Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1996 A novel mechanism for resistance to Colorado potato beetle (Leptinotarsa decemlineata) in wild Solanum species Evans Olonyi Sikinyi Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Ecology and Evolutionary Biology Commons, Entomology Commons, Environmental Sciences Commons, and the Plant Sciences Commons

Recommended Citation Sikinyi, Evans Olonyi, "A novel mechanism for resistance to Colorado potato beetle (Leptinotarsa decemlineata) in wild Solanum species " (1996). Retrospective Theses and Dissertations. 11414. https://lib.dr.iastate.edu/rtd/11414

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly firom the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be fi-om any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMl a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing fi'om left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infonnation Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600

A novel mechanism for resistance to Colorado potato beetle {Leptinotarsa decemlineata) in wild Solanum species

by

Evans Olonyi Sikinyi

A dissertation submitted to the graduate faculty in partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY

Major Horticulture Major Professor David Harmapel

Iowa State University Ames, Iowa 1996 DMI Number: 9712604

UMI Microform 9712604 Copyright 1997, by UMI Company. All rights reserved.

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, MI 48103 a

Graduate College Iowa State university

This is to certify that the doctoral dissertation of

Evans Olonyi Sikinyi has met the dissertation requirements of Iowa State University

Signature was redacted for privacy.

Signature was redacted for privacy. i^mttee member

Signature was redacted for privacy. Committee member

Signature was redacted for privacy.

Signature was redacted for privacy.

Signature was redacted for privacy. For the Major Departmem

Signature was redacted for privacy. Forth Graduate College lU

TABLE OF CONTENTS Page CHAPTER L GENERAL INTRODUCTION 1 Dissertation Organization 5 CHAPTER 2. A NOVEL MECHANISM FOR RESISTANCE TO COLORADO POTATO BEETLE (COLEOPTERA: CHRYSOMELIDAE) IN WILD SOLANUM SPECIES 9 Abstract 9 Introduction 10 Materials and Methods 13 Results 17 Discussions 20 Acknowledgements 23 Literature Cited 24 CHAPTER 3. CYSTEINE PROTEINASE INHIBITOR ACTIVITY IN WILD SOLANUM SPECIES RESISTANT TO COLORADO POTATO BEETLE ( LEPTINOTARSA DECEMLESIEATA) 35 Abstract 35 Introduction 36 Materials and Methods 39 Results 42 Discussions 44 Acknowledgements 46 Literature Cited 47 CHAPTER 4. GENERAL CONCLUSIONS 57 LITERATURE CITED 60 ACKNOWLEDGEMENTS 1

CHAPTER 1. GENERAL INTRODUCTION

The family Solanaceae includes about 90 genera and over 3000 species (Hawkes,

1979). It contains a number of economically valuable crops and vegetables such as potato

{Solanum tuberosum), tomato {Lycopersicon esculentum), peppers {Capsicum spp), and tobacco (Hicotiana tabacum). Others such as Petunia and ornamental tobacco are cultivated as flower crops. Many wild Solanum species, both tuberizing and nontuberizing, like

Solaman brevidens, S. spegazzinii, S. demissum, S. stoloniferum, S. chacoense and S. integrifolium, are known to contain valuable disease resistance or stress tolerance traits

(Foldo, 1987). Although they are not agronomically important, they can still be used as potential breeding materials in traditional breeding programs for cultivated solanaceous species or specific genes in their germplasm could be manipulated, through genetic engineering. Some wild Solanum species such as S. integrifolium, S. vemicosum, S. hjertingii, S. stoloniferum and S. demissum have been transformed with reporter genes

(Rotino, et al. 1992, Kumar, et al. 1995). These transformation systems provide an opportunity to put new genetic markers into wild species for breeding purposes.

The potato {Solanum tuberosum L.) is one of the world's most important food crops with its great potential for both protein and carbohydrate contributions to the diet (Horton,

1987). Its volume of production ranks fourth in the world after rice, wheat, and maize

(Horton, 1987). The Colorado potato beetle (CPB) is the most damaging insect pest of the potato worldwide, causing millions of dollars of economic loss annually in the United States 2 of America alone (Ferro et al., 1985). It is well established in western Europe and continues to spread through eastern Europe and into Asia (Hurst, 1975). Large-scale insecticide applications have been used to control the beetle in the United States and were generally efifective during the 1960s and 70s. However, in the last 30 years the CPB has shown a remarkable ability to develop resistance to all of the most effective chemical treatments.

(Ferro, 1985; Hare, 1990). This early success with insecticides resulted in less emphasis on research in biological control methods and host-plant resistance mechanisms. As a result, we know very little about the specific defense mechanisms employed by plants to defend against the CPB. The host range of the CPB is confined to about 20 species in the family Solanaceae with the most important agronomic hosts being the cultivated potato (Solanum tuberosum

L.), the tomato {Lycopersicon esculentum Mill.), and eggplant {Solanum melongena)

(Hsiao, 1988).

The most practical approach to CPB management is a comprehensive management system incorporating elements of chemical, biological, and cultural control with host-plant resistance. Unfortunately, very little is known about any specific resistance mechanisms to

CPB that may be present in commercial varieties of Solanum tuberosum. Because of the narrow genetic variability in most potato varieties, it is very unlikely that high levels of CPB resistance will be found in existing cultivars of S. tuberosum. The most promising attempts to identify and characterize resistance to CPB have been undertaken with wild species using two approaches: 1) high levels of foliar glycoalkaloids and 2) glandular trichomes. During the

1940s, European scientists screened numerous wild Solamm species for resistance to CPB. 3

Three of the species they identified, S. demissum, S. chacoense Bitter, and S. polyadenium, had demonstrated significant resistance to CPB which could be attributed to the presence of certain steroid glycoalkaloids in their foliage (Dimmock and Tingey 1985). The effectiveness of this defense mechanism depended on the concentration of specific glycoalkaloids and the life stage of the feeding insect (Schreiber, 1958; Sanford et al., 1994; Sanford et al.,1995).

Sinden et al. (1986) showed that acetylated glycoalkaloids (leptines) were the most potent deterrents. Leptine I was a feeding deterrent for both aduhs and larvae in potato (Sinden et al.,1986 ) while the more common glycoalkaloids, chaconine and solanine, would partially deter adult feeding only at very high concentrations (>200 mg/lOOg fi-esh mass). The leptines have been identified only in S. chacoense while chaconine and solanine are widespread and are the major glycoalkaloids present in S. tuberosum foliage.

Despite the obvious potential for utilizing these compounds as a defense against

CPB, breeding efforts have been slowed by several problems. The drawbacks include the polygenic nature of total glycoalkaloid (TGA) inheritance (Sanford and Sinden, 1972), the environmental influence on TGA metabolism and accumulation (Sinden, et al.,1984; Sinden and Webb, 1972), the high correlation between leaf and tuber TGA content (Deahl et al.,

1973), the technical difBculty of screening breeding populations for TGA (Dimmock and

Tingey 1985), and the differential activity of specific glycoalkaloids toward CPB (Tingey,

1984). Breeding programs have attempted to incorporate higher levels of specific glycoalkaloids into tuberosum germplasm with limited success (Sanford et al., 1984).

Solaman chacoense has been used for hybridization, but the resistance of S. chacoense X 4 subsp. tuberosum hybrids tends to be less than the wild parent (Sanford et al., 1984). Upon backcrossing to tuberosum, to improve tuberization characteristics, the levels of resistance decline considerably. Somatic hybrid potato plants between resistant S. chacoense and 5. tuberosum have been produced via protoplast electrofiision with resistance to CPB (Cheng et al. 1994). This is a tool to be used to overcome the crossing barriers between these species.

A second mechanism of host-plant resistance to CPB involves a foliar adaptation.

Some wild potato species may be protected from numerous arthropod pests by an adhesive exudate produced by glandular trichomes on the surface of the foliage (Gibson, 1971;

Gibson, 1976a). The process of this defense mechanism has been characterized in the insect- resistant Bolivian potato species S. berthaultii. Upon contact by the insect, the trichomes release a sticky secretion which effectively deters CPB populations by inhibiting larval growth and egglaying capacity of female adults (Casagrande, 1982; Gibson, 1976b). In addition to physical entrapment, chemical components of the glandular trichome exudate may have toxic effects on the CPB larvae and adults (Ryan et al. 1982). It has been demonstrated that this chemical defense reaction involves the production of a polymeric phenol by the action of polyphenol oxidase (Ryan et al. 1982). Resistance based on glandular trichomes has advantages over the glycoalkaloid mechanism as a pest management tool. Glandular trichomes provide resistance against a wide range of insect pests and S. berthaultii readily crosses with the cultivated potato producing hybrids that exhibited resistance through glandular trichomes (Wright et al. 1985). But despite these advantages, it has been reported 5

that CPB populations can adapt to this resistance mechanism after only two generations

(Groden and Casagrande, 1986).

Insect- plant relationship studies have shown that natural plant products

(allelochemicals) can act as defenses against herbivorous insects (Elden, 1995). Four classes

of plant proteins, including, proteinase inhibitors, alpha-amylase inhibitors, lectins and

lipoxygenases have been identified as the defensive factors (Gatehouse et al. 1992, Felton et al. 1994). A number of proteinase inhibitors have been shown to fiinction as a defense

mechanism against feeding insects (Gatehouse et al. 1992, Felton et al. 1994).

Proteinases are enzymes involved in the degradation of natural proteins. There are

four diflferent classes of proteinases: aspertic (carboxyl), cysteine (thiol), metalloproteins, and

serine which differ in their in vitro properties (Wolfson and Murdock, 1987; Barret, 1986).

The class for a specific proteinase is determined by the pH range over which it is maximally active, ability to hydrolyze specific proteins, similarity to well characterized proteinases, and

the sensitivity to various inhibitors (North, 1982). Insects use a variety of proteinases to

hydrolyze ingested proteins to release fi-ee amino acids necessary for normal growth and development (Thie and Houseman, 1990). A wide range of diversity in midgut proteolytic activity among insect species has been reported (Elden, 1995).

Proteinase inhibitors generally are low molecular weight proteins that form complexes

with proteinases, thereby inactivating these enzymes. Most Lepidoptera species demonstrate

presence of serine proteinases (Christeller et al. 1992), while cysteine proteinases are common in Coleopteran species (Murdock et al. 1987). In vitro activity on the midgut 6

homogenates or insect growth and development on artificial diets have been used to determine the effect of specific proteinase inhibitors on selected insects (Murdock et al. 1987,

Oppert et al. 1993).

Plant proteinase inhibitors are single gene products and are therefore targeted for

transfer of protection against insect pests through genetic engineering (Elden, 1995). Serine

proteinase inhibitor genes fi-om cowpea, {Vigna ungiiiculata L.), tomato (Lycopersicon esciilentum Mill.), and potato, have been transferred into tobacco, {Nicotiana tabacum L.), and expressed adverse effect on two species of Lepidopteran herbivores (Hilder et al. 1987,

Johnson et al. 1989).

Jasmonate (JA) and its methyl ester form, methyl jasmonate (MJ) have been identified in many plant species and have demonstrated hormone-like properties (Vick and Zimmerman

1984). Jasmonate and its derivatives are known to have an effect on many physiological changes in plants such as leaf senescence and abscission, chlorophyll degradation, inhibition of photosynthesis, seed germination, induction of stomata closure, and tuber formation

(Sembdner and Parthier, 1993). Exogenous treatments of methyl jasmonate regulate the expression of several specific proteins such as serine proteinase inhibitor I and II fi-om potato and tomato (Farmer et al. 1992, IDldmann et al. 1992, Pena-Cortes et al. 1992). The accumulation of these proteinase inhibitors is developmentaUy regulated in growing tubers and environmentally regulated in response to wounding in leaves (Sanchez-Serrano et al.

