BROADENING the GENETIC BASE of PIGEONPEA by the UTILIZATION of Cajanus SPECIES from SECONDARY and TERTIARY GENE POOLS a THESIS Submitted

BROADENING THE GENETIC BASE OF PIGEONPEA BY THE UTILIZATION OF Cajanus SPECIES FROM SECONDARY AND TERTIARY GENE POOLS A THESIS Submitted

for the award of the Degree of

DOCTOR OF PHILOSOPHY in BOTANY

Submitted by

Sandhya Rani Vanam

Department of Plant Science Center of Excellence in Life Sciences Bharathidasan University Tiruchirappalli - 620 024 Tamil Nadu, India 2014

BROADENING THE GENETIC BASE OF PIGEONPEA BY THE UTILIZATION OF Cajanus SPECIES FROM SECONDARY AND TERTIARY GENE POOLS

Thesis submitted to Bharathidasan University for the award of the Degree of

DOCTOR OF PHILOSOPHY in BOTANY

SANDHYA RANI VANAM (Ref.14396/PhD1/Botany/ Oct 2010/ FT/Confirmation/ Date.16.10.2012) Under the Guidance of Dr. M.V. Rao Professor Department of Plant Science Bharathidasan University, Tiruchirappalli, Tamil Nadu

& Under the Co-guidance of Dr. Nalini Mallikarjuna Principal Scientist-Biotechnology International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Patancheru 502 324, Andhra Pradesh

DEPARTMENT OF PLANT SCIENCE SCHOOL OF LIFE SCIENCES BHARATHIDASAN UNIVERSITY TIRUCHIRAPPALLI – 620024 TAMIL NADU, INDIA

JUNE 2014

This thesis is dedicated to:

My Lord and Savior Jesus Christ, who brought me into the light and glory of salvation through His example, sacrifice, death, and resurrection, and showed me the meaning of love. First of all I would like to thank my LORD Jesus Christ for not only making the lovely area, in which I could do this study, but also for helping me, and enabling me to carry out this study. All praise, honour and glory to my Lord Jesus Christ for His richest grace and mercy for the accomplishment of this thesis. Without His strengthening and encouragement, this thesis would not exist.

CERTIFICATE-I

This is to certify that the Ph.D. thesis entitled “Broadening the genetic base of pigeonpea by the utilization of Cajanus species from secondary and tertiary gene pools” is the record of bonafide research carried out in the Department of Plant Science, Bharathidasan University, Tiruchirappalli, by Mrs. Sandhy Rani Vanam, under my guidance and also co-guidance of Dr. Nalini Mallikarjuna, Principal Scientist- Biotechnology, International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Patancheru 502 324, Andhra Pradesh. This is submitted for the award of DOCTOR OF PHILOSOPHY in Botany of Bharathidasan University, Tiruchirappalli, Tamil Nadu.

I further certify that the research work is original and the data presented in the thesis are based on her own observations and no portion thereof has been submitted elsewhere in part or full for any other degree, diploma, associateship of fellowship of any other university.

(M.V.RAO) RESEARCH SUPERVISOR

CERTIFICATE-II

This is to certify that the thesis entitled “Broadening the genetic base of pigeonpea by the utilization of Cajanus species from secondary and tertiary gene pools.” that is being submitted by Ms. Vanam Sandhya Rani in partial fulfillment for the award of Ph.D. in Department of Botany to the Bharathidasan University Tiruchirappalli - 620 024, Tamil Nadu is a record of bonafide work carried out by her under our guidance and supervision. The results embodied in this thesis have not been submitted to any other University or Institute for the award of any degree or diploma.

Signature of Co-Supervisor Signature of Supervisor

Dr. Nalini Mallikarjuna Dr. M.V. Rao Principal Scientist Professor and Head Legume Cell Biology Department of Plant Science Grain Legumes Bharathidasan University ICRISAT Tiruchirappalli - 620 024 Patancheru - 302324 Tamil Nadu, India Hyderabad, India

CERTIFICATE-III

This is to certify that the thesis entitled “Broadening the genetic base of pigeonpea by the utilization of Cajanus species from secondary and tertiary gene pools” that is being submitted by Ms. Sandhya Rani Vanam in partial fulfillment for the award of Ph.D. in the Department of Botany to the Bharathidasan University, Tiruchirappalli - 620 024, Tamil Nadu is a record of bonafide work carried out by her at our organization/institution.

Signature of Head/Director of Organization/Institution Name and Designation Dr. Rajeev K. Varshney, Ph. D., FNAAS Research Program Director- Grain Legumes Director- Center of Excellence in Genomics (CEG) ICRISAT, Patancheru-502 324 Email: [email protected] Tel: 040-30713305

DECLARATION

I hereby declare that the work embodied in this thesis has been originally carried out by me under the guidance of Dr. M.V. Rao, Professor and Head, Department of Plant Science, Bharathidasan University, Tiruchirappalli- 620024, Tamil Nadu, India and also co-guidance of Dr. Nalini Mallikarjuna, Principal Scientist-Biotechnology, International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Patancheru 502 324, Andhra Pradesh, India. I further assure that this work has not been submitted either in whole or part for any other degree or diploma at any other university.

Place: Tiruchirappalli Signature of Candidate

Date: (Sandhya Rani Vanam)

Acknowledgements

Endless compassion of Almighty turns my difficult task to a feeling of pleasant journey of my life. Emotions cannot be adequately expressed in words because emotions are transformed into a mere formality. Hence, my acknowledgements are many times more than what I am expressing here.

I have received a lot of ad-hoc response from many generous grandees in many quarters to those pain stacking days encompassing the research work. I wish to vow my genuflexion with deep sense of gratitude to my co-supervisor Dr. Nalini Mallikarjuna, Principal Scientist, Department of Legume Cell Biology, ICRISAT, for her conscientious guidance, gracious, cordial and meticulous explications and encapsilative remarks towards the representation of this dissertation. Her valuable suggestions brought a panacea for me dealing with this research work. I can truly say that whatever she has done for me is not possible for every advisor to do for their advisee.

At this stage I feel pleasure to express my profound regards, indebtness and gratitude to my supervisor Dr. M. V. Rao, Professor and Head, Department of Plant Science, , Bharathidasan University, Tiruchirappalli - 620 024 Tamil Nadu, for his expedient advice, debonair discussion, innovative ideas, abiding interest and invaluable support during this tenure of research work.

This dissertation would not have been possible without the guidance; moral support and the kind help of Dr. Rajeev K Varshney, Research Program Director, Grain Legumes, and Director- Center of Excellence in Genomics (CEG). I am sincerely obliged and indebted for his peer guidance, endless support, encouragement and able direction during last 6 months of my research project.

With respect, regards and immense pleasure, I wish to acknowledge and express sincere thanks from bottom of my heart to several scientists including Dr. K.P.M.S.V. Padmasree, Associate Professor, Dept. of Biotechnology and Bioinformatics, School of Life Science, University of Hyderabad, Dr. Rachit K Saxena, Scientist, Aplied Genomics, ICRISAT and Dr. Riyazul rouf Mir, Scientist, ICRISAT for providing lab facilities to complete some of my research works with their constructive comments, valuable suggestions, immense help and enthusiastic discussions. I wish to express my sincere thanks for their critical evaluation and emending suggestions in the preparation of scientific papers for journals and in achieving the final form of dissertation.

I am heartily thankful to Dr.V.M. Muthukumar, Hon. Vice-chancellor and Dr. E. Ramganesh, Registrar, Bhrathidasan University, Tiruchirappalli for providing facilities and resources during my Ph.D. research programme. I wish to thank Dr. N.Jayabalan- Professor, Dr. M.B.Viswanathan-Professor, Dr. B.D.Ranjitha Kumari- Professor, Dr. A.Lakshmi Prabha- Asst Professor, Dr. M.sathiyabama- Asst Professor and Dr. S.R. Sivakumar-Asst Professor, Dept. of Plant Science, Bharathidasan University.

A depth of gratitude is owed to Dr. William Dollente Dar, Director General and Dr. C.L.L Gowda, Deputy Director General of the ICRISAT for giving me opportunity to work in one of the best labs of International standards. I am thankful to Dr. Shivali Sharma, scientist, Wide hybridization Lab for her administrative help during last few months my stay at ICRISAT. I wish to express my thanks to Dr. Rosana P Mula, the LSU Co-ordinator and other staff of LSU for providing administrative help and Mr. Madan M, Mr. Gautham T.L and other staff for providingthe Library facilities. My sincere thanks to Mr. Nagaraju T from Reprographic unit for his all time support, kind help and encouragement during my stay at ICRISAT. I would like to thank Mr. Sk Meeravali and Mr. MNR Ramesh, Strategic Marketing & Communication for their timely help in preparing Posters.

It is my immense pleasure to render sincere regards to Dr. Deepak Jadhav, scientific officer, Department of Legume Cell Biology for his kind support and encouragement and guiding me to complete a small portion of my Ph. D research. I render special thanks to Mr. Balakrishna, Mr. Satyanaraya, Mr. Gafoor (AGL), Mr. Avinash, Ms. Amrutha and Ms. Lakshmi for providing technical support. I place on record my heartfelt thanks to my labmates and friends Ms. Revathi Lakshmi S, Mr. Parthibhan S, Mr. Ahamed Sherif N, Ms. Senbagalakshmi P, Mr. Muthukumar M, Mr. Muthukrishnan S, Mr. Thaniarasu R, Mr. Baradwaj RG, Ms. Pushpavalli R, Mr. Santosh Kumar VV and Mr. Rakesh Kumar Verma Dr. Lekha P T, Ms. Sameera P, Ms. Krishna Shilpa , Ms. Swath P, Ms Swathi M, Mr. Mohan raj, Mr. Lokya, Mr. Srinivas and Mr. Abhay.

