this study indicates yaupon cultivar ‘Pendula’, currently used as an ornamental hedge, produces caffeine concentrations on par with yerba maté, I suggest the use of

wild-type varieties in the development of cultivars for tea production because they offer a wider range of chemical traits that can be selected according to the desires of consumers. Finally, I hope that this study inspires people living in the range of yaupon to forage for its leaves and to prepare a stimulating and healthful drink that will link them to the history of the region.

Figure 2-1. Cultivated hedge of yaupon (Ilex vomitoria) variety ‘Nana’ on the University of Florida campus, Gainesville, Florida.

Figure 2-2. Mean (+/- SE) A) leaf mass, B) caffeine concentration, and C) total caffeine yield in control and fertilized Ilex vomitoria cultivar ‘Nana’ individuals obtained from the second harvest of this study (analysis based on paired samples, N=6).

Figure 2-3. Mean (+/- SE) A)antioxidant capacity and B) total antioxidant capacity in control and fertilized Ilex vomitoria cultivar ‘Nana’ individuals obtained from the second harvest of this study (analysis based on paired samples, N=6).

caffeine conc. = -0.11 + 0.009 (% full sun) SEb = 0.005 R2 = 0.25 P = 0.1

Figure 2-4. Relationship between % full sun and caffeine concentration (% dry weight of foliage) for yaupon holly (Ilex vomitoria) cultivar ‘Nana.’ Line indicates linear regression. Dots represent control individuals, triangles represent fertilized individuals.

ORAC = 210.4 + 16 (% full sun) SEb =3.2 R2 = 0.72 P < 0.001

Figure 2-5. Relationship between % full sun and antioxidant capacity, as measured by the ORAC method, for yaupon holly (Ilex vomitoria) cultivar ‘Nana.’ Line indicates linear regression. Dots represent control individuals, triangles represent fertilized individuals.

b ab

Figure 2-6. Mean (+1 SE) foliar caffeine concentrations in four varieties of Ilex vomitoria. Lower case letters specify groupings as indicated by Tukey’s Tests for multiple comparisons.

Table 2-1. Mean ± 1 SE ORAC (oxygen radical absorbance capacity) values for yaupon holly (Ilex vomitoria, variety ‘Nana’), wild –type yaupon (I. vomitoria), yerba maté (I. paraguariensis), and green tea (Camellia sinensis). N = 6 for each yaupon holly variety ‘Nana’, N = 24 for each wild-type yaupon, and N = 3 for both yerba maté and green tea.

Species ORAC Value (µmol TE/g)

Yaupon (‘Nana’) Unfertilized 608 ± 61.5 Yaupon (‘Nana’) N-Fertilized 670 ± 87.8 Yaupon (wild-type) Unfertilized 987 ± 83.2a Yaupon (wild-type) N-Fertilized 1034 ± 78.3 a Yerba Maté 1239 ± 55.7b Green Tea 1346 ± 60.0b a values from Chapter 1. b values from Chandra and Gonzalez de Mejia (2004).

LIST OF REFERENCES

Ågren GI, Ingestad T (1987) Root:shoot ratio as a balance between nitrogen productivity and photosynthesis. Plant Cell Environ 10:579-586

Ågren J, Danell K. Elmqvist T, Ericson L. Hjältèn J (1999) Sexual dimorphism and biotic interactions. In: Geber MA, Dawson TE, Delph LF (eds) Gender and Sexual Dimorphism in Flowering Plants. Springer-Verlag, Berlin, pp 217-246

Ames BN, Shigenaga MK, Hagen TM (1993) Oxidants, antioxidants, and the degenerative diseases of aging. Proc Natl Acad Sci 90:7915-7922

Aneja M, Gianfagna T (2001) Induction and accumulation of caffeine in young, actively growing leaves of cocoa (Theobroma cacao L.) by wounding or infection with Crinipellis perniciosa. Physiol Mol Plant P 59:13-16

Baldwin IT (1998) Jasmonate-induced responses are costly but benefit plants under attack in native populations. Proc Natl Acad of Sci USA 95:8113-8118