1986). Proteinase inhibitor genes are activated at the wound site in leaves, and also systemically activated in nonwounded leaves of wounded plants (Pena-Cortes et al. 1988). 7

Since methyl jasmonate can mimic both wound responses in the induction of proteinase

inhibitor genes in potato and tomato leaves (Farmer et al. 1992, Hildmann et al. 1992), it has

been proposed that both jasmonate and methyl jasmonate act as secondary messengers in the

wound signal transduction pathway (Ryan, 1992).

In order to continue the attempts to identify efifective host-plant resistance mechanisms to CPB, a detailed screening of several wild Solanum species with known resistance was undertaken. This pool of resistant germplasm was collected by scientists of the Potato Introduction Station and characterized through a preliminary screening for beetle feeding. Some of the wild species examined in this study showed resistance to CPB in the preliminary screening, with very low levels of foliar glycoalkaloids and no glandular trichomes. Our goal in this research was to identify Solanum germplasm with novel host- plant resistance mechanisms to CPB which either may be incorporated into breeding programs or used as a trial for improvement through biotechnological applications. The objectives of this study were therefore to 1) screen a select number of Solanum species and accessions for resistance to the Colorado potato beetle through bioassays (feeding, beetle growth, and mortality, 2) examine plant leaf morphology factors that could affect resistance, 3) use chemical analysis of leaf compounds mcluding unique or rare glycoalkaloids which may be efifective as antifeedants and attempt to identify them, and 4) use blot hybridization techniques to characterize the expression pattern of cysteine proteinase inhibitor genes present in the Solanaceous species that are induced by beetle feeding and methyl jasmonate. By identifying and characterizing this novel mechanism of resistance, it is our goal to eventually incorporate this mechanism into breeding material for production of commercial varieties through traditional crossing methods or molecular techniques. 8

Dissertation Organization This dissertation is arranged in an alternate format consisting of two papers suitable for publication. The first paper, "A novel mechanism for resistance to Colorado potato beetle (Coleoptera: Chrysomelidae) in wild Solcorum species," has been submitted to ih&Jourtial of Economic Entomology. The second paper, "Cysteine proteinase inhibitor activity in wild Solanum species resistant to Colorado potato beetle," will be submitted to Plant Physiology. Evans Sikinyi was the principal investigator under the supervision of Dr. David J. Hannapel on all research reported herein, and is the first author on the two papers. The papers are preceded by the General Introduction and followed by the General Conclusion. Literature Cited in the General Introduction and General Conclusion are listed in alphabetical order according to author's name following General Conclusion. The arrangement of the papers follows the guidelines set forth by each journal in their Instructions to Authors. 9

CHAPTER 2. A NOVEL MECHANISM FOR RESISTANCE TO COLORADO POTATO BEETLE (COLEOPTERA CHRYSOMELIDAE) IN WILD SOLANUM SPECIES

A paper accepted by the Journal of Economic Entomology

Evans Sikinyi', David Hannapel, Paula M. Imerman*, and H. M. Stahr*

Department of Horticulture, Iowa State University Ames, Iowa 5001 l-llOO

* College of Veterinary Medicine, Veterinary Diagnostic Laboratory, Iowa State

University, Ames, Iowa 50011-1250

'Present address: National Horticultural Research Centre, P.O. Box 220, Thika, Kenya

Abstract

The Colorado potato beetle {Leptinotarsa decemlineata Say) is the most destructive insect

pest of the potato {Solcamm tuberosum L.) worldwide and has shown remarkable adaptability to insecticides. In wild Solamm species, the best known host-plant resistance mechanisms are a high level of foliar glycoalkaloids and the presence of specialized, glandular trichomes.

Through a preliminary field-test screen, three Solamm species, Solamm trifidum, S. raphanifolium, and S. circaeifolium, were identified with low foliar glycoalkaloid levels and no glandular trichomes, but still exhibiting substantial resistance to L decemlineata. These three species along with two species, S. berthaultii and S. chacoense, representing the two 10 known mechanisms of resistance, were examined in bioassays for larval leaf consumption, effect on larval growth, and percent larval mortality. The best S. trifidum accessions had approximately ten-fold less foliar consumption per larva than controls and induced 54 % reduction in larval weight compared with controls during a 24 h feeding period. Forty-eight h mortality rates were 100 % for second and third instars feeding on S. trifidum compared with

22 % for the S. tuberosum controls. Despite very low foliar glycoalkaloid levels and the absence of glandular trichomes, S. trifidum accessions exhibited both an effective antinutritive and deterrent mechanism for resistance to L. decemlimata.

BCEY WORDS ; Host-plant resistance, Leptinotarsa decemlineata, Solamim trifidum

INTRODUCTION

Leptinotarsa decemlineata (Colorado potato beetle) is the most damaging insect pest of the potato worldwide, causing millions of doUars of losses annually in the United States alone (Ferro 1985). It is well established in western Europe and continues to spread through eastern Europe and into Asia (Hurst 1975). The host range of the beetle is confined to about twenty species in the family Solanaceae with the most important agronomic hosts being the cultivated potato {Solartum tuberosum L.), tomato (Lycopersicon esadentimt Mill.), and eggplant {Solanum melongena) (Hsiao 1988). Large-scale insecticide applications have been used to control the beetle in the United States, and were generally effective during the 1960s 11

and 70s. In the last 30 years, however, the L. decemlineata has shown a remarkable ability to develop resistance to all of the most effective chemical treatments. (Ferro 1985, Hare 1990).

The most practical approach to L. decemlineata management is a comprehensive

management system incorporating elements of chemical, biological, and cultural control with

host-plant resistance. Unfortunately, very little is known about any specific resistance mechanisms to L. decemlineata that may be present in commercial varieties of Solamim tuberosum. The most promising attempts to identify and characterize resistance to L decemlineata have been undertaken with wild species using two approaches: 1) high levels of foliar glycoalkaloids and 2) glandular trichomes. Sinden et al. (1986) showed that acetylated glycoalkaloids (leptines) were the most potent deterrents of L decemlineata. Leptine I was a feeding deterrent for both adults and larvae in potato (Sinden et a/. 1986), while the more conmion glycoalkaloids, chaconine and solanine, would partially deter adult feeding only at very high concentrations (>200 mg/lOOg fi-esh mass). The leptines have been identified only in S. chacoense while chaconine and solanine are widespread and are the major glycoalkaloids present in S. tuberosum foliage. Despite the obvious potential for utilizing these compounds as a defense against L. decemlineata, breeding efforts have been slowed by the polygenic nature of total glycoalkaloid (TGA) inheritance (Sanford & Sinden 1972), the high correlation between leaf and tuber TGA content (Deahl et al. 1973) and the differential activity of specific glycoalkaloids toward L decemlineata (Tingey 1984).

A second mechanism of host-plant resistance to L. decemlineata involves a foliar adaptation. Some wild potato species may be protected fi^om numerous arthropod pests by an 12 adhesive exudate produced by glandular trichomes on the surface of the foliage (Gibson

1971, Gibson 1976a). The process of this defense mechanism has been characterized in the insect-resistant Bolivian potato species S. berthaultii. Upon contact by the insect, the trichomes release a sticky secretion which effectively deters L decemlineata populations by inhibiting larval growth and egg-laying capacity of female adults (Casagrande 1982, Gibson

1976b). Glandular trichomes provide resistance against a wide range of insect pests and S. berthaultii readily crosses with the cultivated potato producing hybrids that exhibited resistance through glandular trichomes (Wright et a/. 1985). But despite these advantages, it has been reported that L. decemlineata populations can adapt to this resistance mechanism after only two generations (Groden & Casagrande 1986).

To continue the attempts to identify effective host-plant resistance mechanisms to L decemlineata, we have undertaken a detailed screening of several wild Solanum species with known resistance. This pool of resistant germplasm was collected by scientists at the Potato

Introduction Station, Sturgeon Bay, Wisconsin, and identified through a preliminary screening for beetle feeding. The three wild species examined in this study showed resistance to L decemlineata, but had very low levels of foliar glycoalkaloids and no glandular trichomes. Our goal in this research was to characterize a novel host-plant resistance mechanism to L decemlineata which could potentially be used in potato germplasm enhancement programs. 13

Materials and Methods

Plant culture. Seeds for the plant material were obtained from the Potato

Introduction Station, Sturgeon Bay, WI. Test species included Solanwn trifidum (PI 283064

& PI 283104), S. raphanifolium (PI 473526 & PI 473371), and S. circaeifolhm (PI

498116). Accessions of the test species were selected on the basis of a previous study

screening for L. decemlimata resistance conducted by the Potato Introduction Station. These

species showed resistance to L. decemlimata, had low total foliar glycoalkaloid content, and

did not have glandular trichomes. Putative susceptible accessions of S. trifidum (PI 283064)

and S. raphanifolium (PI 473371) were included as checks. Two species with known

resistance mechanisms were also used: Solanum berthaultii (PI 310926 & PI 265858) with

glandular trichomes and S. chacoense (PI 320123 & PI 320311) with high levels of foh'ar

glycoalkaloids. The commercial variety S. tuberosum 'Superior* was used as a susceptible

control. Plants were grown in greenhouses, maintained at 25 °C day and 23 °C night, with

supplemental lighting from high-pressure sodium lamps providing an irradiance of at least,

300 ^mol • s"' • m'^ at the plant canopy, a photoperiod of 16:8 (L:D) h, and received weekly

fertilization (17-5-16 ; N: P: K). About ten seeds of each accession were germinated and

transplarrted at the 4 to 5 leaf stage into 22-cm azalea pots. Certified seed tubers were used for 'Superior*. No insecticides were applied to plant material used for all feeding assays.

Insect rearing. Colorado potato beetle {Leptinotarsa decemlimata Say.) eggs were obtained from a colony maintained at the Department of Entomolgy, Iowa State University. 14

During the two-year test period, beetles from controlled field studies were introduced on a regular basis to maintain the genetic vigor of the population. Beetle populations were reared on either S. tuberosum 'Superior* whole plants or stem cuttings in growth chambers maintained at 26 °C day and 25 °C night, with a photoperiod of 16:8 (L:D) h. Stem cuttings were maintained for up to 48 h in test tubes with water plugged with cotton. Eggs from the adults were collected and either stored in vials at 5 °C or allowed to hatch by incubating at 25 to 28 °C. Larvae were reared to adults on pre-bloom to bloom-stage

'Superior* potato plants. Mature fourth instar larvae were transferred to containers with moist vermiculite for pupation. Feeding assays were performed with second, third, or fourth instars.

Leaf feeding. Feeding trials were conducted to determine whether there were any preferences m feeding by the beetles among the several accessions and species, and to determine the beetle stage with the highest feeding rate (adults or larvae). In a choice feeding assay, bouquets of all the accessions were evaluated for beetle feeding by either the adults or larvae, and the percentage leaf consumption estimated.

To assess leaf consumption in a no-choice experiment and to determine the effect of feeding on larval development, third and fourth instars were allowed to feed on excised leaves of the different accessions. Ten separate accessions from six species, including

'Superior* as a control, were evaluated in three separate experiments. Each experiment consisted of three replications. An experimental unit consisted of three to eight leaves in a

28x20x10 cm plastic storage box lined with wetted foam at the bottom and filter paper to maintain moisture. Leaves were excised from the mid-section of the plants, placed on moist 15 filter paper, and their area determined using the L1-3100 Area Meter (Li- COR, Inc.,

Lincoln, NE) before and after feeding. Attempts were made to have approximately the same leaf area in all the treatments. Ten larvae were assayed for feeding per box. Leaf consumption and larval mass were measured after feeding for 24 h. Mortality rates for beetles were monitored over a 48 h feeding period using the second and third instars.

Glandular trichome evaluation. Foliar trichome morphology was examined by using a Scanning Electron Microscope (SEM) to determine whether the foliage of any of these species had glandular trichomes similar to S. berthaultii. Fully expanded leaves were excised fi-om greenhouse-grown plants, and leaf discs 1.2 cm in diameter were cut, fi-ozen by liquid nitrogen, coated with gold for 4 min. and observed on a cold stage at 10 kV. The specimens were prepared in an Emscope SP 2000 cryogenic system and examined in a JEOL

JSM-35 SEM. The four species examined were S. berthaiiltii, S. trifidum, S. circaeifolium, and S. tuberosum 'Superior*.