With immense pleasure to express my sincere thanks to Dr. Suri Seghal, Dr. M. D Gupta, Dr. Hanumantharaya and P. Vani Sekhar for providing their support and encouragement during my stay at ICRISAT. Also, Suri Seghal foundation is gratefully acknowledged for providing me the financial support during my Ph D.

I feel blessed to have my husband Mr. T. Srikanth in my life, who was the constant inspiration for me to carry out the research and gave emotional support whenever I had difficult times. I feel happy to have my husband who have traveled every step of this journey by cheering me up and stood by me through the good and hard times. I feel indebted to my mother-in-law Smt. K. Bhagya Rekha for her kind support and encouragement throughout the tenure.

I avail this opportunity to thank my beloved parents who are the Almighty’s most treasured gift to me. I feel scanty of words to magnitude the boundless love and tireless sacrifice, affection and encouragement showed on me by my parents, Shri. V. Nageswar Rao and Smt. V. Shakunthala and affection of my sisters that I could attain academic heights to accomplish my doctoral degree. And I express my deepest adoration to them for their support at every crucial moment of my life.

I convey my whole hearted thanks to many of my well wishers and friends.

Tiruchirappalli June, 2014. (Sandhya Srikanth)

CONTENTS

Chapter Title Page No. No. 1 INTRODUCTION 1-9 2 REVIEW OF LITERATURE 10-51 3 MATERIALS AND METHODS 52-71 4 RESULTS AND DISCUSSION 72-220 5 TABLES 126-154 6 FIGURES 155-220 7 SUMMARY AND CONCLUSION 221-223 8 REFERENCES i-xxxvii

LIST OF TABLES

Table Title Page No. No. 1 Interspecific hybridization in Cajanus species 18 2 Morphological traits of F3 BC 1 population of the cross C. cajan 126 (ICPL87119) X C. cajanifolius (ICPW29) 3 Fertility status of F1 hybrids derived from the cross C. cajan (ICPL 127 85010) X C. lanceolatus (ICP15639) and number of F2/BC1 seeds obtained from hybrids 4 Morphological traits of F1 hybrids derived from the cross C. cajan 128 (ICPL 85010) X C. lanceolatus (ICP15639) 5 Percentage of pod set in sterile F1 hybrids (P-1, P-4, P-7, P-10 and 129 P-12) derived from C. lanceolatus when crossed with different unrelated pigeonpea cultivars (ICPL 85010, ICPL 88039, ICPL 88034, ICPL 85030, MN1, MN5, MN8 and ICPL 92016) 6 Morphological traits of F1 BC1 [(ICPL 85010 X ICP15639) X ICPL 130 85010] and F2 lines from the cross C. cajan (ICPL 85010) X C. lanceolatus (ICP15639) 7 Morphological traits of F1 BC3 [(ICPW 68 X ICPL 85010) X ICPL 131 85010 X ICPL 85010 X ICPL 85010] lines derived from C. platycarpus 8 Number of pods and viable seeds per plant recorded in F1 BC3 133 [(ICPW 68 X ICPL 85010) X ICPL 85010 X ICPL 85010 X ICPL 85010] lines derived from the C. platycarpus. 9 Morphological traits of advanced generation lines derived from the 135 cross C. platycarpus (ICPW 68) X C. cajan (ICPL 85010) 10 Morphological traits of F2 population derived from the cross C. 136 cajan (ICPL 85010) X C. volubilis (ICP15774) 11 Morphological traitsof F4 population of the cross C. cajan X C. 138 volubilis., S. DET-Semi-determinate and DT- Determinate 12 Meiotic analysis of abnormal individual plants of F3BC1 lines 140 derived from the C. cajanifolius indicates more homology between two parental chromosomes involved in the cross. 13 Meiotic studies of hybrids derived from the cross C. cajan (ICPL 141 85010) X C. lanceolatus (ICP 15639) indicates more homology between two parental chromosomes involved in the cross. 14 Segregation of Male sterility in F1 BC1 [(ICPL 85010 X ICP 15639) 142 X ICPL 85010] plants derived from the crosses between three sterile F1s and pigeonpea cultivars.F:Fertile, S:Sterile 15 Meiotic analysis of F1 BC3 [(ICPW 68 X ICPL 85010) X ICPL 143 85010 X ICPL 85010 X ICPL 85010)] lines derived from C. platycarpus 16 Polymorphism status of SSR markers tested on eight F1 hybrids 145 derived from the cross C. cajan (ICPL85010) X C. lanceolatus (ICP15639) and two parental genotypes. 17 Genotype scoring of 14 polymorphic primers among eight F1 146 hybrids & two parental genotypes of the cross C. cajan X C. lanceolatus and percentage of hybrid purity index for each marker 18 Percentage of hybrid purity in F1 individuals of the cross C. cajan X 147 C. lanceolatus. 19 Polymorphism status of the Primers screened to test the F1 hybrid 147 derived from the cross C. cajan (ICPL 85010) X C. volubilis (ICP 15774). 20 Percentage of pod damage by Helicoverpa armigera in the infested 148 field correlated with the Helicoverpa mid gut trypsin inhibitors 21 Percentage of pod damage in F3 BC1 lines of C. cajanifolius 149 22 Percentage of pod damage in interspecific derivatives of C. 152 lanceolatus 23 Percentage of pod damage in advanced generation lines of C. 153 platycarpus 24 Bruchid damage in the F1 hybrids of the cross C. cajan X C. 154 lanceolatus and their parents (ICPL 85010 and ICP 15639)

LIST OF FIGURES

Table Page Title No. No. (A) Male parent ICPW 29 (C. cajanifolius), (B) female parent ICPL 155 1 87119 (C. cajan). 2 (A) Variation in plant morphology in progeny lines of F3 BC176-2. (B) 156 Variation in the leaf size and shape of F3 BC137-1 and (C) Uniformity in the morphology of progeny lines of F3 BC1129-1. 3 (1) Small sized flowers of 82-2-7 F3 BC1. (2) Purple and green mixed 157 small pods with prominent locules of 76-2-3 and (3) Green and purple coloured pods in the different plants of 130-1 F3 BC1 (single row) line. 4 Morphology of leaves in the F3 BC1 population of the cross C. cajan X 157 C. cajanifolius. 5 (A) F3 BC1 plantation of C. cajanifolius derivatives (B) interspecific 158 derivatives in the field. 6 Female parent (ICPL 85010), Male parent (ICP 15639) and tallest F1 159 hybrid P-13 measuring about 380cm. 7 Morphological observations of hybrids (ICPL 85010 X ICPW15639) 160 and parents (cultivar- ICPL 85010 and ICPW15639). 8 (A) Morphology of pods and seeds in parents (cv. ICPL 85010- left and 161 ICP 15639 - right), (B) hybrid (middle) plants. 9 Variation in plant height of C. lanceolatus derivatives. 161 10 Variation in the leaf size and shape of the progeny lines derived from 162 the cross C. cajan X C. lanceolatus. 11 Receme morphology in F1 BC1 population of C. cajan X C. 162 lanceolatus.