Baldwin IT (1999) Inducible nicotine production in native Nicotiana as an example of adaptive phenotypic plasticity. J Chem Ecol 25:3- 29

Balsberg Påhlsson A (1992) Influence of nitrogen fertilization on minerals, carbohydrates, amino acids and phenolic compounds in beech (Fagus sylvatica L.) leaves. Tree Physiol 10:93- 100

Bañuelos MJ, Sierra M, Obeso JR (2004) Sex, secondary compounds and asymmetry. Effects on plant-herbivore interaction in a dioecious shrub. Acta Oecol 25:151-157

Bastos DHM, Ishimoto EY, Marques MOM, Ferri AF, Torres EAFS (2006) Essential oil and antioxidant activity of green mate and mate tea (Ilex paraguariensis) infusions. J Food Comp Anal 19:538-543

Berenbaum M (1995) The chemistry of defense: theory and practice. Proc Nat Acad Sci USA 92: 2-8

Bezemer TM, Jones TH, Newington JE (2000) Effects of carbon dioxide and nitrogen fertilization on phenolic content in Poa annua L. Biochem Syst Ecol 28:839-846

Bracesco N, Dell M, Rocha A, Behtash S, Menini T, Gugliucci A, Nunes E (2003) Antioxidant activity of a botanical extract preparation of Ilex paraguariensis: Prevention of DNA double-strand breaks in Saccharomyces cerevisiae and human low-density lipoprotein oxidation. J Alt Comp Med 9:379-387

Bradstreet RB (1965) The Kjeldahl Method for Organic Nitrogen. Academic Press, New York, NY

Bryant JP, Chapin FS, Klein DR (1983) Carbon/nutrient balance of boreal plants in relation to vertebrate herbivory. Oikos 40:357-368

Bryant JP, Tuomi J, Niemala P (1988) Environmental constraint of constitutive and long-term inducible defenses in woody plants. In: Spencer KC (ed) Chemical Mediation of Coevolution. Academic Press, San Diego, California. pp 367-389

Carini M, Maffei Facino R, Aldini G, Calloni M, Colombo L (1998) Characterization of phenolic antioxidants from mate (Ilex paraguayensis) by liquid chromatography/mass spectrometry and liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 12:1813-1819

Chandra S, Gonzalez de Mejia E (2004) Polyphenolic compounds, antioxidant capacity, quinone reductase activity of an aqueous extract of Ardisia compressa in comparison to mate (Ilex paraguariensis) and green (Camellia sinensis) teas. J Agr Food Chem 52:3583-3589

Chapin, FS, Schulze E, Mooney HA (1990) The ecology and economics of storage in plants. Annu Rev Ecol Syst 21:423-447

Dennis DT, Blakeley SD (2000) Carbohydrate metabolism. In: Buchanan BB, Gruissem W, Jones RL (eds.) Biochemistry and Molecular Biology of Plants. American Society of Plant Physiologists, Rockville, MD, pp. 630-675

Dietsch TV, Philpott SM, Rice RA, Greenberg R, Bichier P (2004) Conservation policy in coffee landscapes. Science 303:625

Donaldson JR, Krueger EL, Lindroth RL (2006) Competition- and resource-mediated tradeoffs between growth and defensive chemistry in trembling aspen (Populus tremuloides). New Phytol 169:561-570

Dong X, Mindros M, Davis KR, Ausubel FM (1991) Induction of Arabidopsis defense genes by virulent and avirulent Pseudomonas syringe strains and cloned avirulent gene. Plant Cell 3:61-72

Doormann CF, Skarpe C (2002) Flowering, growth and defence in two sexes: consequence of herbivore exclusion for Salix polaris. Functional Ecol 16:649-656

Edwards Al, Bennett BC (2005) Diversity of methylxanthine content in Ilex cassine L. and Ilex vomitoria Ait: Assessing sources of the North American stimulant cassina. Econ Bot 59:275-285.