Glycoalkaloid analysis. Glycoalkaloid analysis was performed on third and fourth leaves fi'om the shoot tip, fi'eshly harvested fi"om three-month old, greenhouse-grown plants.

The extraction methods previously described by Bushway et al. (1986) with modifications were used. After harvest, leaves were cut into small pieces, weighed in 2.0 g samples and extracted with 20 ml of tetrahydrofiiraniwateracetonitrileiacetic acid, (50:30:20:1). The samples were extracted for 40 min by using rotary extraction. The extract was vacuum filtered (Whatman No. 4), and a 5-ml sample was transferred to a 20 ml scintillation vial to which 10 ml of 0.4 % 1-heptanesulfonic acid containing 1.0 % acetic acid was added. Five 16

milliliters (representing 0.17 g of fresh tissue) were added to a pre-conditioned Cig SepPak®

cartridge (Millipore Water Corp., Milford, MA) using the cleanup procedures of Carmen et

al. (1986) with modifications of Bushway et al. (1986). Quantitation and chararcterization of total foliar glycoalkaloids were accomplished using thin layer chromatography (TLC). The

TLC analysis was performed using 20x20 cm reverse phase plates, LKCig with pre-absorbent strip, (Whatman Inc., Fairfield, NJ). A composite standard was prepared at a concentration of

0.05 lig/^l for chaconine, solanine, solanidine, and demissidine; 0.1 |ig/|il for tomatine and

tomatidine; and 1.0 |ig/|il for lepidine (Sigma Chemical Co., St. Louis, MO) in HPLC grade

methanol (Fisher Scientific, Chicago, IL). Standard curves for all compounds were constructed by spotting 20,40, and 80 |il of the composite standard along with 100 (il of the

samples. The TLC plates were developed in ethanol:water acetic acid (65:35; 1) previously

used for ergot alkaloids (Stahr 1991). The glycoalkaloids were visualized by iodine fiimes as described by Deahl and Sinden (1987). The total glycoalkaloids were calculated by comparison of like Rf s to the standard curve and the unknowns were calculated by using the

standard curve of the closest Rf to the unknown band.

Data analysis. Data from the feeding bioassays were collected as the leaf area consumed per beetle and larval mass (g) change. Results from the bioassays are presented as the average of the means of three separate experiments carried out over a two-year period.

The means were treated as replications in a pooled data analysis. Leaf consumption, larval

mass change, and larval mortality data were analysed using the general linear model (GLM)

procedure of SAS (SAS Institute, 1995) and least significant difference (LSD) for multiple 17 comparisons. The glycoalkaloid data were reported as the mean of three determinations with standard deviations.

Results

Leaf feeding. Significant differences in beetle leaf consumption were shown among the species (F = 84.8; df = 8,14;P = 0.0001). In a 24 h period, three species (two accessions of S. trifidum, two of S. berthaultii and one S. chacoense) showed considerable inhibition of leaf consumption (Fig. 1) (LSD = 1.45, df = 14, F = 0.0001, Table I). One S. trifidum (0.84 cm^/larva) and one S. berthaultii (0.87 cm^/larva) accession had approximately ten-fold less foliar consumption per larva than 'Superior* (8.4 cm^/larvae). Leaf consumption levels for both S. circaeifolium (8.7 cm^/larva) and S. raphanifoliiim (7.2 cm^/Iarva) were not different fi-om the control (Table I), despite preliminary field screening based on leaf consumption, indicating some resistance to L. decemlineata. Two different seedlots of S. circaeifolium PI

498116 were used during the test. The initial lot used in preliminary feeding tests indicated that S. circaeifolium PI 498116 and S. trifidum PI 283104 had the least leaf damage with only 1 to 2 % of the leaf consumed compared with 45 % consumption for the 'Superior* control after a 24 h feeding assay (data not shown). In these preliminary feeding trials, even after 72 h, leaf consumption was 10 % and 5 % for S. circaeifolium and S. trifidum^ respectively, compared with 95 % consumption for the control. Plants fi'om this seedlot were used for first and second replicates for S. circaeifolium. We observed that, overall, adult beetle leaf consumption rate was significantly lower than the larval rate, but even with adult 18

feeding, 'Superior* and S. raphanifolium leaves were the most damaged while S. trifidian had

the least damage. There were no replicate effects between the first and second experiments.

However, there were significant effects with the third replicate using the second seedlot for S. circaeifolitmt and for this reason the S. circaeifolium data was not included in Table 1.

Significant differences were shown for larval mass change among the species tested

(F = 22.25; df = 10, 17; P = 0.0001). All of the species examined showed a significant difference in reduction of larval mass compared with 'Superior* except S. raphanifolium

(LSD = 0.12, df = 17, P = 0.0001, Table I). The highest levels of growth reduction were exhibited by larvae feeding for 24 h on S. berthaultii PI 310926 (57 %), S. berthaiiltii PI

265858 (54 %), and S. trifidimi PI 283104 (54 %) (Fig. 2). Solanum chacoense PI 320123 and S. trifidum PI 283064 demonstrated a 42 % reduction compared with 'Superior*, whereas, the lowest mass reductions were exhibited by larvae feeding on S. circaeifolium (20

%) and S. raphanifolium (5 %). These differences in larval mass reduction were consistent even after 96 h of leaf feeding (data not shown). There was no significant replication effect on this set of data.

There were significant differences in larval mortality rates between beetles feeding on

'Superior* and accessions of S. trifidum, S. berthaultii, and S. chacoeme (LSD = 30.6, df=

13, f = 0.0034, Table I). The highest larval mortality rates were on S. trifidum PI 283104

(100 %), S. chacoense PI 320123 (90 %), and S. berthaultii (84 and 80 % for the two accessions. Fig. 3) (F = 12.50, df = 7, 13, P = 0.0034). Solanum chacoense PI 320311,

'Superior*, and S. circaeifolium had 18, 22, and 49 % mortality, respectively. 19

Glandular trichome evaluation. Solamm berthmiltii had the highest density of foliar trichomes (Fig. 4A) with the two expected types of trichomes. Type A, the short type with a four-lobed gland at its apex (Fig. 4B), and the longer type B, with an ovoid gland at the tip, occurred in approximately equal numbers (see arrows. Fig. 4A). Solatium trifidum leaves had approximately one-third the number of trichomes present on S. berthmiltii leaves

(Fig. 4C), and although short and long trichomes were evident (see arrows. Fig. 4C), both types lacked any kind of glands. Further magnification of the short trichome type (Fig. 4D) revealed an amorphous mass at the tip, and the presence of a waxy-like substance on the leaf surface. Solanum circaeifolium had no visible trichome structures (Fig. 4E), and upon fiirther magnification (Fig. 4F), waxy deposits were evident on the leaf surface. Solamim tuberosum 'Superior* leaves had only a few trichomes and none of these showed any visible evidence of glands (not shown).

Glycoalkaloid content analysis. Estimates by TLC of total glycoalkaloids quantities in fi-eshly harvested foliage, indicated very low amounts (0 to 17 mg/100 g fi-esh mass) in S. circaeifolium, and S. trifidum accessions (Table n), medium to low amounts (24 to 60 mg/100 g fi-esh mass) in S. raphanifolium, S. chacoense (PI 320311), and 'Superior*, and high amounts (96-186 mg/100 g fresh mass) in S. berthaultii and S. chacoense (PI 320123). As expected, the most abundant glycoalkaloids identified were chaconine and solanine. A unique compound that could be a glycoalkaloid (or aglycone), but did not match any standards, was detected in the HPLC profile from leaf extracts of S. trifidum PI 283104. 20

Discussion

The results of this study showed that even though accessions of S. trifidum have low levels of foliar glycoalkaloids and no glandular trichomes, they demonstrated substantial reduction in beetle feeding, induced retardation of larval growth, and increased the rate of mortality relative to the S. tuberosum control. Foliar trichome morphology was examined to determine if the foliage of any of these Solanum species had glandular trichomes similar to S. berthaultii which are known to play a role in resistance to L. decemlineata and other insects

(Tingey & Laubengayer 1981, Tingey & Yencho 1994). Solanum berthaultii has high densities of the two types (A and B) of trichomes that are associated with insect resistance

(Tingey & Laubengayer, 1981). Exudates of type A trichomes darken and harden after rupture of the four-lobed gland, whereas those of type B hairs remain clear and viscous

(Gibson & Turner 1977). The basis for this chemical defense mechanism involves the production of a polymeric phenol by the action of polyphenol oxidase (Ryan et al. 1982).

Though trichome-based resistance has been attributed to physical entrapment and the chemical composition of the trichome exudates. Franca and Tingey (1994), suggested the major impact of S. berthaultii on L decemlineata to be in the physiology of digestion and reproduction. In our study, beetle larvae feeding on the sticky exudate from the trichomal glands retarded the insect's mobility and feeding habits, leading to reduced leaf consumption, retarded development, and a high mortality rate. Solanum trifidum, which has no glandular trichomes, demonstrated antifeeding and developmental retardant levels comparable to those of S. berthaultii. We conclude from our results that these levels of resistance are not due to 21

the same glandular trichome mechanism of S. berthaultii. It is conceivable that the waxy-like

substances on the leaf surfaces of resistant S. trifidum and S. circaeifolium accessions play a

role ui resistance to the beetle, although, we did not observe any waxy substances on beetles

feeding on these species. In chickpea, components of leaf exudates have been shown to act as

insecticidal agents as part of the plant's defense mechanism against insects (Yoshida et al.

1995).

Foliar glycoalkaloid analysis on the species and accessions was conducted to

determine the levels of total glycoalkaloids, and to attempt to correlate the presence of any

specific glycoalkaloids with resistance to Colorado potato beetles. High foliar glycoalkaloid

levels have been a basis for resistance breeding using S. chacoense (Sanford et al. 1994). We

compared two S. chacoense accessions in this study, one with a relatively high glycoalkaloid

level (PI 320123) and the other with a 69 % lower level ( PI 320311, Table II). In the

mortality study, PI 320123 had a five-fold greater rate of mortality than PI 320311 (Fig. 3

and Table I). Our data indicate lower levels of total glycoalkaloids in S. trifidum, S.

circaeifolium and 'Superior", while S. chacoense and S. berthaultii had relatively higher levels

(Table II). In a preliminary study by the Potato Introduction Station, Sturgeon Bay,

Wisconsin, S. trifidum and S. circaeifolium accessions had undetectable levels of foliar

glycoalkaloids (John Bamberg, personal communication). This indicates that the resistance in

S. trifidum is not due to high levels of foliar glycoalkaloids. However, specific glycoakaloids,

such as the leptines of S. chacoense, can confer resistance at much lower levels. Deahl and

Sinden (1987), demonstrated that leptine I at 100 mg/100 g fresh mass concentration 22 deterred feeding by adult beetles using potato leaf discs. However, for levels below 20 mg/100 g fresh mass, feeding was only slightly deterred. The S. trifidim accessions analyzed so far had total foliar glycoalkaloids less than 20 mg/100 g fresh mass, thus likely eliminating leptines as the resistance factor in this species. It is possible there may be a novel, potent glycoalkaloid present in S. trifidum that is effective against the beetle at very low concentrations. This seems unlikely, however, because some S. trifidum accessions with substantial resistance had no detectable levels of foliar glycoalkaloids (accession PI 283064, for example).