12 F1 BC1 plants segregated for flower colour and pattern of streaks on 163 standard petal. 13 Variation in the pod colour and shape in the F2 and F1 BC1 pods. 163 14 F1 BC1 [(ICPL 85010 X ICP15639) X ICPL 85010] and F2 progeny 164 plantation in the field. 15 Female parent C. platycarpus ICPW 68 (left) and male parent ICPL 164 85010 (right). 16 Morphology of the F1BC3 plants derived from C. platycarpus. 165 17 Variation in Anther bundles and pollen load in F1BC3 lines derived 166 from C. platycarpus. 18 Pods obtained by cross and self pollinations of F1BC3 lines derived 167 from C. platycarpus. 19 Seeds obtained from F1 BC3 lines derived from C. platycarpus were 168 compared with cultivated parent ICPL 85010. 20 Morphology of F1 BC3 B6 plant and their pods. 169 21 Variation in leaf let shape and size in the C. platycarpus derivatives. 169 22 Morphology of F1 BC5 lines of the cross C. platycarpus X C. cajan in 170 the field. 23 Morphology of parents involved in the wide cross. 171 24 Frequency polygon graphs showing Days to first flowering (A), Plant 172 height (B), Days to 50% pod maturity (C) and Number of pods/plant (D). 25 Morphology of hybrids in C. volubilis derivatives 173 26 Colour and pattern flowering in F4 population of C. cajan X C. 173 volubilis. 27 Comparisons of F2 and F4 hybrid plants derived from the C. volubilis 174 with different cultivars. 28 F4 plantation of the C. volubilis derivatives in the glass house. 175 29 Meiotic analysis of F1 hybrid derived from C. cajan (ICPL 87119 ) X 176 C. cajanifolius (ICPW 29). 30 Meiosis in BC1 F3 73-1-3 (A-E), 76-2-1(F-H) and 82-2-2 (I-K). 177 31 Summary charts (stacked columns) for the meiotic analysis of C. 178 lanceolatus F1 hybrids. 32 Meiotic analysis in F1 hybrids derived from C. lanceolatus. 179 33 Meiosis in F1 hybrids P4, P5, P6 and P7. 180 34 Meiosis in fertile hybrid F1 P8. 180 35 Meiotic analysis in BC3 F1 A lines derived from C. platycarpus. 181 36 Meiotic analysis in BC3 F1 B lines derived from C. platycarpus. 182 37 Meiotic analysis in BC3 F1 C and D lines derived from C. platycarpus. 183 38 Meiotic analysis in BC3 F1 E lines derived from C. platycarpus 184 39 Pollen fertility in F1 BC3 A, F1 BC3 B, F1 BC3 C, F1 BC3 D and F1 BC3 186 E lines derived from C. platycarpus 40 Anther morphology in F1 BC3 A, F1 BC3 B and F1 BC3 C lines derived 187 from C. platycarpus. 41 Anther morphology in cv.ICPL 85010, F1 BC3 D and F1 BC3 E lines 188 derived from C. platycarpus. 42 Meiotic behavior in F1 hybrid derived from C. cajanx C. volubilis. 189 43 Abnormal meiosis in F4 P1-6-9 derived from C. volubilis. 190 44 Abnormal meiosis in F4 P1-6-9 derived from C. volubilis. 191 45 Meiosis in F4P1-3-8 derived from C. volubilis. 192 46 Meiosis in (ICPL 85010 X ICP15639) P12 F1 X MN1 male sterile 192 hybrid 47 Microtomy sections of flower buds from Fertile ((ICPL 85010 X 193 ICP15639) P-12 X MN1) and sterile ((ICPL 85010 X ICP15639) P-12 X MN1) plants. 48 Primer CcM0047, Electropherogram of SSRs obtained with software 194 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids 49 Primer CcM 0057, Electropherogram of SSRs obtained with software 195 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 50 Primer CcM 0710, Electropherogram of SSRs obtained with software 196 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 51 Primer CcM 0974, Electropherogram of SSRs obtained with software 197 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 52 Primer CcM 1459, Electropherogram of SSRs obtained with software 198 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 53 Primer CcM 1991, Electropherogram of SSRs obtained with software 199 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 54 Primer CcM 2012, Electropherogram of SSRs obtained with software 200 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 55 Electropherogram of SSRs obtained with software Genemapper. Primer 201 CcM 2066 yielded both parental amplicons in F1 hybrids. 56 Primer CcM 2071, Electropherogram of SSRs obtained with software 202 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 57 Primer CcM 2176, Electropherogram of SSRs obtained with software 203 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 58 Primer CcM 2228Electropherogram of SSRs obtained with software 204 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 59 Primer CcM 2505, Electropherogram of SSRs obtained with software 205 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 60 Primer CcM 2639, Electropherogram of SSRs obtained with software 206 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids 61 Primer CcM 2855, Electropherogram of SSRs obtained with software 207 Genemapper. Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids. 62 Electropherogram of SSRs obtained with software Genemapper. 208 Hybridity confirmed with CcM 0008. Female parent ICPL 85010 (first line), Male parent ICP15774 (middle line) and F1 hybrid (third line). 63 Electropherogram of SSRs obtained with software Genemapper. 209 Hybridity confirmed with CcM 0008. Female parent ICPL 85010 (first line), Male parent ICP15774 (second line) and then F2 hybrids. 64 Inhibition of bovine pancreatic trypsin by crude extracts of proteinase 210 inhibitors from the parents (cultivated and wild) and their derivatives of pigeonpea. 65 Inhibition of bovine pancreatic chymotrypsin by crude extracts of 211 proteinase inhibitors from the parents (cultivated and wild) and their derivatives of pigeonpea. 66 Inhibition of midgut trypsin- like proteinases of Helicoverpa armigera 212 by crude extracts of proteinase inhibitors from parents (cultivated and wild) and their derivatives of pigeonpea. 67 Inhibition profiles of parents (wild and cultivated) and their derivatives 213 of pigeonpea against pancreatic trypsin (A), chymotrypsin (B) and in presence of Helicoverpa armigera midgut proteinases (C). 68 Effect of proteinase inhibitors from parents (cultivated and wild) and 214 their derivatives of pigeonpea on Human pancreatic trypsin. 69 Interspecific derivatives at flowering stage in natural pod borer infested 215 field of ICRISAT. 70 (A) F3 BC3 C7-13-2 (derived from C. platycarpus) plant with healthy 216 pods. (B) Healthy and damage pods from the same plant 71 Healthy and Helicoverpa damage pods in C. cajanifolius derivatives. 217 72 Healthy and Helicoverpa damage pods in C. lanceolatus derivatives. 217 73 Healthy and Helicoverpa damage pods in C. platycarpus derivatives. 218 74 Bruchid damage in F1 hybrids seeds of C. cajan x C. lanceolatus 219 (showing low egg) laying. 75 Expression of IDT/SDT or DT associated T/A specific alleles in 220 CcTFL1 76 Phenotypic variation in flowering pattern (SDT or DT). 220

ABBREVIATIONS

% Per cent / Per > More than < Less than 0C Degree celsius cm Centimeter et al. Et alia (and others) etc. Etcetera ed. Edited by Fig. Figure g Gram kg ha-1 Kilogram per hectare ha Hectare hrs. Hours i.e. That is Kg Kilogram(s) M Metre(s) No./no. Number GA3 Gibberellic acid CMS Cytoplasimic Male Sterility GMS Genomic Male Sterility PIs Proteinase inhibitor(s) ml milliliter mg/L Milligrams per Liter μl Microliter FAA Formaldehyde: Glacial acetic acid: alcohol TBA Tert-Butyl Alcohol BAPNA N--benzoyl-DL arginine-p-nitroanilide GLUPHEPA N-Glutaryl-L-pheny-alanine p-nitroanilide BBI Bowman-Birk inhibitor SKTI Soybean Kunitz trypsin inhibitor CBB Coomassie Brilliant Blue TEMED N,N,N',N'-Tetramethylethylenediamine TI Trypsin inhibitor CI Chymotrypsin inhibitor HGPI Helicoverpa mid-Gut trypsin like proteinase inhibitors HPTI Human pancreatic trypsin inhibitors bp base pair CTAB Cetyl Trimethyl Ammonium Bromide DNA Deoxyribonucleic Acid SSR Simple Sequence Repeats SNP Single Nucleotide Polymorphism Mb Million bases PAGE Polyacrylamide Gel Electrophoresis PCR Polymerase Chain Reaction ICRISAT International Crops research Institute for the Semi-Arid Tropics ICPL ICRISAT Pigeonpa line ICPW ICRISAT pigeonpea wild

ABSTRACT

It is known that pigeonpea rests on a narrow genetic base. One of the ways to broaden the genetic base of pigeonpea is by utilization of wild relatives of pigeonpea in wide crosses. Wild relatives are important sources of genetic and agronomic traits, including resistance to various biotic and abiotic stresses, yield and seed quality. The present investigation was initiated with a broad objective of “Broadening the genetic base of pigeonpea by utilization of Cajanus species from secondary and tertiary gene pool”. The present study involved four wide crosses namely Cajanus cajan (ICPL 87119) X C. cajanifolius (ICPW 29), C. cajan (ICPL 85010) X C. lanceolatus (ICP 15639), C. platycarpus (ICPW 68) X C. cajan (ICPL 85010) and C. cajan (ICPL 85010) X C. volubilis (ICP 15774). Of these wide crosses, C. cajnifolius and C. lanceolatus were secondary gene pool species while C. platycarpus and C. volubilis were tertiary gene pool species.

In the cross C. cajan X C. cajnifolius successful pod set was obtained in direct crosses involving C. cajan as ovule parent and C. cajanifolius as pollen parent. F1 hybrids from the cross ICPL87119 × C. cajanifolius was vigorous in growth and exhibited intermediate leaf and flower morphology with predominance of C. cajanifolius plant type.

Chromosome pairing was regular in the F1 hybrid with a few abnormalities such as univalents, laggards, stickiness, bridges, precocious separation. Normal chromosome pairing coupled with seed set suggested the possibility of incorporating traits of economic importance such as insect resistance from C. cajanifolius into C. cajan.

In the present study, a successful wide-cross between Cajanus lanceolatus, a wild relative from secondary gene pool, native to Australia, with desirable traits such as frost and drought resistance, and cultivated pigeonpea is reported. A range of F1 progeny were obtained and the resultant F1 hybrids set mature pods/seeds. Hybrids had intermediate morphology sharing traits of both the parents. All the F1 hybrids flowered profusely. Some of the hybrids were completely male sterile and some were partially fertile with the pollen fertility ranging from 35 % to 50 %. Meiotic analysis of the fertile F1 hybrids revealed high degree of meiotic chromosome pairing between the two parental genomes.

Meiotic analysis of the sterile F1 hybrids revealed that the break-down in the microsporogenesis was at the post meiotic stage after the formation of tetrads. Fertile plants formed regular bivalents with normal disjunction except for occasional asynchrony at meiotic II division. The present study reports a second source of CMS developed by using the cultivated pigeonpea as the female parent and one of its wild relative C. lanceolatus as the pollen donor, similar to the A5 CMS system derived from C. acutifolius. All the F1 hybrids were evaluated to confirm hybridity using 27 Simple sequence repeats (SSR) markers. SSR marker analysis of parents that were involved in the generation of the F1 hybrids have shown that 9 out of 27 SSRs were polymorphic among the parents and these were used to confirm the hybridity of the F1 plants. Sterile F1 hybrid plants were crossed with a range of pigeonpea cultivars to identify maintainers of male sterility. This study besides summarizing the morphology of the F1 and backcross generations (BC1), reports cytology of the sterile as well as fertile floral buds derived from the crosses between sterile F1 hybrids and pigeonpea cultivars. An important observation made was that male sterility was a post meiotic process. Microsporogenesis was normal until the tetrad stage, but none of them formed pollen grains. Instead, they grouped together within the pollen mother cell (PMC) wall and the tetrads did not separate into individual pollen grains. Also, few F1 hybrids such as P1, P2, P4, P5, P6, P7 and P8 exhibited good levels of resistance against Callosobruchus maculates.