Fairbanks, CH (1979) The function of black drink among the creeks. In: Hudson CM (ed) Black Drink: A Native American Tea. The University of Georgia Press, Athens, GA. pp 120-149

Fritz C, Palacios-Rojas N, Feil R, Stitt M (2006) Regulation of secondary metabolism by the carbon-nitrogen status in tobacco: nitrate inhibits large sectors of phenylpropanoid metabolism. Plant J 46:533-548

Fuller HM, Arnold RI, Murphy MK (2002) Using chemistry to understand culture: Why did Native Americans vomit after drinking yaupon tea. J Undergrad Chem 1:47-51

Gilbert RM (1986) Caffeine: The most popular stimulant. Chelsea House Publishers, New York, NY

Grace SC, Logan BA (2000) Energy dissipation and radical scavenging by the plant phenylpropanoid pathway. Philos T Royal Soc B 355:1499-1510

Graham HN (1998). Maté. In: G.A. Spiller ed. Caffeine. CRC Press, Boca Raton, FL pp 193-197

Green TH, Mitchell RB, Gjerstad DH (1994) Effects of nitrogen on the response of loblolly pine to drought. New Phytol 128:145-152

Hale EM (1891) Ilex cassine: The aboriginal North American tea. USDA Division of Botany 14: 1-22

Hamilton JG, Zangerl AR, DeLucia EH, Berenbaum MR (2001) The carbon-nutrient balance: its rise and fall. Ecol Lett 4:86-95

Harborne JB, Williams CA (2000) Advances in flavonoid research since 1992. Phytochemistry 55:481-504

Herms DA, Mattson WJ (1982) The dilemma of plants: to grow or defend. Q Rev Bio 67:283- 335

Hollingsworth RG, Armstrong JW, Campbell E. (2003) Caffeine as a novel toxicant for slugs. Ann Appl Biol 142:91-97

Hudson C (1979) Introduction. In: Hudson CM (ed) Black Drink: A Native American Tea. The University of Georgia Press, Athens, GA. pp 1-9

Hudson, C (1995) “Black drink” and the Indians of the southeastern United States. In: Affolter J and Henry M (eds), People and plants: Cultural and historical connections. The State Botanical Gardens of Georgia, Athens, GA. pp 15-21

Ikonen A, Tahvanainen J, Roininen H (2001) Chlorogenic acid as an antiherbivore defence of willows against leaf beetles. Entomol Exp Appl 99:47-54

Ikonen A, Tahvanainen J, Roininen H (2002) Phenolic secondary compounds as determinants of the host plant preferences of the leaf beetle, Agelastica alni. Chemoecology 12:125-131

Jamieson RW (2001) The essence of commodification: Caffeine dependencies in the early modern world. Journal of Social History 35:269-294

Jing SW, Coley PD (1990) Dioecy and herbivory: the effect of growth rate on plant defense in Acer negundo. Oikos 58:369-377

Johnson AF, Barbour MG (1990) Dunes and maritime forests In: Myers RL, Ewel JJ (eds) Ecosystems of Florida. The University of Central Florida Press, Orlando, FL, pp 429 – 480

Kaakeh W, Pfeiffer DG, Marini RP (1992) Combined effects of spirea aphid (Homptera Aphididae) and nitrogen fertilization on shoot growth, dry matter accumulation, and carbohydrate concentration in young apple trees. J Econ Entomol 85:496-506

Kannan D, Paliwal K (1997) Fertilization response om growth, photosynthesis, starch accumulation. And leaf nitrogen status of Cassia siamea Lam, seedlings under nursery conditions. J Sustain For 4:141-157

Kervinen T, Peltonen S, Teeri TH, Karjalainen R (1998) Differential expression of phenylalanine ammonia-lyase genes in barley induced by fungal infection or elicitors. New Phytol 139:292-300

Koricheva J, Larsson S, Haukioja E, Keinanen M (1998) Regulation of woody plant secondary metabolism by resource availability: hypothesis testing by means of meta-analysis. Oikos 83:212-226

Kumar NS, Hewavitharanage P, Adikaram NKB (1995) Attack on tea by Xyleborus fornicatus: inhibition of the symbiote, Monacrosporium ambrosium, by caffeine. Phytochemistry 4:1113-1116

Lou Y, Baldwin IT (2004) Nitrogen supply influences herbivore-induced direct and indirect defenses and transcriptional responses in Nicotiana attnuata. Plant Physiol 135:496 – 506