Using choice feeding tests, we observed the consistent avoidance of S. trifidim PI

283104 and S. berihaultii PI 265858 and PI 310926 by both the adults and larvae. When given no choice in feeding, beetle larvae and adults consumed significantly lower amounts of the resistant leaves compared with the 'Superior* control. The first seedlot for S. circaeifolitim was not preferred while the second lot was consumed by the beetles indicating possible segregation and variation among different seedlots. Larvae that fed on these

Solamim species (S. trifidim, S. berihaidtii and high glycoalkaloid S. chacoense), had significantly reduced larval weight (development) and had a higher mortality rate. Solamm berthaultifs resistance is due to the high density of glandular trichomes which inhibit feeding by retarding body movement through entrapment, via the trichome exudate (Casagrande

1982, Franca & Tingey 1994). Solanum chacoense's resistance is due to high levels of foliar total glycoalkaloids, which are toxic to the insects (Sinden et al. 1986). For S. trifidum, reduction in larval growth could be explained by both the larvae eating less or the presence of 23

some toxic foliar compound. Solarmm trifidum accessions had the best resistance against L.

decemlineata demonstrated in this study, equal to, if not better than S. berthaultii and S.

chacoense. Despite very low foliar glycoalkaloid levels and the absence of glandular

trichomes, S. trifidim accessions exhibited effective resistance to Colorado potato beetle

which could be due to an antinutritive or deterrent mechanism.

Acknowledgements

We thank John Obrycki and Loren Stephens for reviewing the manuscript and Paul

Hinz for assistance with data analysis. Journal paper no. J-16951 of the Iowa Agriculture

and Home Economics Experiment Station, Ames, Iowa, Project no. 3056, and supported by

Hatch Act and State of Iowa fimds. We gratefully acknowledge fiinding for this research

from a Carver Trust Grant through Iowa State University and the U.S. Agency for

International Development- Kenya National Agricultural Research Project.

References Cited

Bushway, R. J., J. L.Bureau & J. King. 1986. Modification of the rapid high liquid

chromatographic method for the determination of potato glycoakaloids. J. Agric. Food

Chem. 34; 277-287. 24

Carmen, Jr., A. S., S. S. Kuan, G. M. Ware, O. J. Francis, Jr. & G. P. Kirschenheuter.

1986. Rapid high-performance liquid chromatograhic determination of the potato glycoalkaloids alpha-solanine and alpha-chaconine. J. Agric Food Chem 34: 279-282.

Casagrande, R A. 1982. Colorado potato beetle resistance in a wild potato, Solamim berthaultii. J. Econ. Entomol. 75: 368-372.

Deahl, K. L., R. J. Young, & S. L. Sinden. 1973. A study of the relationship of late blight

resistance to glycoalkaloid content in fifteen potato clones. Am. Potato J. 50: 248-253.

Deahl, K. L. & S. L. Sinden. 1987. A technique for rapid detection of leptine glycoalkaloids in potato foliage. Am. Potato J. 64: 285-290.

Ferro, D. N. 1985. Pest status and control strategies of the Colorado potato beetle.

Massachusetts Ag. Expt. Sta. Res. Bull. 704: 1-8.

Franca, F. H. & W. M. Tingey. 1994. Solamm berthaiiltii Hawkes affects the digestive system, fat body and ovaries of the Colorado potato beetle. Amer. Pot. J. 71: 405-410.

Gibson, R. W. 1 971. Glandular hairs providing resistance to aphids in certain wild potato species. Ann. Appl. Biol. 68: 113-119.

Gibson, R.W. 1976a. Glandular hairs are a possible means of limiting aphid damage to the

potato crop. Ann. Appl. Biol. 82: 143-146.

Gibson, R. W. 1976b. Glandular hairs of Solcamm polyadeniim lessen damage by the

Colorado potato beetle. Ann. Appl. Bio. 82: 147-150.

Gibson, R. W. & R. H. Turner. 1977. Insect-trapping hairs on potato plants. PANS (Pest

Artie. News Summ.) 22: 272-277. 25

Groden, E. & R. A. Casagrande. 1986. Population dynamics of the Colorado potato beetle,

Leptinotarsa decemlineata (Coleoptera: Chrysomelidae) on Solanum berthcailtii. J. Econ.

Entomol. 79: 91-97.

Hare, J. D. 1990. Ecology and management of the Colorado potato beetle. Annu. Rev.

Entomol. 35; 81-100.

Hsiao, T. H. 1988. Host specificity, seasonality, and bionomics of Leptinotarsa beetles.

Biology of Chrysomelidae, Dordrecht, Kluwer, pp. 581-599.

Hurst, G. W. 1975. Meteorology and the Colorado potato beetle. World Meteorological

Organization Tech. Note 137. 51 pp.

Ryan, J. D., P. Gregory & W. M. Tingey. (1982). Phenolic oxidase activities in glandular

trichomes of berthaultii. Phytochem. 21: 1885-1887.

Sanford, L. L. & S. L. Sinden. 1972. Inheritance of potato glycoalkaloids. Am. Potato J.

49: 209-217.

Sinden, S. L., L. L. Sanford, W. W. Cantelo & K. L. Deahl. 1986. Leptine

glycoalkaloids and resistance to the Colorado potato beetle in Solanum chacoense. Env.

Entomol. 15: 1057-1062.

Stahr, H.M. 1991. Ergot alkaloids in feed: Thin layer chromatography. In Analytical

Methods in Toxicology, pp. 109-111, John Wiley and Sons, Inc., New York.

Tingey, W. M. 1984. Glycoalkaloids as pest resistance factors. Am. Potato J. 61: 157-167.

Tingey, W. M. & J. E. Laubengayer. 1981. Defense against green peach aphid and potato

leafhopper by glandular trichomes of Solanum berthaultii. J. Econ. Entomol. 74: 721-725. 26

Tingey, W. M. & G. C. Yencho. 1994. Insect resistance in potato; A decade of progress, pp. 405-425. In G. W. Zehnder, M. L. Powelson, R. K. Jansson & K. V. Raman [eds.].

Advances in potato pest biology and management. APS Press, St. Paul, Minnesota.

Wright, R. J., M. B. Dimock, W. lVI.Tingey & R. L. Plaisted. 1985. Colorado potato beetle (Coleoptera: Chrysomelidae): expression of resistance in Solcamm berthaultii and interspecific potato hybrids. J. Econ. Entomol. 78: 576-582.

Yoshida, M., S. E. Cowgill & J. A. Wightman. 1995. Mechanism of resistance to

Helicoverpa armigera (Lepidoptera: Noctuidae) in chickpea: Role of oxalic acid in leaf exudate as an antibiotic factor. J. Econ. Entomol. 88: 1783-1786.

Received for publication ; accepted . 27

Table 1. Multiple comparison analysis of effects of L. decemlineata feeding on Solamm

species and accessions. Each column represents separate test parameters for either feeding

habits (Leaf consumption) or beetle development (Larval mass change. Percent larval

mortality). The species and accessions used were S. trifidum PI 283104 (trf 104), PI 283064

(trf064), and PI 283537 (trf 537), S. berthaultii PI 310926 (ber 926) and PI 265858 (ber

858), S. chacoense PI 320123 (chc 123) and PI 320311 (chc 311), S. circaeifoliiim PI

498116 (crc 116), S. raphanifolittm PI 473526 (rap 526), and S. tuberosum 'Superior*.

Solamm species Leaf consumption Larval mass change Percent (cm^/larva) (relative to 1.0) larval mortality trf 104 0.84 c* 0.83 egh 100 a trf 064 1.69 c 0.95 be NT trf 537 1.12c 0.98 b NT ber 926 0.87 c 0.80 fh 80 a ber 858 1.66 c 0.83 cdef 84 a chc 123 2.32 c 0.95 bdg 90 a chc 311 NT NT 18 b rap 526 7.16 b 1.32 a NT 'Superior* 8.40 ab 1.37 a 22 b LSD 1.45 0.12 16.3

•Numbers followed by the same letter are not significantly different. NT= Not tested for this parameter. 28

Table H. Total glycoalkaloid levels in fresh leaf tissue of several Solatium species determined

by thin layer chromatography. Levels in the Previous Results column were based on a study

done by the Potato Introduction Station, Sturgeon Bay, Wisconsin, and are included only as a

reference. The species and accessions used were S. berthaultii PI 265858 (ber 858), S.

berthaultii PI 310926 (ber 926), S. trifidum PI 283104 (trf 104), S. trifidum PI 283064 (trf

064), S. circaeifolium PI 498116 (crc 116), S. tuberosum 'Superior*, S. raphanifolhm PI

473371 (rap 371), 51 chacoeme PI 320123 (chc 123), and S. chacoense PI 320311 (chc

311).

Solcmum species Total glycoalkaloid level (mg/100 g fr mass)

Present Study Previous Results

ber 858 96 ±4 96 ber 926 104 ± 14 165 trf 104 17±4 <5 trf 064 ND* <5 crc 116 ND <4

'Superior* 24 ±2 30 rap 371 24 ±3 8 chc 123 192 ±6 186 chc 311 60 ±5 156

* Not Detected 29

Figure Legends

Fig. I. Mean leaf area consumption per larvae for the various Solanum species in a no choice

feeding test over a 24 h period. The means were calculated from three experiments

containing three replicates, each replicate having ten larvae. Three to eight leaves were used

(depending on leaf size) per test sample to achieve approximately, equivalent leaf area for all

species at the start of the feeding. Leaf area was measured by scanning with an LI-3100

Area Meter (Li- COR, Inc., Lincoln, NE). Third and fourth instars were used. The species

and accessions used were S. trifidum PI 283104 (trf 104), S. trifidum PI 283064 (trf064), S.

berthaultii PI 265858 (ber 858), S. berthaultii PI 310926 (ber 926), S. chacoense PI 320123

(chc), S. circaeifoliiim PI 498116 (crc), S. raphanifolium PI 473526 (rap) and S.

tuberosum 'Superior*.

Fig. 2. Percent larval mass reduction for fourth mstar Colorado potato beetles after feeding

on the foliage of various Solamm species for 24 h, compared to 'Superior* control. The

means were calculated from three experiments with three replicates each and ten larvae per

replication. The species and accessions used were S. trifidum PI 283104 (trf 104), S. trifidum

PI 283064 (trf 064), S. berthaultii PI 265858 (ber 858), S. berthaultii PI 310926 (ber 926),

S. chacoense PI 320123 (chc), S. circaeifolium PI 498116 (crc), S. raphanifoliimi PI

A13526 (rap) and S. tuberosian 'Superior*. 30

Fig. 3. Percent Colorado potato beetle larva mortality after feeding on leaf cuttings of various Solanum species for 48 h period. The percent mortality represents the mean of two experiments. Each experiment consisted of three replications with ten second/third instars per replication. The species and accessions used were S. trifidum PI 283104 (trf 104), S. berthaultii PI 310926 (ber 926), S. berthaultii PI 265858 (ber 858), S. chacoense PI 320123

(chc 123), S. circaeifolium PI 498116 (ore 116), S. chacoeme PI 320311 (chc 311) and

S.tuberosum 'Superior'.