Cajanus platycarpus (Benth) van der Maesen, a wild relative from the tertiary gene pool of pigeonpea, has many desirable traits. Hence an important wild relative for broadening the genetic base and introduction of useful traits into cultivated pigeonpea. Stable F1BC2 A, B, C, D and E lines which were not screened for any trait of interest were taken up for the present study. Morphological observations and cytological analyses were carried out for F1BC3 A, B, C, D, and E lines as well as their subsequent backcrossing generations

(F1BC4 and F1BC5) and selfing generations (BC3F2 and BC3F3).

Cajanus volubilis (Blanco) belongs to the tertiary gene pool of pigeonpea. There are no reports of using C. volubilis as the male parent in the crosses with pigeonpea. This study details the successful cross between pigeonpea and C. volubilis. Cajanus volubilis has a climbing growth habit, and pigeonpea has erect growth habit and both the parents have a raceme inflorescence. F1 hybrid had erect growth habit and was shorter than the female parent with the flowering inflorescence in the form of a raceme. When F2 population was grown in kariff season of 2011, all of them were short stature and with determinate flowering inflorescence, unlike the one observed in both the parental plants as well as in the F1 hybrid. F2 plants flowered earlier than the female parent and matured early by 85 days. More number of F2 plants were raised in the kariff season of 2012. All the plants were short statured with determinate flowering inflorescence. Short duration is a desirable trait in pigeonpea improvement and there is an emphasis to look for this trait while breeding for pigeonpea especially for higher altitudes and latitudes. Although early flowering with short stature are available in cultivated pigeonpea germplasm, the source presently being reported is derived from C. volubilis, a tertiary gene pool species and hence these lines had a different genetic background. Molecular analysis of the F1 hybrid and few plants in F2 population were confirmed the hybridity of the derivatives with band patterns resembling that of the male and female parents. There was drastic change in the morphology of the F2 plants, in being dwarfs in comparison to the F1 hybrid, and both the parental plants. They were compact and were segregated for determinate (DT) or semi- determinate (SDT) growth habits with terminal clusters inflorescences resulting pods in bunches. The expression of DT/SDT growth habit was validated using allele (A/T) specific TERMINAL FLOWER 1(TFL1) primers.

A range of interspecific derivatives derived from C. lanceolatus, C. cajanifolius, C. volubilis and C. platycarpus along with their cultivated and wild pigeonpea parents were screened for the pod borer resistance under unprotected field conditions at ICRISAT. Biochemical basis of resistance was also identified by studying the levels of defense proteins i.e. proteinase inhibitors (PIs) active against bovine pancreatic trypsin, chymotrypsin as controls and trypsin-like enzymes of Helicoverpa armigera mid-gut proteinases. PI profiles of parents (both wild and cultivated) and interspecific derivatives differed in terms of activity units, number and intensities of activity bands visualized on gelatine activity-PAGE. High level of Helicoverpa gut proteinase inhibitor (HGPI) units were further screened for Human pancreatic trypsin inhibitory (HPTI) activity levels. Samples with high ratio of HGPI/HPTI represented less or no effect against Human pancreatic trypsin and high effect on Helicoverpa insect gut proteinases.

INTRODUCTION

Introduction

Pigeonpea [Cajanus cajan (L.) Mlilspaugh], also known as red gram, is the 6th most important pulse crop of the tropics and sub tropics. It is cultivated in about 50 countries in Asia, Africa and the Americas. Globally, the pigeonpea is cultivated on 5,2,00,000ha of land with an annual production of 3.35million tons and productivity of 780kg ha-1 (www.fao.org) and 77% of its area is in India. It is followed by Myanmar (6,20,000ha) and China (1,50,000ha). In sub-Sahara Africa (Kenya, Malawi, Tanzania, Uganda and Mozambique) long duration pigeonpea constitute an important component of rain-fed agriculture. Considering the vast natural genetic variety of cultivated pigeonpea and presence of its wild relatives in India, it is believed that pigeonpea originated in India and is concluded as primary center of origin (van der Maesen, 1980).

Pigeonpea is widely grown in the Indian subcontinent which accounts for almost 90 per cent of the world's crop (Nene and Sheila, 1990). It is grown in almost all states, but the major concentration is in Maharashtra (11,00,000ha), Karnataka (5,80,000ha), Andhra Pradesh (5,10,000ha), Uttar Pradesh (4,10,000ha), Madhya Pradesh (3,20,000ha), and Gujarat (3,50,000ha). These six states account for over 70% of the total pigeonpea production. The trends in pigeonpea area, production and productivity over the last five decades show about 2% annual increase in its production area, but the yield levels have remained low and unchanged at around 700 kg ha-1 (Saxena et al., 2005; Saxena, 2008).

The pigeonpea crop is adaptable in a number of cropping systems and is grown on marginal to rich soils. It is grown in a wide range of soils from sandy to heavy pH of 5.0 to 8.0. It is widely used as a pulse, green vegetable, fodder, and for a variety of other purposes (Nene and Sheila, 1990). It is a hardy, well-adapted and drought-tolerant crop. There is a large variation in its maturity that helps in its wide adaptation including diverse locations and cropping systems. Some of the salient points about pigeonpea are that the short-duration (100-140days) varieties of pigeonpea are grown as a sole crop, while the medium (160-180days) and long-duration (> 200days) types are invariably grown as intercrop or mixed crop with other short-duration cereals and legumes. The main use of the pigeonpea is in the form of de-hulled and split peas (locally known as dhal), while the tender seeds and pods are consumed as fresh vegetable. The broken and damaged seeds are fed to animals, while the leaves are used as fodder. The dry stems are used as fuel

1 wood. In addition, the perennial type pigeonpea is grown on sloping mountains for reducing soil erosion (Saxena et al., 2006a; 2008).

The high nutritive value of pigeonpea is perhaps the most important reason why it finds an important place among the small-holder poor farmers. Pigeonpea is mainly traded for consumption as food. It is a rich source of protein, carbohydrate, and certain minerals. The protein content of commonly cultivated pigeonpea varieties ranges between 17.9 to 24.3g/100g sample (Salunkhe et al., 1986) in whole grain samples, and between 21.1 to 28.1g/100g sample in the split seed samples. The wild species of pigeonpea are a promising source of high-protein and several high-protein genotypes have been developed with protein content as high as 32.5% (Singh et al., 1990). The pigeonpea seeds contain about 57.3 to 58.7% carbohydrate, 1.2 to 8.1% crude fiber, and 0.6 to 3.8% lipids (Sinha, 1977). The vegetable pigeonpea types are important in Central America, Western and Eastern Africa, where green peas are consumed (Morton, 1976). Vegetable types are generally large podded with large, sweet green seeds. Canned pigeonpea is also sold in certain parts of the world (Morton, 1976). The by-products of split and shrivelled seed are used as livestock feed. It provides excellent fodder for livestock and there is a great scope for selecting cultivars with not only higher grain yields but also higher forage yields and crude protein. The dry sticks, obtained after threshing, are used for various purposes such as fuel, thatching roof, fencing the sides of bullock carts and basket making. Pigeonpea produces more nitrogen from plant biomass per unit area of land than many other legumes although it usually produces fewer nodules than other legumes (Onim, 1987). The residual effect on a following cereal crop can be as much as 40 kg N/ ha (Rao et al., 1983). With so many benefits at low cost, pigeonpea has become an ideal crop for sustainable agriculture systems in rain-dependent areas.

Crop protection specialists estimate that the pests and diseases that attack pigeonpea are responsible for US$ 1.1billion in annual losses across the semi-arid tropic (exploreit.icrisat.org/page/pigeonpea/687). Principal plant diseases include Fusarium wilt and Sterility mosaic; major insect pests are pod borers, mainly Helicoverpa armigera, Maruca and podfly. Resistance to some of these constraints is not present in the cultivated genotypes, but the wild relatives have been found to be good sources of resistance (Mallikarjuna et al., 2011a).

The primary objective in plant breeding is to widen the genetic base of a cultivated species. If the genetic variation is limited, as in the case of pigeonpea, breeders can incorporate alien variation through introgression, induce mutations or exploit somaclonal variation. The genetic potential of wild relatives is widely demonstrated in plant breeding and in the evolutionary studies. Wild relatives have helped to fill the voids in traditional breeding programmes. Pigeonpea germplasm collection developed from 13,771 pigeonpea accessions collected from 76 countries and is maintained at International Crops Research Institute for Semi Arid Tropics (ICRISAT), Patancheru, India. This includes 555 accessions of wild relatives, which represent six genera and 57 species (Upadhyaya et al., 2007). The majority of the collection has been characterized for morpho-agronomic traits of importance in crop improvement.

Pigeonpea belongs to the sub tribe Cajaninae and contains 13 genera. Earlier, the genus Atylosia and Cajanus were considered to be closely related, however, recently the genus Atylosia has been merged with Cajanus (van der Maesen, 1980). Subsequently, the Cajanus has 32 species, 18 of which are endemic to Asia and 13 to Australia and one to western Africa (van der Maesen, 1986). Apart from these, there are other related genera, namely Rhynchosia, Dunbaria, Flemingia, Paracalyx, Eriosema, Adenodolichos, Bolusafra, Carissoa, Chrysoscias and Baukea. Cajanus species, which are endemic to Australia, are Cajanus lanceolatus, C. confertiflorus, C. viscidus, C. acutifolius, C. aromaticus, C. crassicaulis, C. lanuginosus, C. latisepalus, C. reticulates, C. pubescens, C. cinereus, C. marmoratus and C. mareebensis, and C. kerstingii is endemic to Africa.