Mattson WJ (1980) Herbivory in relation to plant nitrogen content. Annu Rev Ecol Syst 11:119- 161

Mazzafera P (1994) Caffeine, theobromine, and theophylline distribution in Ilex paraguariensis. Rev Bras Fisiol Veg 6:149-151

Merrill WL (1979) The beloved tree. In: Hudson CM (ed) Black Drink: A Native American Tea. The University of Georgia Press, Athens, GA. pp 40-82

Montagnon C, Guyot B, Cilas C, Leroy T (1998) Genetic parameters of several biochemical compounds from green coffee, Coffea canephora. Plant Breeding 117:576-578

Nathanson, JA (1984) Caffeine and related methylxanthines: possible naturally occurring pesticides. Science 226:184-187

Nitao JK, Zangerl AR, Berenbaum MR (2002) CNB: Requiescat in pace? Oikos 98:540-546

O’Brien TG, Kinnaird MF (2003) Caffeine and conservation. Science 300:587

O’Brien TG, Kinnaird MF (2004) Response to: Conservation policy in coffee landscapes. Science 303:625-626

Onyilagha JC, Lazorko J, Gruber MY, Soroka JJ, Erlandson MY (2004) Effect of flavonoids on feeding preference and development of the crucifer pest Mamestra configurata. J Chem Ecol 30:109 -124

Ou BX, Hampsch-Woodill M, Prior RL (2001) Development and validation of an improved oxygen radical absorbance capacity assay using fluorescein as the fluorescent probe. J Agr Food Chem 49:4619-4626

Platt WJ, Schwartz MW (1990) Temperate hardwood forests. In: Myers RL, Ewel JJ (eds) Ecosystems of Florida. The University of Central Florida Press, Orlando, FL, pp 194 – 229

Power FB, Chestnut VK (1919) Ilex vomitoria as a native source of caffeine. J Amer Chem Soc 41:1307-1312

Punyasiri PAN, Abeysinghe ISB, Kumar V (2005) Performed and induced chemical resistance of tea leaf against Exobasidium vexans infection. J Chem Ecol 31: 1315-1324

Reginatto FH, Athayde ML, Gosmann G, Schenkel EP (1999) Methylxanthine accumulation in Ilex species – Caffeine and theobromine in erva-mate (Ilex paraguariensis) and other Ilex species. Soc Bras Quim 10: 443-446

Reichardt PB, Chapin FS, Bryant JP, Mattes BR, Clausen TP (1991) Carbon/nutrient balance as a predictor of plant defense in Alaskan balsam poplar: potential importance of metabolic turnover. Oecologia 88:401-406

Ritchie GA, Tanaka Y, Duke SD (1992) Physiology and morphology of Douglas-fir rooted cuttings compared to seedlings and transplants. Tree Physiol 10:179-194

Scheible W, Gonzáles-Fontes A, Lauere M, Müller-Röber B, Caboche M, Stitt M (1997) Nitrate acts as a signal to induce organic metabolism and repress starch metabolism in tobacco. Plant Cell 9:783-798

Scheible WR, Morcuende R, Czechowski T, Fritz C, Osuna D, Palcio-Rojas N, Schindelasch D, Thimm O, Udvardi MK, Stitt M (2004) Genome-wide reprogramming of primary and secondary metabolism, protein synthesis, cellular growth processes, and the regulatory infrastructure of Arabidopsis in response to nitrogen. Plant Physiol 136:2583-2499

Schinella GR, Troiani G, Dávila V, de Buschiazzo PM, Tournier HA (2000) Antioxidant effects of an aqueous extract of Ilex paraguariensis. Biochem Biophys Res Comm 269: 357-360

Slansky F, Wheeler GS (1992) Caterpillar’s compensatory feeding response to diluted nutrients leads to toxic allelochemical dose. Entomol Exp Appl 65:171-186

Spiller MA (1984) The Methylxanthine Beverages and Foods: Chemistry, Consumption, and Health Effects. Alan R Liss, Inc. New York, NY