Fig. 4. Scanning electron micrographs of the leaf surfaces of S. berthaultii (panels A and B)

S. trifidum (panels C and and S. circaeifolium (panels E and F). Panels A, C, and E have a magnification of 60X, the bar representing 100 pim. Panels B, D, and F have a magnification of440X, the bar representing 20 ^m. The top-left arrow in panel A indicates the shorter trichome type (type A) of S. berthaultii, whereas, the center arrow of panel A indicates the four-lobed, type B trichome. The arrows in panel C indicate the two types of trichomes present on leaves of S. trifidum. Panel D is a magnification of the short, knobby type of trichome. Waxy-like substances were observed on the surfaces of S. trifidum and S. circaeifolium (panels D and F). 0 "1 Leaf area consumption (cm larva ) Oi-JtOOJtls^OlONVlcXlvOO ' I' 111111111111111111111111 11 11 1111 11 trf 104 -ni

ber 926

trf 064

ber 858

Superior Percent reduction in larval growth

trf 104

ber 926 c/) trf 064 o

ber 858 t/i Q. U)ro

Superior - Percent larval mortality

trf 104

chc 123

bcr 926 - 31 oC/5 U> d5" bcr 858

U) TS n P. crc 116 U)n>

chc 311

Superior

35

CHAPTER 3. CYSTEINE PROTEINASE INHIBITOR ACTIVITY IN WILD SOLANUM SFECmS RESISTANT TO COLORADO POTATO BEETLE (LEPTINOTABSA DECEMLJNEATA SAY)

A paper prepared for submission to Plant Physiology

Evans O. Sikinyi*, David Hannapel, and Caroline J. Bolter^

Department of Horticulture, Iowa State University Ames, Iowa 50011-l 100 ^Agriculture and Agri- Food Canada, Pest Management Research Centre, 1391 Sandford St., London, Ontario, N5V 4T3, Canada. *Present address: National Horticultural Research Centre, P.O. Box 220, Thika, Kenya

ABSTRACT Plant proteinase inhibitors function as a defense mechanism against feeding insects by disrupting gut proteolytic digestion. Potato multicystatin (PMC) is a member of the cystatin superfamily of cysteine proteinase inhibitors and is effective in retarding insect growth by inhibiting the cysteine proteinases commonly found in the guts of numerous beetle species. In potatoes, the accumulation of these proteinase inhibitors is developmentally regulated in growing tubers and environmentally regulated in response to wounding in leaves. In this study, three Solanum species with resistance to Colorado potato beetle were examined for the induction of PMC and its inhibitory activity against papain. Two of the species, S. chacoense and S. berthaultii, have known resistance mechanisms, whereas, S. trifidum 36

exhibits a novel host-plant resistance mechanism against the Colorado potato beetle. PMC induction occurred in all three resistant species in response to methyl jasmonate treatment and wounding by beetle feeding. These species showed accumulation of PMC transcripts within 24 h in response to methyl jasmonate treatment in leaf cuttings and on whole plants in response to wounding by beetle feeding. Two of the resistant species showed very high levels of PMC activity as measured by papain inhibition in response to beetle feeding. It is likely that PMC activity in the resistant species serves as an antinutrutive mechanism of defense but is only one component in the plant's overall defense system.

KEY WORDS : Host-plant resistance, Leptinotarsa decemlineata, Solanum trifidiim. Potato multicystatin proteinase inhibitor.

INTRODUCTION There are a number of plant proteinase inhibitors that function as a defense mechanism against feeding insects (Elden 1995). These inhibitors act to disrupt protein digestion in the insect's gut. In potatoes, the accumulation of these proteinase inhibitors is developmentally regulated in growing tubers and environmentally regulated in response to wounding in leaves (Sanchez-Serrano et al. 1986). Proteinase inhibitor genes are not only activated at the wound site in leaves, but are also systemically activated in nonwounded leaves of wounded plants (Pena-Cortes et al. 1988). Methyl jasmonate has been shown to mimic both wound responses in the induction of proteinase inhibitor genes in potato and tomato leaves (Farmer et al. 1992, Hildmann et al. 1992), and it has been proposed that methyl jasmonate acts as a secondary messenger in the wound signal transduction pathway (Ryan 1992). Most proteinase inhibitors contain a signal peptide that is cleaved during packaging of the mature form into vacuoles.(Walker-Simmons 1977, Ishikawa et al. 1994). 37

Proteinase inhibitors generally are low molecular weight proteins that form complexes with proteinases, thereby inactivating these enzymes. Proteinases are protein degrading enzymes utilized by insects to breakdown plant proteins for digestion and assimilation (Wolfson and Murdock 1990). Insects use a variety of proteinases to hydrolyze ingested proteins to release free amino acids necessary for normal growth and development (Thie and Houseman 1990). A wide range of diversity in midgut proteolytic activity among insect species have been reported (Elden 1995). Digestive proteinases utilized by insects include serine proteinases ( e.g., trypsin and chymotrypsin), active in the pH 7.0-10.0 and inhibited by Bowman-Birk, Kunitz and lima bean inhibitors; cysteine proteinases (e.g., papain), active at mildly acidic range, (5.0-7.0), and inhibited by heavy metals, cystatin and E-64; aspartic proteinases (e.g., pepsin), active in the acidic range below pH 4.5, and inhibited by pepstatin (Barret 1977). Many Coleoptera, including the Colorado potato beetle, use cysteine and aspartic proteinases to digest proteins (Wolfson and Murdock 1987, Thie and Houseman 1990, Purcell et al. 1992, Michaud et al. 1993). In vitro assays on midgut homogenates or analysis of insect growth on artificial diets have both been used to determine the efifect of specific proteinase inhibitors on selected insects (Murdock et al. 1987, Oppert et al. 1993). Detrimental effects on the growth of specific Coleoptera species by cysteine proteinase inhibitors has been reported (Chen et al., 1992; Orr et al., 1994). E-64, which inhibits all known cysteine proteinases, significantly retards Colorado potato beetle and other Coleoptera larva growth (Wolfson and Murdock 1987, Murdock et al. 1988, Oppert et al. 1993). Potato multicystatin (PMC), a multigene family encoding an 85-kD polypeptide, occurs within subepidermal cells of potato tubers and is an effective inhibitor of cysteine proteinases, which include papain, ficin and chymopapain (Rodis and Hoflf 1984). Amino acid sequence analysis reveals that PMC is a member of the cystatin superfamily of cysteine proteinase inhibitors and that it is composed of eight cystatin domains (protease binding 38

sites). PMC can be induced to high levels in leaves in response to methyl jasmonate and wounding, and like other proteinase inhibitors is linked to the plant defense system (Walsh and Strickland 1993). Specifically, PMC is effective against insects that utilize cysteine proteinases by inhibiting gut proteinases and retarding insect growth (Orr et al. 1994). The Colorado potato beetle is the most damaging insect pest of the potato worldwide, causing millions of dollars of economic loss annually in the United States alone (Ferro 1985). It is well established in western Europe and continues to spread through eastern Europe and into Asia (Hurst 1975). Large-scale insecticide applications have been used to control the beetle in the United States, and were generally effective during the 1960s and 70s. However, in the last 30 years the Colorado potato beetle has shown a remarkable ability to develop resistance to all of the most efifective chemical treatments. (Ferro 1985; Hare 1990). Wild Solanum species have demonstrated resistance to Colorado potato beetle that is absent in the cultivated potato. The best known host-plant resistance mechanisms in these species are a high level of foliar glycoalkaloids and the presence of specialized, glandular trichomes on leaves. Accessions fi'om three species were selected for evaluation of the cysteine proteinase activity. These included S. chacoense (high foliar glycoalkaloid content) and S. berthaultii (specialized foliar glandular trichomes), representing the two known mechanisms of resistance, and S. trifidum, demonstrating an effective antinutrient and deterrent mechanism for resistance, despite very low levels of total glycoalkaloids and the absence of glandular trichomes (Sikinyi et al. 1996). Solanum trifidum accessions demonstrated tenfold less foliar consumption by Colorado potato beetle larvae than the control, a 54% reduction in larval growth, and a 100% mortality rate (22% for control) after 48 h. These results indicated the existence of novel host-plant resistance mechanisms against the Colorado potato beetle in Solanum trifidum. The objectives of the study were to examine the expression of a potato cysteine proteinase inhibitor in Solanum species that are resistant to Colorado potato beetle 39

and to evaluate the relation between cysteine proteinase inhibitor activity and resistance to the Colorado potato beetle.

MATERIALS AND METHODS Plant culture. Solanum species used in the study were grown from seed obtained from the Potato Introduction Station, Sturgeon Bay, WI. These included Solamm trifidum (PI 283104), Solanum berthaultii, (PI 310926), S. chacoense (PI 320123 and PI 320311). Solcmum tuberosum 'Superior*, a commercial susceptible type was grown from certified seed tubers obtained from Cornell University. Using feeding assays and by monitoring larval development, these species were tested for resistance to Colorado potato beetle by Sikinyi et O O al. (1996). Plants were grown in greenhouses, maintained at 25 C day and 23 C night, with supplemental lighting from high-pressure sodium lamps providing an irradiance of at least, 300 [imol*s"^*m"^ at the plant canopy, a photoperiod of 16:8 (L:D) h, and a weekly fertilization (17-5-16 ; N: P: K) applied. About ten seeds of each accession were germinated and transplanted at the 4 to 5 leaf stage into 22-cm azalea pots. No insecticides were applied to plant material used for all feeding assays. Methyl jasmonate treatments. Leaf-petiole cuttings from Solanum tuberosum 'Superior*, S. trifidum, S. chacoense and S. berthaultii plants were used for methyl jasmonate treatments. Leaves were excised from the midsection to the top of the plant (avoiding the very young leaves) by cutting the petiole close to the stem. The cut ends were placed in 40 ml of solution of either 100 |iM methyl jasmonate (Bedoukian Industries, Danbury, CT) or water. Methyl jasmonate was dissolved in 95% ethanol and diluted in sterile deionized water to make a final concentration of 100 |jM methyl jasmonate. Control treatments consisted of water with equivalent levels of ethanol without the methyl jasmonate. AH methyl jasmonate treatments consisted of adequate leaf-petiole cuttings for at least lOg sample each. The treatments were 40

kept in a growth chamber at 25°C with light, for 24 hours. The control treatments were kept separate from other treatments to prevent exposure to volatile methyl jasmonate derivatives. After the incubation periods, leaf blades from the treatments were harvested (midveins removed), frozen in liquid nitrogen, and stored at -80 °C. Whole plant beetle feeding. Third and fourth instars were loaded on the test plants, covered with nylon cloth to confine the beetles. These plants were kept in a growth chamber maintained at 25 °C with light, a photoperiod of 16:8 (L:D) h, for 24 hours. Control plants had no insects and were kept under same conditions. Leaf samples with visible signs of feeding damage were harvested, frozen in liquid nitrogen, and stored at -80 °C . Insects from these plants were saved for gut extraction and analysis. Leaf cuttings feeding. Each replicate consisted of a 27x20x10 cm plastic crisper box lined with wetted foam at the bottom and filter paper to maintain moisture. Excised leaves were placed on moist filter paper, loaded with at least ten third or fourth instars and allowed to feed for 24 h. Nucleic Acid Analysis Southern hybridization. Genomic DNA was extracted from fresh shoot tips of the four Solanum species using the CTAB (cetyltrimethylammonium bromide) method (Rogers and Bendich 1985). Gels were stained with ethidium bromide, visualized, and photographed under UV light. For Southern hybridization, after denaturation and neutralization, gels were blotted onto a nylon membrane (Micron Separations Inc.). The membrane filter was hybridized overnight at 42 °C with a 0.99 kb PMC probe (Waldron et al. 1993) that was P -labeled by nick translation. The hybridization buffer and conditions were the same as described by Hansen and Hannapel (1992). Northern hybridization. Total leaf RNA from the frozen leaf samples was isolated by the phenol/chloroform method described by Suh et al. (1991). For northem blots, RNA was 41

subjected to electrophoresis in 1.4% agarose gels containing 5 mM methylimercuric hydroxide (Alfe Products, Danvers, MA, USA) in tris-borate buffer (pH 7.5) and blotted onto nylon membranes. RNA was detected with ethidium bromide under UV light to ascertain consistent loading of gels and efiBcient transfer to nylon membranes. A 1.2 kb wheat 18S ribosomal RNA probe was used to confirm uniform loading on replicate blots. Blots were hybridized with the 0.99 kb PMC probe in 50% (v/v) formamide, 5x SSC, 50x Denhardt's, 25 mM sodixmi phosphate (pH 6.8), 0.5% SDS, and 250 (ig/ml sheared sahnon sperm DNA. After overnight hybridization at 42 °C, blots were washed two times with Ix SSC / 0.1% SDS at room temperature for 5 and 30 min. Filters were then washed twice in O.lx SSC / 0.1% SDS at room temperature and at 60 °C for 30 and 20 min, respectively. Hybridized probe was detected by exposure of the filters to X-OMAT AR film for 12 to 48 hours. Measurement of papain inhibitor activity Preparation of leaf extracts. Third and fourth leaves from the apex were excised from three month-old plants and used for feeding experiments over a 24 h period. After feeding, the leaves were frozen at -20 °C for 48 h, then thawed, and centrifiiged at lOOOg at 6-8 °C for 20 min to extract the cell sap and stored at -80 °C. The analysis and measurement of papain inhibitor activity in the plant extracts were done according to Bolter (1993). Briefly, 100 |il of 2 jiM papain enzyme was incubated with 1 to 20 |jJ of plant extract (25X diluted in MES buffer) at 37 °C for 10 min before k was stopped with TCA. Leaf juice was incubated with casein alone as a control. The results were reported as percent inhibition of papain activity. Preparation of gut extracts. Early fourth mstar larvae were fed on the four Solanum species (berthaultii, trifidum, chacoense, and 'Superior") for 6-12 h. The larvae used for the extraction weighed between 100-130 mg, and those with empty guts were discarded. Gut 42

extraction was the same as described by Bolter and Jongsma (1995) with minor modifications. The larvae were placed at -20 °C briefly to immobilize them, pinned on an aluminum foil-covered styroform board behind the head capsule. The last abdominal segment was cut off and a probe used to gently rub the body to force out the gut. After blot drying the guts on tissue paper, they were weighed, fi^ozen in liquid nitrogen and stored at -80 °C. Gut assay of proteolytic activity was the same as described by by Boher and Jongsma (1995)