Pigeonpea is the only cultivated species under the genus Cajanus and the remaining 31 species belonging to this genus are wild. In addition to India, the greatest diversity of wild species in Cajanus is found in Myanmar, Yunnan China, and northern Australia. Several species, such as C. villosus, C. elongatus, C. granadiflorus and C. niveus that were earlier known to exist in northeastern India are either rare or extinct (Aruna, 2003). Harlan and de Wet (1971) proposed a systematic means of grouping the germplasm of a crop species and their wild relatives. Mallikarjuna et al. (2011a) classified Cajanus germplasm into three basic gene pools namely primary gene pool (GP1), secondary gene pool (GP2) and tertiary gene pool (GP3) with the related species in the quaternary gene pool (GP4).

Pigeonpea has a narrow genetic base due to which the crop is susceptible to a range of diseases and pests such as Fusarium wilt, sterility mosaic disease, insect pests such as pod borer [Helicoverpa armigera (Hub.), also known as H. armigera], spotted pod borer [Maruca vitrata (Geyer)], pod fly [Melanagromyza obtuse (Malloch)], pod bug (Clavigralla spp.), Lima bean pod borer [Etiella zinckenella (Tr.)], and the bruchids [Callosobruchus chinensis (F.)] (Parde et al., 2012). The levels of resistance to many of these pests and diseases are low to moderate in the cultivated germplasm (Sharma et al., 2005), but the wild relatives of pigeonpea have shown high levels of resistance to many of the constraints (Green et al., 2006; Sharma et al., 2009; Sujana et al., 2009). Wild relatives of crop plants are important sources of resistance to various biotic and abiotic constraints (Mallikarjuna et al., 2011a). They possess several valuable traits, including cytoplasmic and genetic male sterility systems (Reddy and Faris, 1981; Ariyanayagam et al., 1995; Saxena et al., 2010; Mallikarjuna et al., 2012) and partial cleistogamous trait, which ensues very high purity of genotypes (Saxena et al., 1988b), pod borer resistance (Lateef et al., 1981; Dodia et al., 1996; Shanower et al., 1997; Mallikarjuna et al., 2007; Jadhav et al., 2012), nematode resistance (Sharma, 1993), and salinity tolerance (Subba Rao, 1988; Srivastava et al., 2006).

Efforts have been made in past to increase the average productivity of pigeonpea by developing a number of high yielding pure varieties and in spite of releasing more than 100 varieties, the yield levels remained unchanged (Saxena 2006b). In this scenario, the use of hybrid technology in enhancing productivity has a potential. A stable male sterility system in conjunction with existing natural out-crossing can make it possible. Since the pigeonpea had no male sterility system before the development of hybrid breeding program, a deliberate search for male sterile genotypes in the germplasm was made at ICRISAT that led to the identification of male-sterile plants in ICP 1596. In this accession the GMS was associated with translucent anthers and it was controlled by a single recessive gene ms (Reddy et al., 1978).

Later, another source of male sterility, characterized by brown anthers and controlled by non-allelic single recessive gene ms2 was also reported (Saxena et al., 1983). These GMS sources were used for developing hybrids and the world‟s first pigeonpea hybrid ICPH 8 was released by ICRISAT for cultivation in 1991 (Saxena et al., 1992). Since it was the first commercial hybrid to be available (in any pulse crop), the release of ICPH 8 is considered a milestone in the history of legume breeding. In spite of

4 high yield ICPH 8 did not become popular due to seed production problems. To overcome such constraints, it was found necessary to have a CMS system.

So far nine different cytoplasmic sources have been identified for the utilization in practical pigeonpea hybrid breeding programs (Saxena et al., 2010a; Mallikarjuna et al.,

2012). These are: A1 CMS system from Cajanus sericeus (Saxena et al., 1997), A2 CMS system from C. scarabaeoides (Tikka et al., 1997; Saxena and Kumar, 2003), A3 CMS system from C. volubilis (Wanjari et al., 2001). A4 CMS system from C. cajanifolius

(Saxena et al., 2005), A5 CMS system was developed from Cajanus cajan (L.) Millsp.

(Mallikarjuna and Saxena, 2005), A6 CMS system from C. lineatus (Saxena et al., unpublished), A7 CMS system from Cajanus platycarpus (Mallikarjuna et al., 2011b) and

A8 cytoplasm were developed from C. reticulates (Saxena et al., 2013). More recently, traits for CMS were observed in the progeny lines derived from C. lanceolatus (Srikanth et al., in Press). Once published, this will be named as A9 CMS system.

Of all the pests/ insects the cotton bollworm/legume pod borer, H. armigera is one of the most damaging pests worldwide. Losses due to pod borer valued over US $325million annually. Helicoverpa armigera has a wide host range, and feeds on more than 300 plant species, of which pigeonpea is highly preferred. The larvae of H. armigera feed on leaves, flowers and pods. When activity of H. armigera occurs during vegetative stages, significant amount of leaf feeding can occur. However, once flowering commences, feeding occurs preferentially on reproductive plant parts. The usual sequence followed by a H. armigera on pigeonpea appears to be for moth to lay eggs on flowers, young pods or leaves in the upper part of the crop.

Knowledge of the resistance mechanisms and associated factors involved is essential to effectively use the sources of resistance in breeding programs. Despite large scale screening of the germplasm, lines with high levels of resistance were not obtained (Sharma, 2005). Hence, it felt that there is ample scope for substantially improving host plant resistance (HPR) in pigeonpea to H. armigera.

During the course of evolution, plants acquire several defence mechanisms against insect pests. The major mechanisms are antixenosis (non-preference), antibiosis, tolerance, and escape (Painter, 1951). Plants use these mechanisms through different component traits. Using specific assays to monitor the effects of particular physical and chemical characteristics on insect behavior and physiology, it has been reported that

5 antibiosis than antixenosis or tolerance has been reported in legume crops (Clement et al., 1994).

Plants synthesize various proteinaceous and non-proteinaceous compounds against insect attack, amongst these, proteinase inhibitors (PIs) are the most-studied class of plant-defense proteins. PIs reduce the digestive capability of insects by inhibiting proteinases of the midgut, thereby arresting their growth and development (Broadway and Duffey, 1986; Delano et al., 2008), and have been deployed in defense against insects through transgenic technology (Jouanin et al., 1998; Lawrence and Koundal, 2002).

Analysis of digestive proteinases of H. armigera has revealed the presence of serine proteinases, predominantly trypsin and chymotrypsin like enzymes (Johnston et al., 1991; Xu and Qin, 1994; Bown et al., 1997). In polyphagous insects such as H. armigera, diverse specificities and intricate changes in the expression of proteinases are responsible for the inactivation of host plant and newly exposed PIs (Bown et al., 1997; Jongsma et al., 1995; Broadway, 1996; Gruden et al., 2004; Koiwa et al., 1997).

Most PIs are small proteins that are found in all plant species investigated so far, and occur in both reproductive and vegetative tissues. In herbivorous insects, PIs act by inhibiting protein digestive enzymes in the gut of insect larvae or adults, resulting in amino acid deficiencies; these deficiencies lead to serious developmental delay, mortality, or reduced fecundity. In legumes, PIs accumulate in large amounts during seed maturation, and play an important role both in the depositing storage protein and in plant defense (Koiwa et al., 1997). Protease inhibitors are induced under various stress-prone conditions such as insect chewing, mechanical wounding, pathogen attack, drought and UV exposure (Schaller and Ryan, 1995; Conconi et al., 1996).

Hence, identifying PIs that impair insect gut proteinases with high binding efficiency is necessary to effectively inhibit midgut proteinases. Such PIs will have direct relevance and application in the development of plants with resistance insects. This is most relevant to H. armigera.

Even with the availability of germplasm with wide variability for various economic characters within the cultivated types, little progress has been made in evolving cultivar hybrids with durable resistance to biotic stresses. Resistant and less susceptible cultivars would provide an equitable, environmentally sound, and sustainable pest

6 management tool. More than 15,000 pigeonpea accessions have been screened at ICRISAT and in other national agricultural research programs (Sharma, 2005). Though several genotypes (with no or less pod damage) have been identified, these genotypes have not been widely used. The level of tolerance in the cultivated genotypes is low, and some of the genotypes are susceptible to major pigeonpea fungal, viral pathogens and insects (Sharma, 2005).

Wild Cajanus species have been identified as a potential source of resistance (Pundir and Singh, 1987; Saxena et al., 1990; Shanower et al., 1997; Sharma et al., 2009). There is also some evidence that the wild species have different mechanisms of resistance against pod borer than in the cultivated types (Chougule et al., 2003: Parde et al., 2012; Jadhav et al., 2012). However, despite the number of sources of resistance, their utility in pigeonpea improvement has not been fully explored. A few on the utilization of wild relatives in pigeonpea improvement include their exploitation as sources of resistance to pod borer, pod wasp and Phytophthora blight (Reddy et al., 1982; Saxena et al., 1996; Sharma et al., 2000, Mallikarjuna et al., 2011, and Jadhav et al., 2012). Thus, opening an avenue to breed pigeonpea resistant to many biotic constraints which were not possible before.

The pigeonpea also faces other production constraints. These include post-harvest insect pest problems [25 species of insects attack pulses (Prabhakar, 1979)), improper sanitation and storing causes both qualitative and quantitative loss to pulses. Of these, Coleopteran insect pests cause major damage to stored grain and grain products worldwide. It is estimated that about 8.5% of the total damage to stored grains is inflicted by insect pests amounting to great loss. Among them, Callosobruchus species belonging to the family Bruchidae, are very serious pests of stored legumes. Pulse beetles also cause a lot of damage both in the field and storage (Anonymous., 1970). About hundred and seventeen species of bruchids belonging to 11 genera have been recorded from India (Arora, 1977). The genus Callosobruchus includes a number of important species that attack stored pulses throughout the world. Callosobruchus chinensis (L.) and C. maculatus (F.) are predominant species and cause heavy damage to pigeonpea and cowpea. Pingale et al., (1956) recorded zero % germination due to the bruchids (C. chinensis) infestation after six months in stored green gram. During storage, pulses undergo some chemical changes due to the presence of insects. They alter the flavour and nutritive value of grains which reduce the marketability and acceptability of pulses.