Stewart AJ, Mullen W, Crozier A (2005) On-line high performance liquid chromatography analysis of the antioxidant activity of phenolic compounds in green and black tea. Mol Nutr Food Res 49: 52-60

Stitt M, Krapp A (1999) The interaction between elevated carbon dioxide and nitrogen nutrition: the physiological and molecular background. Plant Cell Environ 22:583-621

Sturtevant WC (1979) Black drink and other caffeine-containing beverages among non-Indians. In: Hudson CM (ed) Black Drink: A Native American Tea. The University of Georgia Press, Athens, GA. pp 150-165

Sultan SE (2000) Phenotypic plasticity for plant development, function and life history. Trends Plant Sci 5:537-542

Suzuki T, Ashihara H, Waller GR (1992) Purine and purine alkaloid metabolism in Camellia and Coffea plants. Phytochem 31(2575-2584)

Swain T, Hillis WE (1958) The phenolic constituents of Prunus domestica I. The quantitative analysis of phenolic constituents. J Sci Food Agr 10:63-68

Talcott ST, Howard LR, Brenes CH (2000) Antioxidant changes and sensory properties of carrot puree processed with and without periderm tissue. J Agri Food Chem 48:1315-1321

Terao J, Piskula M, Yao Q (1994) Protective effect of epicatechin, epicatechin gallate, and quercetin on lipid peroxidation in phospholipids bilayers. Arch Biochem Biophys 308:278- 284

Tuomi J, Niemelä P, Chapin FS, Bryant JP, Sirén, S. (1998) Defensive responses of trees in relation to their carbon/nutrient balance. In: Mattson J, Levieux J, Bernard-Dagan C (eds) Mechanisms of Woody Plant Defenses Against Insects: Search for Pattern. Springer- Verlag, NY. pp 57-72

Vandemeer J (2003) Tropical Agroecosystems. CRC Press, Boca Raton, FL van der Graff E, Hooykass P, Lein W, Lerchl J, Kunze G, Sonnewald U, Boldt R (2004) Molecular analysis of “de novo” purine biosynthesis in solanaceous species and in Arabidopsis thaliana. Front Biosci 9:1803-1816

Weaver LM, Herrmann KM (1997) Dynamics of the shikimate pathway in plants. Trends Plant Sci 2:1360-1385

Weinberg BA, Bealer BK (2002) The World of Caffeine: The Science and Culture of the World’s Most Popular Drug. Routledge Press, New York, NY

Yokozawa T, Dong E, Nakagawa T, Kashiwagi H, Nakagawa H, Takeuchi S, Chung HY (1998) In vitro and in vivo studies on the radical-scavenging activity of tea. J Agri Food Chem 46:2143-2150

Young IS, Woodside JV (2001) Antioxidants in health and disease. J Clin Pathol 54:176-186

BIOGRAPHICAL SKETCH

Matthew Palumbo was born in Manhasset, New York on September 20, 1970. His love of nature began when he was three years old with walks through the forest in rural New Jersey with

his father. He distinctly remembers being joined on one of these walks by his mother who subsequently had to turn back because she was too pregnant (with his brother) to balance herself on a log used to cross the creek that crossed their wooded path. His love of nature was fostered during childhood when his father brought home a garter snake from Vermont that became the family pet, which was followed with collections of fish, newts, fresh water crabs, and more snakes. Although music became his passion during high school and continues to be a great influence, he became enamored with anthropology, biochemistry, and ultimately botany during his undergraduate education, which he finished at SUNY at Stony Brook in 1998 with a bachelor’s degree in biochemistry. Matt then took a job as a chemist for a vitamin and herbal supplement company to increase his knowledge and skills in phytochemistry while taking classes at the New York Botanical Gardens and reading books on tropical biology and ethnobotany in his spare time. After taking a year of classes as a post-baccalaureate at the University of Florida, he began a Master’s degree program in botany there in January 2004, which he finished in

December 2006. During the last year of his master’s program, he had the opportunity to teach a course called “Plants in Human Affairs.” He found that teaching is hard but rewarding work. He plans to continue his research on the ecology and ethnobotany of plant secondary metabolites while pursuing a doctoral degrees and plans to teach as the opportunity arises.