RESULTS Using a partial-length PMC cDNA probe, the Southern blot analysis of Figure 1 shows that there are numerous genomic fragments which hybridize to the PMC sequence. As previously reported, it would appear that the PMC family of genes is composed of from three to six genes, depending on the species. The most intensely hybridizing bands came from genomic DNA of S. trifidum (Fig. I, lane 1). Even though bands were identified by hybridization, this is no indication that these bands represent fiinctional genes. Because there is evidence that methyl jasmonate treatments of leaves can induce PMC expression (Bolter 1993), we used methyl jasmonate induction on petiole leaf cuttings to quickly ascertain the functionality of any potential PMC genes in these four species. Figure 2 shows the results of northern hybridization of RNA from leaf cuttings treated with methyl jasmonate to determine whether the cysteine proteinase inhibitor genes could be induced in the four Solanum species. There was PMC expression at the RNA level in all four species examined. The strongest induction was in S. trifidum and S. berthaultii, with lower levels of transcript accumulation in the leaves of 'Superior" and S. chacoense. To verify consistent loading of total RNA for all samples, an 18S ribosomal RNA probe was used (Fig. 2, panel B). Knowing that the cysteine proteinase inhibitor genes could be 43

activated at the RNA level, we next determined if wounding in the form of beetle feeding would induce the PMC genes in these resistant species. To accomplish this, we used beetles feeding on whole plants for 24h. The results of the northern blot analysis of the total RNA extracted from leaves of these Solanum species is shown in Fig 3. There was accumulation of PMC RNA for all species, the highest shown with S. trifidum and S. berthaiiltii while S. chacoense had slightly less induction. Solanum tuberosum 'Superior* had the lowest level of transcript accumulation and this accumulation was constitutive for all three samples. These results indicated that the PMC genes were induced by beetle feeding at the RNA level in all three resistant species. To conjBrm equal loading of the RNA samples, an 18S ribosomal RNA probe was used (Fig. 3, panel B). To measure the activity of PMC induced in leaves fed upon by third instar Colorado potato beetles for 24 h, a papain inhibitor assay was perfomed on leaf extracts from all four species (Fig. 4). Papain is used for assaying PMC activity because of its high sequence homology with both cathepsin B and H, the cysteine proteinases found in Colorado potato beetle gut (Thie and Houseman 1990), PMC activity is represented in Fig. 4 as percent inhibition of papain activity. Excised leaf cuttings were used for this beetle feeding induction experiment because they provide more uniform feeding and facilitate monitoring. The highest papain inhibition was shown with S. trifidum (78%) and S. chacoense ( 54% and 62 % for accessions PI 320123 and PI 320311, respectively). Papain inhibition was only 7% for 'Superior* and less than 2% for S. berthaultii. Papain inhibition activity was greater than 35% for all S. trifidum accessions tested and even nonwounded plants showed constitutive activity (data not shown) 44

DISCUSSION Potato muiticystatin represents a famUy of proteinaceous inhibitors of cysteine proteases present in a diverse number of plant species including tomato, rice, and maize (Waldron et al. 1993, Xu et al. 1993). It is an important proteinase inhibitor because many Coleopteran species, including the Colorado potato beetle, use cysteine proteases for most of the proteolytic activity present in their digestive tracts (Orr et al. 1994, Murdock et al. 1987). Our results showed induction of PMC genes in all three species resistant to Colorado potato beetle, S. trifidum, S. chacoense, and S. berthaultii. These species showed accumulation of PMC transcripts within 24 h in response to methyl jasmonate treatment in leaf cuttings and on whole plants in response to wounding by beetle feeding. Papain inhibition levels in wounded leaves, however, did not necessarily correlate with PMC transcript accumulation (compare Fig. 4 and Fig. 3). The highest transcript levels were detected in wounded leaves from S. trifidum and S. berthaultii. Whereas S. trifidum leaves had consistently high levels of papain inhibitor activity, S. berthaultii leaves had very little, if any, papain inhibition activity. All of the S. chacoense accessions tested here had high levels of papain inhibition but S. chacoense PI 320123 had a relatively low level of PMC transcript accumulation in beetle-fed leaves. The discrepancy between S. trifidum and S. berthaultii could be explained by increased efficiency of RNA translation with S. trifidum, producing more PMC polypeptide resulting in greater activity. If leaf PMC protein levels were proportional, then it is possible that the S. trifidum PMC form has more binding afBnity for papain. Because of the complexity of the PMC polypeptide, it is possible that a change in structure or size could modify the effectiveness of the proteinase binding domains. Both the S. trifidum and S. chacoense forms could represent modified polypeptides, with greater accessibility to the cystatin domains resulting in more efBcient binding at lower concentrations. PMC activity as measured by papain inhibition was consistently very high 45

(up to 78% inhibition. Fig. 4) for the accessions of the resistant species, S. trifidwn. In a previous study by Sikinyi et al. (1996), S. trifidum exhibited a tenfold reduction in leaf consumption and a 54 % reduction in larval growth compared to the control. Mortality rates were 100 % for larvae feeding on S. trifidum leaves after 48 h compared to 22 % for the control. In choice feeding experiments, Colorado potato beetle larva consistently avoided feeding on S. trifidum leaves. From these results, we concluded that S. trifidum exhibited both an antinutritive and deterrent mechanism against Colorado potato beetle (Sinden et al. 1986). The question arises then, can the high levels of PMC activit>' in S. trifidum account for its high degree of resistance. Within the S. chacoeme accessions, it is not entirely clear what PMC activity contributes to overall resistance. Both S. chacoeme accessions had high levels of papain inhibition (>50%), but in earlier studies (Sikinyi et al. 1996) the high total glycoalkaloid accessions (PI 320123) had twofold less leaf consumption (2.32 cm^/larva for PI 320123 compared to 4.92 cm^/larva for PI 320311) and a fivefold greater mortality rate (90% vs 18%) than the low glycoalkaloid accession PI 320311. These two accessions were equivalent only in larval weight reduction with 42% less weight decrease for both compared to the control (Sikinyi et al. 1996). It is possible that the high induced levels of PMC could explain the comparable levels of larval growth retardation by reducing digestibility and that the difference in total glycoalkaloids levels could explain the discrepancy in mortality rates. It seems unlikely that PMC accumulation in the leaves of S. trifidum could alone account for its remarkable resistance against Colorado potato beetle. Bolter and Jongsma (1995) showed that Colorado potato beetle larvae rapidly adapted to induced foliar papain inhibitors by compensating for inhibited gut proteolytic activity by synthesizing insensitive proteinases. Proteinase inhibitors normally do not cause such rapid rates of insect mortality and there is no evidence that proteinase inhibitors contribute to a deterrent mechanism of defense that can be detected by insects before feeding. All of the S. trifidum accessions 46

examined demonstrated, at least, a low level of constitutive PMC accumulation in leaves. If PMC works on beetle gut proteases, inhibition could begin immediately upon beetle feeding enhancing the inhibitory effect on larval growth. Both the Colorado potato beetle and S. trifidum originated from similar habitats of the Mexican plateau and so it is likely that S. trifidum has evolved a number of defense systems against this aggressive predatory insect (Lu and Lazzell 1996). It is plausible that PMC induction in the leaves of S. trifidum fimctions as an antinutritive mechanism of defense but that it represents just one of the potentially numerous host-plant resistance mechanisms against the Colorado potato beetle present in S. trifidum.

ACKNOWLEDGEMENTS We thank for review of the manuscript. Journal paper no. J- of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa. Project no. 3056. We gratefully acknowledge fimding for this research from a Carver Trust Grant through Iowa State University and the US Agency for International Development- Kenya National Agricultural Research Project. 47

LITERATURE CITED Barrett A J (1977) Introduction to the history and classification of tissue proteinases in A.J. Barret (ed). Proteinases in mammalian cells and tissues. Amsterdam; Elsevier / North- Holland Biomedial pp. 1-21 Bolter CJ (1993) Methyl jasmonate induces papain inhibitor(s) in tomato leaves. Plant Physiol. 103: 1347-1353 Bolter CJ, Jongsma MA (1995) Colorado potato beetles (Leptinortarsa decemlineata) adapt to proteinase inhibitors induced by methyl jasmonate. J Insect Physiol 41: 1071-1078 Chen MS, Johnson B, Wen L, Muthukrishnan S, Kramer KJ, Morgan TD, Reeck GR (1992) Rice cystatin: Bacterial expression, purification, cysteine proteinase inhibitory activity, and insect growth supression activity of a truncated form of the protem. Prot Express Purif 3; 41-49 Elden TC (1995) Selected proteinase inhibitor effects on alfalfa weevil (Coleoptera; Curculionidae) growth and development. J Econ Entomol 88:1586-1590 Farmer EE, Johnson RR, Ryan CA (1992) Regulation of expression of proteinase inhibitor gene by methyl jasmonate and jasmonic acid. Plant Physiol 98: 995-1002 Ferro DN (1985) Pest status and control strategies of the Colorado potato beetle. Massachusetts Ag Expt Sta Res Bull 704: 1-8 Hare JD (1990) Ecology and management of the Colorado potato beetle. Annu Rev Entomol 35: 81-100 Hansen JD, Hannapel DJ (1992) A wound-inducible potato proteinase inhibitor gene expressed in non-tuber bearing species is not sucrose inducible. Plant Physiol 100: 164-169 Hildmann T, Ebneth M, Pena-Cortes H, Sanchez-Serrano JJ, Willmitzer L, Prat S (1992) General role of abscisic and jasmonic acids in gene activation as a result of mechanical wounding. Plant Cell 4:1157-1170 48

Hurst GW (1975) Meteorology and the Colorado potato beetle. World Meteorological

Organization Tech Note 137; 1-51

Ishikawa A, Ohta S, Matsuoka K, Hattori T, Nakamura K (1994) A family of potato genes that encode Kunitz-type proteinase inhibitors: Structural comparisons and differential expressions. Plant Cell Physiol 35: 303-312

Lu W, Lazell J (1996) How did an innocuous Mexican weed-eating beetie become an agricultural scourge of global proportions ? Natural BDstory 1:36-39

Michaud D, Nguyen-Quoc, Yelle S (1993) Selective inhibition of Colorado potato beetie cathepsin H by oryzacystatins I and H. FEBS Lett 331: 173-176