The most common method to minimize the loss due to the above reasons has been to use chemical pesticides. However, pesticides cause environmental and health hazards. What‟s more, most importantly, insects develop a resistance to chemical pesticides over time. In such a situation, using sustainable methods to minimize loss is the only alternative to the problem.

Gene-based resistance is one of the most sustainable methods of pest control, particularly as a basic element in integrated pest management approach. Legume seeds are a rich and varied source of secondary plant compounds many of which are toxic to or anti-metabolic for insect pests. Knowledge of the biochemical mechanisms of resistance to insects enables plant breeders to assess their potential and to exploit such traits in plant breeding programs directly. Among the anti-metabolites, protease inhibitors are usually present in storage tissues in seeds and tubers and their expression are both regulated in development and are induced in response to insect and pathogen attack.

The possibility of wide hybridization using wild relatives from different gene pools has opened up new vistas for broadening genetic base of variation. It will also help in improving pigeonpea production and transferring insect/pest resistance genes from the wild into cultivated pigeonpea. Thus, with an aim to further understand wild Cajanus species and their potential significance in improving insect/pest resistance and to identify the genetic basis of different qualitative and quantitative traits (including the resistance against pod borer), the present investigation was undertaken with the following

Objectives:

 To develop interspecific hybrids through wide hybridization utilizing the closest wild relatives C. cajanifolius, C. lanceolatus from secondary gene pool, C. volubilis from tertiary gene pool, and study their progeny

 To back cross selected lines from C. platycarpus, with cultivated parent and study their progeny lines

 To determine the homology of both parents involved in crossing programme and reason for pollen fertility/sterility, through cytological studies

 To get an insight into the developmental stages of microsporogenesis from the tetrad stage up to pollen formation in sterile hybrids derived from the C. lanceolatus, by by anatomical analysis using microtomy

 To evaluate hybrid purity of hybrids derived from C. lanceolatus and C. volubilis by using simple sequence repeat (SSR) markers

 Biochemical analysis:

To estimate of protein levels in crude extracts of seeds from parents and hybrids and examination of proteinase inhibitory activity against bovine trypsin and chymotrypsin

To evaluate insecticidal properties of PI against trypsin like proteinases of H. armigera

 To screen hybrids (C. lanceolatus, C. cajanifolius and C. platycarpus derived lines) for pod borers and bruchid resistance (only for C. lancoelatus derived lines).

REVIEW OF LITERATURE

Review of Literature

Pigeonpea

Pigeonpea is one of the major grain legumes of tropics and sub-tropics and second most important pulse crop of India. It ranks second in area, production and productivity after chickpea. Since pigeonpea is widely grown in Indian sub-continent and has a variety of end usages, special attention needs to be given to this crop for enhancing its productivity. The deep roots of pigeonpea can break the hard plough pans and improve soil structure. It can tolerate salinity and alkalinity but not excessive acidity. It has special mechanism to use phosphorus from soil to meet its needs (Saxena, 2008). Extensive ground cover by pigeonpea prevents soil erosion by wind and water and encourages filtration and minimizes sedimentation. Being a legume crop, pigeonpea fixes nitrogen and the leaf fall at maturity not only adds to the organic matter in the soil but also provides additional nitrogen for the succeeding crop. It has been estimated that around 40kg nitrogen ha-1 added from the leaf fall and nitrogen fixation (Kumar Rao et al., 1983).

Pigeonpea is a perennial shrub with grooved silky branches. The point on the main stem where branching starts, the number of secondary branches and the angle at which these are given off also vary. The root consists of a deep, strong, woody tap root with well developed lateral roots in the superficial layers of the soil. Roots are nodulated by rhizobia, usually by a slow growing Rhizobium species. Stem is woody and straited, branching normally begins from sixth to tenth node. The plants grow into woody shrubs, 1-2m tall when annually harvested. It may attain a height of 3-4m when grown as a perennial plant in fence rows or agroforestry plots. When sown under optimal moisture and temperature (29 to 36°C), the seed testa splits open near the micropyle on the 2nd day. The tip of the radical elongates and emerges from the seed coat. On the 3rd day the hypocotyl appears as an arch and continues to grow upward. The hypocotyl turns light purple. The seedling epicotyl elongates 3-7 cm before the first trifoliate leaf emerges (Reddy, 1990).

Predominate stem color in pigeonpea germplasm of Indian origin is green; while in African germplasm predominant color is purple. In certain cases, unstable purple stem pigmentation due to the exposure of stems to direct sunlight is observed. Purple stem

10 color was found dominant to green and was found to be controlled by a single factor, Pst (D'Cruz and Deokar, 1970; D'Cruz et al., 1971; D'Cruz et al., 1974). However, in a cross between cultivars, N. Black x Purple grained, Deokar and D'Cruz (1972) reported an F2 ratio of 45 purple: 19 green, and suggested that three genes Psta, Pstb, and Pstc. These inherited contrasting stem colors i.e., have been used as markers to detect the extent of natural out crossing in pigeonpea (Bhatia et al., 1983).

Deshpande and Jeswani (1952) and Deokar and D'Cruz (1971) reported that the prostrate growth habit was recessive to normal erect type, and controlled by a single gene.

However, Patil and D'Cruz (1965) and Shinde et al., (1971) observed the F2 ratio of 13 normal: 3 creeping types. Deokar et al., (1971) observed that the growth habit was controlled by three genes Cgra, Cgrb and Cgrc, giving a ratio of 45 erect: 9 creeping: 10 prostrate in the F2 generation.

A number of genetic studies have been reported on plant height and branching habit (erect, compact, spreading). Sen et al. (1966) identified bushy dwarf pigeonpea phenotypes where the dwarf was controlled by a recessive gene, d. Waldia and Singh (1987) reported that dwarf phenotype in the Do dwarf line was governed by two non - allelic recessive genes t1 and t2. Saxena et al. (1989a) studied inheritance of three dwarfs D6, PDI, and PBNA and reported that the dwarfing trait in each line was controlled by a single recessive gene. Shaw (1931) observed dominance of erect growth habit over spreading type. D' Cruz and Deokar (1970) reported that a single dominant gene Sbr controlled spreading habit, and the erect 30 types were homozygous recessive. According to De Menezes (1956) branching angle is quantitatively inherited. D'Cruz et al., (1971) observed that the branching habit was governed by three duplicate complementary factors

Sbra2, Sbrb2 and Sbrc2, giving an F2 ratio of 54 spreading: 10 erect types.

The branching pattern in pigeonpea depends on genotype and spacing between rows and plants. At a wide spacing, it may form a bush and at narrow spacing it may remain compact and upright. For agronomic purposes pigeonpea plants can be grouped as compact (erect), semi-spreading (semi erect) and spreading types.

Based on the flowering pattern it may be determinate or non determinate. The determinate type completes the vegetative phase and then enters into the reproductive phase. In this type, the apical bud of the main shoot develops into an inflorescence, and the sequence of inflorescence production is basipetal (developing in the direction of

11 base). The indeterminate type shows continuous vegetative and reproductive phases. In this type, the flowering starts at nodes behind the apex and proceeds both acropetally and basipetally. Another group is semi-determinate between the determinate and indeterminate types. It includes late-maturing genotypes where branching starts from different angles, but most of the pods are at the upper region of the plant. Kumar et al. (1985) and Pundir and Singh (1985) reported the twining growth habit of C. scarabaeoides and C. albicans, controlled by two genes with epistatic gene action resulting in a ratio of 13 non- twining: 3 twining. The erect growth habit of pigeonpea was dominant to the spreading growth habit of C. scarabaeoides. The plants were intermediate between erect and spreading habit and in F2 generation they observed a ratio of I erect: 1 spreading: 14 intermediate, suggesting that two genes (Egl and Eg2) with partial dominance were responsible for the growth habit. Pundir and Singh (1986) studied inheritance for pod length and ovule number in six interspecific F2 populations. The interspecific crosses of C. lineatus and C. scarabaeoides showed transgressive segregation for pod length. However, in the interspecific crosses involving Pigeonpea a restricted segregation was observed which was attributed to a negative gene interaction in the two species.

The first two leaves in the seedling called primary leaves, are simple, opposite, and caduceus. The later leaves are pinnately trifoliate with lanceolate to elliptical leaflets that are acute at both ends and are spirally arranged. The leaflets are borne on a rachis, which is swollen at the base (pulvinus). The leaf size varies from 6 to 17cm in length and are about the same width. The rachis varies from 2-4cm and the terminal leaflets are 4- 8cm by 2-3.5cm. The lateral leaflets are slightly smaller. There is genetic variability in the size, shape, and color of leaves. The first two leaves are simple, opposite and caduceus. They are narrowly ovate with a chordate to truncate base and an acute to acuminate apex. Subsequent leaves are compound, trifoliate and arranged in a two to five types of spiral. Terminal leaflets are mostly symmetrical, but the side leaflets are broader than other leaves. Terminal leaflets are usually bigger than lateral leaflets. The hair types are simple or glandular. The latter are spherical and contain a yellow oily material, probably responsible for the fragrance of pigeonpea plants (Bisen and Sheldrake, 1981). In general, the trifoliate leaflet of pigeonpea is lanceloate, but some morphological variations in the leaflet shape have been reported. The first report of inheritance of leaflet shape in pigeonpea was published by Pandya et al. (1954). They referred to both obovate

12 and round shaped leaflets and reported a F2 ratio of 3 lancelolate: 1 rounds leaflets. The monogenic inheritance of lanceolate leaflet shape was also confirmed by Deokar et al. (1971), D'Cruz and Deokar (1970) and D'Cruz et al. (1971). Deshpande and Jeswani (1956) observed segregation ratio of 3: l for lanceolate and 15: l for obcordate leaflets in the F2 generation of two different crosses. D'Cruz et al. (1971) reported a ratio of 117 oblong or oval (round) leaflets with obtuse species: 75 lanceolate leaflets with acute species: 64 obcrodate leaflets with retuse species in the F2 population of a cross, involving obcordate and round leaflet types.