Murdock LL, Brookhart G, Dunn FE, Foard DE, Kelley S, Kitch L, Shade RE,

Shukle RH, Wolfson JL (1987) Cysteine digestive proteinases in Coleoptera. Comp

Biochem Physiol STB: 783-987

Murdock LL, Shade RE, Pomeroy NU (1988) Effects of E-64, a cysteine proteinase inhibitor, on cowpea weevil growth, development and fecundity. Environ Entomol 17: 467-

469

OppertB, Morgan TD, Culberston C, Kramer KJ (1993) Dietary mixtures of cysteine proteinase inhibitors exhibit synergergistic toxicity toward the red flour beetie, Tribolium castcmeum. Comp Biochem Physiol 105C: 379-385

Orr GL, Strickland JA, Walsh TA (1994) Inhibition of Diabrotica larval growth by multicystatin from potato tubers. J Insect Physiol 40: 893-900 49

Pena-Cortes H, Sanchez-Serrano J, Rocha-Sosa M, Willmitzer L (1988) Systemic induction of proteinase inhibitor II gene expression in potato plants by wounding. Planta 174:

84-89

Pureed JP, Greenplate JT, Sammons RD (1992) Examination of midgut luminal proteiase activities in six economically important insects. Insect Biochem Mol Biol 22: 41-47

Rodis P, HofF JE (1984) Naturally occurring protein crystals in the potato. Plant Physiol

74: 907-911

Rogers SO, Bendich A J (1985) Extraction ofDNA from milligram amounts of fresh, herbarium, and mummified plant tissues. Plant Mol Biol 5: 69-76

Ryan CA (1992) The search for the proteinase inhibitor-inducing factor, PUF. Plant Mol

Biol 19:123-133

Sanchez-Serrano JR, Schmidt J, Schell J, Willmitzer L (1986) Nucleotide sequence of

proteinase inhibitor n encoding cDNA of potato (Solanum tuberosum) and its mode of expression. Mol Gen Genet 203:15-20

Sinden SL, Sanford LL, Cantelo WW, Deahl KL (1986) Leptine glycoalkaloids and resistance to the Colorado potato beetle in Solanum chacoense. Env. Entomol. 15: 1057- 1062. Sikinyi EO, Hannapel DJ, Immerman PM, Stahr HM (1996) A novel mechanism for resistance to Colorado potato beetle (Coleoptera: Chrysomelidae) in wild Solanum species. J Econ Entomol In press Suh SG, Stiekema WJ, Hannapei DJ (1991) Proteinase-inhibitor activity and wound- inducible gene expression of the 22-kDa potato-tuber proteins. Planta 184: 423-430 50

Thie NMR, Houseman JG (1990) Identification of Cathepsin B, D and H in the larval midgut of Colorado potato beetle, Leptinotarsa decemlineata Say (Coleoptera; Chrysomelidae). Insect Biochem 20: 313-318 Waldron C, Wegrich LM, Merio PO, Walsh TA (1993) Characterization of a genomic sequence coding for potato multicystatin, an eight-domain cysteine proteinase inhibitor. Plant Mol Biol 23: 801-812 Walker-Simmons M, Ryan CA (1977) Immunological identification of proteinase inhibitor I and n and in isolated tomato leaf vacuoles. Plant Physiol 60: 61-63 Walsh TA, Strickland JA (1993) Proteolysis of the 85-kilodalton crystalline proteinases fi-om potato releases flmctional cystatin domains. Plant Physiol 103: 1227-1234 Wolfson JL, Murdock LL (1987) Suppression of larval Colorado potato beetle growth and development by digestive proteinase inhibitors. Entomol Exp Appl 44: 235-240 Wolfson JL, Murdock LL (1990) Diversity in digestive proteinase activity among insects. J ChemEcol 16: 1089-1102 Xu DD, McElory R, Thomburg W, Wu R (1993) Systemic induction of a potato pin2 promoter by wounding, methyl jasmonate, and abscisic acid in transgenic rice plants. Plant Mol Biol 22: 573-588 51

Figure legends Fig 1. Southern blot analysis of genomic DNA from the four Solcmum species S. trifidum PI 283104 (lane I), S. berthaultii PI 310926 (lane 2), S. chacoense PI 320123 (lane 3), and S. tuberosum 'Superior* (lane 4). Ten micrograms ofDNA was digested with EcoRI, subjected to electrophoresis in a 0.9 % agarose gel, transferred to a nylon membrane (Micron Separations Inc.), and probed with a 0.99 kb PMC fragment (Waldron et al., 1993) that was ^^P -labeled by nick translation. Size markers are shown in kilobases (kb) in the far right lane.

Fig 2. Northern blot hybridization of total RNA extracted from the leaves of leaf-petiole cuttings of four Solomon species, S. trifidum (trf), S. chacoense (chc), S. berthaultii (ber), and S. tuberosum 'Superior" (Sup) induced by 100 jiM methyl jasmonate for a 24 h period. Ten micrograms of total RNA for each sample was subjected to electrophoresis in 1.4% agarose gels containing 5 mM methylmercuric hydroxide and blotted onto nylon membranes. Blots were hybridized with a 0.99 kb PCM probe (panel A). As a control, RNA extracted from nontreated leaves at 0 and 24h was used. An IBS wheat ribosomal RNA probe was used to confirm uniform loading of total RNA in each lane (panel B) 52

Fig. 3. Northern blot hybridization of total RNA extracted from the leaves of four Solaman species, S. trifidum (trf), S. chacoense (chc), S. berthaultii (ber), and S. tuberosum 'Superior* (Sup) wounded by Colorado potato beetle feeding for a 24 h period. Ten micrograms of total RNA for each sample was subjected to electrophoresis in 1.4% agarose gels containing 5 mM methylmercuric hydroxide and blotted onto nylon membranes. Blots were hybridized with a 0.99 kb PCM probe (panel A). As a control, RNA extracted from leaves of nonwounded plants at 0 h and 24 h (C) was used. An 18S wheat ribosomal RNA probe was used to confirm uniform loading of the total RNA in each lane (panel B)

Fig 4. Percent inhibition of papain activity in leaf extracts from various Solanum species. The Colorado potato beetle larva were fed on excised leaf cuttings incubated on moist filter paper in petri dishes for a 24 h period. Leaves from the mid-section of three month-old plants were used for feeding experiments over a 24-hour period. The percent papain inhibition in leaf extracts was measured against 100 |al of 2 (iM papain in 1 to 20 (il of leaf extract. The activity of any papain inhibitor present in the leaf was reported as percent inhibition of papain activity. The percent represents the mean of three replications consisting of ten second or third instars per replication. The species and accessions used were S. trifidum PI 283104 (trf 104), S. berthaultii PI 310926 (ber 926), S. chacoense PI 320123 (chc 123), S. chacoense PI 320311 (chc 311) and S. tuberosum 'Superior*. 53

1 2 3 4 kb

-9.4

-4,3

-0.5

Fig. 1

o

k' trf

' . •.' ! •';•- chq 21 w' ., '• \ •• - •• •' u> . ^.S^; ber

Sup 56

trf 104 ber 926 chc 123 Sup chc 311 Solanum species IHI Oh »B8gH 24h

Fig. 4 57

CHAPTER 4. GENERAL CONCLUSIONS

Colorado potato beetle {Leptinotarsa decemlineata Say) is the most destructive

insect pest of the potato (Solanum tuberosum L.) worldwide causing millions of dollars of

economic loss annually in the United States of America alone (Ferrol985). It is well

established in western Europe and continues to spread through eastern Europe and into Asia

(Hurst 1975). Large-scale insecticide applications have been used to control the beetle in the

United States, and were generally effective during the 1960s and 70s. However, in the last

30 years the Colorado potato beetle has shown a remarkable ability to develop resistance to

all of the most effective chemical treatments. (Ferro 1985, Hare 1990). Wild Solanum species have demonstrated resistance to the Colorado potato beetle absent in the cultivated

potato. In wild Solanum species, the best known host-plant resistance mechanisms are a high level of foliar glycoalkaloids and the presence of specialized, glandular trichomes on leaves.

The overall objectives of the research project were to evaluate the host-plant resistance in wild Solanum species and attempt to identify possible resistance mechanism that can be incorporated into cultivated potatoes. This study focuses on Solanum germplasm with novel host-plant resistance mechanisms to the Colorado potato beetle for use in genetic enhancement. Through a preliminary field-test screen, three Solanum species, Solanum trifidinn, S. raphanifolium, and S. circaeifolium were identified with low foliar glycoalkaloid levels and no glandular trichomes, but still exhibiting substantial resistance to Colorado potato beetle. The criteria for choosing these species were for those that had resistant and susceptible accessions, had low glycoalkaloids and were diploids. These three species along with two species, S. chacoense and S. berthaultii representing the two known mechanisms of resistance, were assayed for larval leaf consumption, effect on larval growth, and percent mortality rate. The best S. trifidum accessions had approximately ten-fold less foliar consumption per larva than controls and induced a 41% reduction in larval weight compared to controls over a 24 h feeding period. Mortality rates were 100 % for second and third instars feeding on S. trifidum for 48 h compared to 21 % for the S. tuberosum controls.

Despite very low foliar glycoalkaloids levels and the absence of glandular trichomes S. trifidum accessions had among the best resistance demonstrated in this study. These accessions exhibited both an effective antinutritive and deterrent mechanism for resistance to

Colorado potato beetle. These results indicated the existence of novel host-plant resistance mechanisms against the Colorado potato beetle in Solamtm trifidum.

Further studies were done in an attempt to identifying the resistance mechanism(s) in these species. First, studies on the leaf morphology of these species showed no trichomes on

S. trifidum, indicating that the resistance was not due to presence of specialized glandular trichomes. Components of leaf exudates in some species have been shown to act as insecticidal agents as part of the plant's defense mechanism against insects (Yoshida et al.

1995). A waxy-like substance that was observed on the leaf surface could probably play a role in resistance of S. trifidum to the Colorado potato beetle. Studies were then initiated to determine the types and amounts of foliar glycoalkaloids present in these species.

Glycoalkaloid extraction and evaluation using thin layer chromatography and high performance liquid chromatography were done on these species. Results from the thin layer 59 chromatography analysis for total foliar glycoalkaloids were consistent with earlier reports by the Potato Introduction Station. Solanum trifidum showed low total glycoalkaloids, though in this species resistance could be due to presence of a more potent glycoalkaloid at low concentration.

Another host-plant resistance mechanism against insect pests are proteinase inhibitors.

These prevent proteolytic digestion of plant proteins by proteinases in insect guts. The objectives of the second study, therefore, were to examine the expression of a potato cysteine proteinase inhibitor in Solanum species that are resistant to Colorado potato beetle and to evaluate the relation between cysteine proteinase inhibitor activity and resistance to the

Colorado potato beetle. Proteinase inhibitors generally are low molecular weight proteins that form complexes with proteinases, thereby inactivating these enzymes. In potatoes, the accumulation of proteinase inhibitors is developmentally regulated in growing tubers and environmentally regulated in response to wounding in leaves (Sanchez-Serrano et al. 1986).

Proteinase inhibitor genes are not only activated at the wound site in leaves, but are also systemically activated in non wounded leaves. There are a number of plant proteinase inhibitors that fimction as a defense mechanism of wounded plants (Pena-Cortes et al. 1988).

We examined the potato multicystatin (PMC), a cysteine proteinase inhibitor. Accessions from three species were selected for evaluation of the cysteme proteinase activity. These included S. chacoense (high foliar glycoalkaloid content), S. berthaultii (specialized foliar glandular trichomes), representing the two known mechanisms of resistance, and S. trifidum.

Solanum trifidum demonstrated an eflfective antinutrient and deterrent resistance mechanism 60 despite very low levels of total glycoalkaloids and with no specialized foliar glandular trichomes. Our results showed induction of PMC genes in all three species resistant to

Colorado potato beetle, S. trifidum, S. chacoense, and S. berthaultii. These species showed accumulation of PMC transcripts within 24 h in response to methyl jasmonate treatment in leaf cuttings and on whole plants in response to wounding by beetle feeding. Papain inhibition levels in leaves woimded by beetle feeding, however, did not necessarily correlate with PMC transcript accumulation. Our conclusion from these studies is that PMC induction and activity represents one among several resistance mechanisms against Colorado potato beetle present in S. trifidum.