In most cultivars, flowers are borne on terminal or auxiliary racemes (4 to 12cm) and are carried on a long peduncle. The raceme inflorescence forms a terminal panicle in non-determinate types and a somewhat corymb-shape bunch in the determinate types. These are grouped together at the end of branches in late types and distributed along the branches in early, medium, and indeterminate types (Sharma and Green, 1980). The number of racemes plant-1 in the pigeonpea world collections ranged from 6 to 915 (Remanandan et al., 1988). Flowering proceeds acropetally (in the direction of apex) both within the raceme and on the branch.

The flowers are clustered at the top of the peduncle. The peduncles are 1 to 8cm long. Flowers are frequently yellow. The bracts are small with a thick middle nerve. They are ovate lanceolate with hairy margins and curved inwards to form a boat like structure to enclose one to three young lateral buds. The pedicel is thin, 5 to15mm long, and covered with hair. The flowers are mostly yellow and papileonaceous or completely bisexual and zygomorphic (Sundararaj and Thulasidas, 1980). The colorcolor of the flowers varies from yellow to yellow with purple veins and from yellow to diffused red. The flowers are usually 2cm in length.

For days to flower Gupta et al. (1981) reported the predominance of additive gene effects, while Pandey (1972), Sharma et al. (1973b) and Dahiya and Satija (1978) observed additive gene action with partial dominance for earliness. Time of flowering plays an important role as the growth season is restricted by climatic factors like drought and high temperatures. Duration of flowering period majorly impacts the yield in the indeterminate growth habit of certain pigeonpea genotypes. Sharma et al. (1973a), Dahiya and Brar (1977), Dahiya and Satija (1978), Gupta et al. (1981) and Reddy et al. (1981b) reported additive gene action for days to flower, and non-additive gene action was

13 reported by Reddy et al. (1 981 b). Both additive and non-additive gene action for days to flower was reported by Sidhu and Sandhu (l981), Saxena et al. (1981b) and Chaudhari et al. (1980). Pandey (1972) and Sharma et al. (1972) reported additive gene action for days to maturity.

The calyx tube is companulate with glandular hairs. The corolla is zygomorphic and generally yellow in color. The petals are imbricate in the bud. The standard petals are erect and spreading more or less orbicular, base clawed, biauriculate with two callosities. Stamens are 10 and diadelphous (9+1). Anthers are ellipsoid, dorsifixed and yellow in color. The ovary is superior with two to nine ovules. The stigma is capitate and glandular- papillate. The style is long, filiform, upturned beyond the middle and glabrous. A pod is formed 15-20days after fertilization. Pods are oblong, straight or sickle shaped, green at the younger stage but mature pods vary as dark purple, purple and green. Pod length varies from2-8cm and pod width ranges from 0.4 to 1cm. The seeds per pod range from two to seven, and sometimes up to nine. The seeds are in separate locules and the cross walls develop during the 1st week after fertilization. The pod wall develops more rapidly than the young seeds. Seed development is visible 7days after pollination.

Some Cajnaus species are characterized by the presence of a prominent strophiole on the seed surface. Reddy (1973) reported 9:7 F2 ratio, from a cross between C. cajan and Atylosia species, suggesting that the involvement of two genes with complementary gene action. Reddy et al. (1981a) and Kumar et al. (1985) reported that in C. scarabaeoides, C. sericeus and C. albicans the presence of strophiole was controlled by two genes (NS and SDI) with inhibitory action. But Pundir and Singh (1985) reported that seeds with strophioles in Cajanus spp. are due to the presence of two genes (s1 and s2) with duplicate gene action. Singh, (2000) reported that the strophioled character is dominant over the non- strophioled character and is governed by a single gene.

Pundir and Singh (1985) studied inheritance of seed color in C. scarabeoides and C. cajanifolius in crosses with orange seeded pigeonpea lines. They reported that a single partially dominant gene, Osc, governed the dark seed color of Cajanus spp. Reddy et al. (1981a) and Kumar et al. (1985) found that seed mottling, was controlled by two complementary genes, Msda, and Msdb.

The seed shape generally varies in four namely oval, pea, square and elongate. The oval shape is most common. Seed coat color ranges from white to black. Germination

14 is hypogeal and cotyledons remain under ground. Under suitable conditions the seedlings appear above the ground in five to six days.

Unlike most legume species, pigeonpea flowers are prone to natural out-crossing and thus it is considered as a partially cross-pollinated species. Self-pollination occurs in the bud before the flowers open while cross pollination takes place after the opening of flowers with the help of insects. In the young buds, the stigma lies above the level of anthers and the style is curved at the tip in a way that the stigmatic surface is directed towards the anthers. These are arranged around the style in two groups of five in each. As the bud develops, the filaments elongate, bringing the top five anthers dehisce in the bud a day before the flowers open. Thus, self-pollination takes place.

Although the stigma is completely covered with the pollen of its own flower, considerable out- crossing occurs in pigeonpea (Saxena et al., 1990). The percentage of “selfs” was negligible when flower buds were pollinated with foreign pollen without emasculation (Reddy and Mishra, 1981). This indicates that foreign pollen has an advantage over native pollen in fertilization. Although anthers dehisce during the bud stage, they do not start germinating until the flowers start to wither 24 to 28hours after dehiscence (Onim, 1981). It has been found that the receptivity of stigma starts 68hours before anthesis and continues for 20hours after anthesis (Prasad et al., 1977). These mechanisms provide a sufficient gap for foreign pollen to be introduced onto stigma and thus favour out-crossing in pigeonpea.

Inter-varietal or intraspecific crosses are preferred because the hybrids offer the fittest options. Intraspecific hybrids are most viable and fertile, and hence are favored under both nature and domestication. In contrast, wide crosses suffer either from non- viability or sterility or both. As a consequence, the forces of natural selection promptly eliminate such hybrids of wide crosses. In crop improvement programs, the parents used in the hybridization are generally differed varieties of the same species. But in many cases it may be desirable or even necessary to cross individuals belonging to two different species or genera. In certain crops, plant breeders in the 20th century have increasingly used interspecific hybridization for transfer of genes from a non-cultivated plant species to a crop variety in a related species.

The first recorded interspecific hybrid was in 1717 between a carnation and sweet William by Thomas Fairchild (Allard, 1960). The first man-made cereal, "Triticale" was

15 an outcome of inter-generic hybridization. Extensive studies on distant hybridization have been made in crops like wheat (Sears, 1975), barley (Bothmer et al., 1983), maize (Mangelsdrof, 1974; Harlan and De wet, 1977), Solanum (Swaminathan and Magoon, 1961; Motskaitis and Vinitskus, 1975), cotton (Blank et al., 1972; Meyer, 1973; Meyer, 1974), Nicotiana (Mann et al.,1963; Smith, 1968; Berbec, 1974), tomato (Rick, 1982) and rice (Nayar, 1973).

Distant hybridization studies have also been carried out in several important legumes such as, Phaseolus, Vigna, Vicia, Pisum and Arachis. Species of genus Phaseolus have been a subject of wide interest. The possibility of gene exchange between species had led to several studies on interspecific hybridization, especially between P. vulgaris and P. coccineus (Mendel, 1866; Tschermak-Seysenegg, 1942; Lamprecht, 1948; Rudrof, 1953; Kedar and Bemis, 1960; Thomas, 1964; Al-yasiri and Coyne, 1964; Rutger and Beckman, 1970; Smart, 1970; Marechal, 1971; Haq et a1., 1980; Savova, 1981; Shii et al., 1982).

Hybridization with other Phaseolus species including the wild forms has received wide attention (Honma, 1956; Coyne, 1964; Braak and Koistra, 1975; Le Merchand et al., 1976; Tan Boun Suy, 1979; Hwang, 1979). Interspecific hybridization in Phaseolus in recent years has been widely used in the improvement of the Phaseolus species with respect to disease resistance, insect resistance, nitrogen fixation and several agronomic characters (Lapinskas, 1980; Bannerot et al., 1981; Zapata et al., 1982; Hunter et al., 1982). Wide hybridization has also received a fair degree of attention in the genus Glycine. Studies on hybridization between G.max and G.soja have been extensive (Karasawa, 1936; Ting, 1946; Williams, 1948; Tang and Chen, 1959; Tang and Tai, 1962; Kiazuma et al., 1980).

Interspecific hybridization in Cajanus dates back to 1956 when Deodikar and Thakur (1956) made the first cross between C. cajan with C. linearus. The hybrid was fairly fertile. Kurnar et al. (1958) extended the earlier work on to hybrid cytology and found regular bivalent formation in the hybrid. A hybrid between C. cajan and C. scarabeoides was obtained by Roy and De (1967) who expressed doubts about the generic status of Cajanus. Reddy (1973) analysed pachytene chromosome pairing in Cajanus cajan, C. lineata, C, scarabaeoides and C. sericeus and their hybrids. These pachytene studies in general revealed a high degree of chromosome homology between

C. cajan and the three species of wild Cajanus. Ariyanayagam and Spence (1978) reported hybrids between C. cajan and C. platycarpus while further attempts (Reddy et al., 1980; Pundir, 1981) to cross Cajanus cajan with C. platycarpa failed. Further attempts in C. cajan and wild relatives hybridization by Pundir (1981) involved karyotype comparisons between cultivated and wild species and meiotic pairing in the FI hybrids. These studies revealed a great degree of karyotypic similarities between species. All the studies on Cajanus and wild hybridization revealed a close relationship between the species and regular pairing in their hybrids which nevertheless exhibited a fair degree of sterility.