Solanum trifidum hybridizes with the dihaploid Solanum tuberosum 'Superior' allowing the transfer of the resistance which can then be tested in the cultivated potato.

Results from these studies forms a basis for fiirther investigation of Colorado potato beetle resistance by this species. Other accessions in S. trifidum should be studied for resistance,

PMC and papain inhibition activity. Identification and characterization of any other compounds on the leaf surface or unique glycoalkaloids needs to be done. Results from these fiirther studies may indcate the mechanism responsible for resistance in S. trifidum.

However, it is possible that a number of resistance mechanism are involved. Solanum trifidum and the Colorado potato beetle have co-evolvd from the same region, therefore S. trifidum developed appropriate mechanism to prevent serious beetle damage. 61

LITERATURE CITED Barrett, A. J. 1986. The classes of proteolytic enzymes, pp. 1-16. In M. J. Dalling (ed.). Plant proteolytic enzymes. CRC, Boca Raton, FL. Casagrande, R. A. 1982. Colorado potato beetle resistance in a wild potato, Solcamm berthaultii. J. Econ. Entomol. 75: 368-372. Cheng, J., J. A. Saunders, and S. L. Sinden. 199S. Colorado potato beetle resistant somatic hybrid potato plants produced via protoplast electrofiision. In Vitro Cell. Dev. Biol- 31: 90-95. Christeller, J. J., W. A. Laing, N.P. Markwick, and E.P.J. Burgess 1992. Midgut protease activities in 12 phytophagous Lepidoptera larvae: dietary and protease inhibitor interactions. Insect Biochem. Mol. Biol. 22: 735-746. Deahl, K. L., R. J. Young, and S. L. Sinden. 1 973. A study of the relationship of late blight resistance to glycoalkaloid content in fifteen potato clones. Am. Potato J. 50: 248-253. Dimock, M. B. and W. M. Tingey. 1 985. Resistance in Solanum spp. to the Colorado potato beetle: Mechanisms, genetic resources and potential. Massachusetts Ag. Expt. Sta. Res. Bull. 704: 79-106. Elden, T. C. 1995. Selected proteinase inhibitor effects on alfalfa weevil (Coleoptera: Curculionidae) growth and development. J. Econ. Entomol. 88:1586-1590. Farmer, E. E., R. R. Johnson, and C. A. Ryan. 1992. Regulation of expression of proteinase inhibitor gene by methyl jasmonate and jasmonic acid. Plant Physiol. 98: 995-1002 Feiton G.W., J. L. Bi, C. B. Summers, A. J. Mueller, and S. S. Dufiy. 1994. Potential role of lipoxygenases in defense against insect herbivory. J. Chem. Ecol. 20: 651-

666 62

Ferro, D. N. 1985. Pest status and control strategies of the Colorado potato beetle. Massachusetts Ag. Expt. Sta. Res. Bull. 704:1-8. Ferro, D. N., J. A. Logan, R. H. Voss, and J. S. EUdngton. 1985. Colorado potato beetle (Coleoptera: Chrysomelidae) temperature-dependent growth and feeding rates. Environ. Entomol. 14: 343-48. Franca, F. H. and W. M. Tingey. 1994. Solanum berthaultii Hawkes affects the digestive system, fat body and overies of the Colorado potato beetle. Amer. Pot. J. 71: 405-410 Gatehouse A.M.R., V. A. HUder, and D. Boulter. 1992. Plant genetic manipulation for crop protection. CAB Wallingford, UK. Gibson, R W. 1 971. Glandular hairs providing resistance to aphids in certain wild potato species. Ann. Appl. Biol. 68: 113-119. Gibson, R. W. 1976a. Glandular hairs are a possible means of limiting aphid damage to the potato crop. Ann. Appl. Biol. 82: 143-146. Gibson, R. W. 1976b. Glandular hairs of Solanum polyadeniwn lessen damage by the Colorado beetle. Ann. Appl. Bio. 82: 147-150. Groden, E., and R A. Casagrande. 1986. Population dynamics of the Colorado potato beetle, Leptinotarsa decemlineata (Coleoptera: Chrysomelidae) on Solammt berthaultii. J. Econ. Entomol. 79: 91-97. Hare, J. D. 1990. Ecology and management of the Colorado potato beetle. Annu. Rev. Entomol. 35: 81-100. Hilder, V. A., A.M.R Gatehouse, S. E. Sheennan, R. F. Barker, and D. Boulter. 1987. A novel mechanism of resistance engineered into tobacco. Nature 330: 160-163 63

Hildmann, T., M. Ebneth, H. Pena-Cortes, J. J. Sanchez-Serrano, L Willmitzer, and S. Prat. 1992. General role of abscisic and jasmonic acids in gene activation as a result of meclianicai wounding. Plant Cell 4:1157-1170 Hsiao, T. H. 1988. Host specificity, seasonality, and bionomics of Leptinotcasa beetles. Biology of Chrysomelidae, Dordrecht, Kluwer, pp. 581-599. Horton, D. 1987. in Potatoes, Westview Press, Boulder, CO, pp. 5-20. Hurst, G. W. 1975. Meteorology and the Colorado potato beetle. World Meteorological Organization Tech. Note 137. 51 pp. Johnson, R., J. Navarez, G. An, and C. A. Ryan. 1989. Expression of proteinase inhibitors I and n in transgenic tobacco plants: effects on natural defense agahit Maduca sexta larva. Proc. natn. Acad. Sci. U.S.A 86: 9871-9875. Kumar, A., M. Miller, P. Whitty, J. Lyon, and P. Davie. 1995. Agrobacterium-mod^siQd transformation of five wild Solanum species. Plant Cell Reports 14: 324-328 Murdock L.L., G. Brookhart, F. E. Dunn, D. E. Foard, S. Kelley, L. Kitch, R. E. Shade, R. H. Shukle, and J. L. Wolfson. 1987. Cysteine digestive proteinases in Coleoptera.Comp. Biochem Physiol 87B: 783-787 North M.J. 1982. Comparative biochemistry of the proteinases of eucaryotic microorganisms. Microbiol. Rev. 46: 308-340. Oppert, B., T. D. Morgan, C. Culberston, and K- J. Kramer 1993. Dietary mixtures of developing tubers. Planta 186: 495-502 Pena-Cortes, H., J. Sanchez-Serrano, M. Rocha-Sosa, and L. Winmitzer 1988. Systemic induction of proteinase inhibitor II gene expression in potato plants by wounding. Planta 174: 84-89 Pena-Cortes, H., X. Liu, J. Sanchez-Serrano, R. Schmidt, and L. Willmitzer. 1992. Factors affecting gene expression of patatin and proteinase-inhibitor II gene 64

families in detached potato leaves: implications for their co-expression in developing tubers. Planta 186: 495-502 Rotino, J., D. Perrone, P. Ajmone-Marsen and E. Lupotto. 1992. Transformation of Solanum integrifolium poir to Agrobacteriim tumefaciens-. plant regeneration and progeny analysis. Plant Cell Reports 11: 11-15 Ryan, J. D., P. Gregory and W. M. Tingey. 1982. Phenolic oxidase activities in glandular trichomes of berthcmltii. Phytochem. 21: 1885-1887. Ryan, C. A. 1992. The search for the proteinase inhibitor-inducing factor, PHP. Plant Mol. Biol. 19:123-133 Sanchez-Serrano J. R., J. Schmidt, J. Schell, and L. Willmitzer 1986. Nucleotide sequence of proteinase inhibitor n encoding cDNA of potato (Solamim tuberosum) and its mode of expression. Mol. Gen. Genet. 203:15-20 Sanford, L. L. and S. L. Sinden. 1972. Inheritance of potato glycoalkaloids. Am. Potato J. 49: 209-217. Sanford, L. L., K. L. Deahl and S. L. Sinden. 1994. Glycoalkaloid content in foliage of hybrid and backcross populations from a Solanum tuberosum X S. Chacoense cross. Am. Potato J. 71 : 225 - 235. Sanford, L. L., K. L. Deahi, S. L. Sinden and R. S. Kobayashi. 1995. Glycoalkaloid content in tubers of hybrid and backcross populations from a Solcmum tuberosum X S. Chacoense cross. Am. Potato J. 72 : 261 - 271. Sembdner, G. and B. Partlier, 1993. The biochemistry and the physiological and molecular actions of jasmonates. A. Rev. Physiol. Plant Molec. Biol. 44: 569-589. Schreiber, K. 1958. Concerning several chemical components of the Solanaceae and their importance in resistance to the Colorado potato beetle. Entomol. Exp. Appl.1: 28-37. 65

Sinden, S. L., Sanford, L. L., Cantelo, W. W., & Deahl, K. L. 1986. Leptine glycoalkaloids and resistance to the Colorado potato beetle in Solomon chacoense. Env. Entomol. 15:1057-1062. Sinden, S. L., L. L. Sanford, and R. E. Webb, 1984. Genetic and environmental control of potato glycoalkaloids. Am. Potato J. 61: 141-156. Sinden, S. L. and R. E. Webb, 1972. Effect of variety and location on the glycoalkaloid content of potatoes. Am. Potato J. 49:334-338. Thie N.M.R-, and J. G. Houseman 1990. Identification of Cathepsin B, D and H in the larval midgut of Colorado potato beetle, Leptinotarsa decemlineata Say (Coleoptera: Chrysomelidae). Insect Biochem. 20: 313-318 Tingey, W. M. and G. C. Yencho. 1994. Insect resistance in potato: A decade of progress. Advances in potato pest biology and management. APS Press, St. Paul, Minnesota. Vick, B. A., and Zimmennan, D. C. 1984. Biosynthesis of jasmonic acid by several species.

Plant Physiol. 75: 458-461

Wright, R J., M. B. Dimock, W. M. Tingey, and R. L. Plaisted, 1985. Colorado

potato beetle (Coleoptera: Chrysomelidae): expression of resistance in Solanum

berthaultii and interspecific potato hybrids. J. Econ. Entomol. 78: 576-582.

Wolfson, J. L. and L.L. Murdock. 1987. Method for applying chemicals to leaf surfaces for

bioassays with herbivorous insects. J. Econ. Entomol. 44: 235-240.

Yoshida, M., S. E. Cowgill and J. A. Wightman. 1995. Mechanism of resistance to

Helicoverpa armigera (Lepidoptera: Noctuidae) in chickpea: Role of oxalic acid in

leaf exudate as an antibiotic factor. J. Econ. Entomol. 88: 1783-1786. 66

ACKNOWLEDGMENTS I wish to express my sincere appreciation and gratitude to my major professor. Dr. David Piannapel for the guidance, encouragement, continuous support, and patience during my research. I would also like to thank the members of my committee, Drs. Amel Hailauer, John Obrycki, Loren Stephens and Richard dadon for their time, patience and constructive suggestions in my program. Special thanks go to Dr. John Obrycki for the thorough review and editing of my journal paper manuscripts, advice and supply of the Colorado potato beetles. I thank Joseph Munyaneza and his colleagues. Department of Entomology at Iowa State University, for the cooperation and assistance given in my work. Thanks go to Mary Tymeson for all the loving care and assistance she gave to all of us in Dr. Hannapel's Laboratory. It was great working with you. I received assistance from many individuals, however, special thanks go to William Gonzalez, Angela Tedesco, Tony Aiello, Dr. Sangu- Kang, and Annie Liu their friendship. Finally, I wish to thank the Department Chair, Dr. Michael Chaplin for his personal assistance. Most of all, I would like to thank my wife, Theresa and children, Consolata and Luke for their patience and support during my studies. To my mother, brothers and sisters and their families for their prayers, support and assistance while away. Thanks go to the US Agency for International Development- Kenya National Agricultural Research Project who made it possible for my studies at Iowa State University.