Most of the interspecific hybridization work was done at ICRISAT. Such programs mostly focused on improving breeding high protein lines (Reddy et al., 1979) and did not focus on breeding for insect resistance, dwarfs and isolation of cytoplasmic male sterile involving a few wild accessions of Cajanus. Genome relationships between wild and cultivated Cajanus species are still obscure. C. cajanifolius, which is morphologically very similar to Cajanus except for the seed strophiole and a few other traits that separate the two (Mallikarjuna et al., 2012), was identified as early as 1920 (Van der Maesen, 1980) but an attempt to cross these two species was made by Pundir (1981).

Distant hybridization is mostly aimed at introducing new genetic variability or to achieve a new genomic constitution in such a way that the characters of the parental species are recombined effectively. These possibilities are directly related to the degree of genetic closeness between the parents. It has been found that the closer the genome relationship between the cultivated and the wild species the greater the amount of genetic recombination, and consequent variability. Assessment of genome relationships is a first step in the exploitation of wild species in the improvement of any cultivated species.

Table 1. Interspecific hybridization in Cajanus species. Cajanus species Author(s) C lineatus Deodikar and Thakur (1956) C. lineatus Kumar et al. (1958) C. scarabaeoides Roy and De (1967) C. lineatus, C. scarabaeoides and C. sericues Reddy et al. (1973) C. platycarpus and C. sericeus Mallikarjuna and Moss (1995) C. albicam,C. sericeus, C. lineatus, C. scarabaeoides, C. trivervea and C. cajanifolius Pundir (1981) C. reticulatus (sub-sp. Reticulata), C. pluriflorus Dundas (1983); and C. acutifolius Mallikarjuna et al. (2011) C. lanceolatus Sateesh kumar (1985); Srikanth et al. (2013) C. lineatus, C. scarabaeoides and C. reticulatus Saxena et al. (2010, 2003 and 2013) C. platycarpus, C. cajanifolius, C. acutifolius, Mallikarjuna et al. (1995, 2002, 2006 R. bracteata and R. sublobata and 2012) C. sericeus Singh et al. (2000) C. lanceolatus and C. volubilis Mallikarjuna and Srikanth (2012)

In backcross breeding, the hybrid and the progenies in the subsequent generations are repeatedly backcrossed to one of their parents. As a result, the genotype of backcross progeny becomes increasingly similar to that of the parent from which the backcrosses are made. At the end of 6 to 8 backcrosses, the progeny would be almost identical with the parent used for backcrossing. The objective of the backcross method is to improve one or two specific defects of a high yielding variety (which is well adapted to the area and has other desirable characteristics). The characters lacking in this variety are transferred to it from a donor parent without changing the genotype of this variety, except for the genes being transferred. Thus the end result of such a backcross program is a well- adapted variety with one or two improved characters.

Backcross method has been used to transfer simply inherited characters, mostly insect and disease resistance, from related species into a cultivated species. For example, transfer of resistance into wild fire and black fire from Nicotiana longiflora to N. tobaccum a leaf and stem rust resistance from Triticum timopheevii, T, monococcum, Aegilops speltoides and rye (S. cereale) to T. aestivum, of black-arm resistance from several Gossypium species to G. hirstutum etc. Interspecific transfer of genes is easy when the chromosomes of the two species pair regularly. But often chromosomes of the concerned species are differentiated by structural changes that reduce pairing between also transferred along with the desirable gene. Another difficulty in interspecific gene

18 transfers is that the transferred gene may not be able to function in the same way in the genetic environment of the new species.

Wild relatives of wheat are a rich source of new genes for resistance to various wheat pathogens (Sharma and Gill, 1983; Gale and Miller, 1987; Jauhar, 1993; Jaing et al., 1994; Friebe et al., 1996; Hajit-Singh et al., 1998). A number of new genes for resistance to various wheat diseases included the three rusts viz., leaf rust (Puccinia econdita f, sp. Tritici), stripe rust (P. striiformis) and stem rust (P. graminis tritici), have been transferred from closely and distantly related wild species (Mclntosh, 1998). A number of genes for disease resistance transferred from wild relatives of wheat Viz. Lr9, Yr9, Pm8, Lr26 etc. have been overcome by the emergence of virulent pathotypes of pathogens when deployed in wheat cultivars.

Helicoverpa armigera has a wide host range and feeds on more than 300 plant species, of which pigeonpea is highly preferred. Prior to 1975, less than 20% farmers used insecticides on pigeonpea. However, 1993 onwards, there is a widespread adoption of insecticides for pest management on pigeonpea. Due to widespread use of insecticides, H. armigera has developed of resistance to conventional insecticides, including synthetic pyrethriods (Armes et al., 1992). Natural enemy activity on H. armigera in pigeonpea is quite low (as compared to that on other crops such as sorghum (Bhatnagar et al., 1980)) and as a result, there is greater survival of the insect on pigeonpea and results in heavy loss in grain yield.

The identification and utilization of cultivars resistant/tolerant to H. armigera would have a number of advantages, particularly for a relatively low value crop such as pigeonpea. Screening of germplasm (> 15,000 pigeonpea accessions) for resistance to H. armigera has revealed very low levels of resistance to this pest (Reed and Lateef, 1990). Several lines of pigeonpea such as ICPL 7703, ICPL 332, ICPL 87088, ICPL 84060 and ICPL 87089 with low to moderate levels of resistance have been identified (Lateef, 1992; Sachan, 1992).

The larvae of H. armigera feed on leaves, growing points, flowers and pods. When periods of H. armigera activity occur during vegetative stages, significant amount of leaf feeding can occur. However, once flowering commences, feeding occurs preferentially on reproductive plant parts. The usual sequence followed by H. armigera

19 on pigeonpea appears to be for moth to lay eggs on flowers, young pods or leaves in the upper part of the crop.

Knowledge of the resistance mechanisms and associated factors involved is essential for effective utilization of resistance in the breeding programs. Despite large scale screening of the germplasm, it has been felt that there is a scope for substantially improving HPR in pigeonpea to H. armigera. This can be done by understanding the mechanisms by which the pod borer is either attracted to or repelled from pigeonpea.

Totally, 25 species of insects attack pulses (Prabhakar, 1979). Of these, coleopteran insect pests cause major damage to stored grain and grain products worldwide. Among them, Callosobruchus spp. belonging to the family Bruchidae, are very serious pests of legumes in storage. It is estimated that about 8.5% of the total damage to stored grains is inflicted by insect pests amounting to great loss. Pulse beetles assume greater importance as they damage both in the field and storage (Anonymous, 1970). About hundred and seventeen species of bruchids belonging to 11 genera have been recorded from India (Arora, 1977). The genus Callosobruchus includes a number of economically important species that attack stored pulse throughout the world. C. chinensis (L.) and C. maculatus (F.) are predominant species and cause heavy damage to pigeonpea and cowpea. Pingale et al. (1956) recorded zero % germination due to the bruchids (C. chinensis) infestation after six months in stored green gram. During storage, pulses undergo some chemical changes due to the presence of insects. They alter the flavour and nutritive value of grains which reduce the marketability and acceptability of pulses.

Wild species have coexisted with pests and pathogens on an evolutionary time scale and have developed alleles that are pest and pathogen resistance (Acosta-Gallegos et al., 1998). These natural defense mechanisms for diseases and pests have been lost during domestication and intense selection for agriculturally desirable traits such as high yield, improved nutritional quality and other desirable agronomic traits. Abiotic (e.g. drought, salinity) and biotic (e.g. diseases and pests) stresses constrain and adversely affect pigeonpea production and cause huge economic damage. The major diseases affecting pigeonpea production are Fusarium wilt (FW), sterility mosaic, Phytophthora blight disease and major pests causing severe damage are pod borer (H. armigera and Maruca vitrata) and pod fly (Melanagromyza obtusa) (Minja et al., 2000).

Breeding strategies to tackle these problems in pigeonpea have been attempted by various researchers (Reviewed by Saxena, 2008). The breeding programs for developing disease resistant cultivars using resistance gene sources from cultivated pigeonpea germplasm did not succeed in controlling devastating pests (e.g. pod borer). The cultivated gene pool has low genetic polymorphism and lack resistance alleles. Alternative approaches of utilizing wild species as a source of resistance have showed promising results, as there is wider genetic diversity and presence of resistance genes in the wild gene pool. Most wild species have unique traits (e.g. presence of trichomes) that confer resistance to these diseases and pests (Aruna et al., 2005).

Wild relatives of pigeonpea such as Cajanus scarabaeoides, C. sericeus, C. acutifolius, C. albicans, Rhynchosia aurea, R. bracteata and Flemingia bracteata have shown high resistance to the pod borer (Chogule et al., 2003, Sujana et al., 2008, Sharma et al., 2009, Parde et al., 2012). Cajanus platycarpus has shown resistance to the most virulent race of Phytoptora blight disease (Mallikarjuna et al., 2011a) and some of the wild relatives of pigeonpea have shown high level of resistance to pod fly (Melanagromyza obtusa) and pod wasp (Tanaostigmodes cajaninae) (Sharma et al., 2003).

Cytoplasmic Nuclear Male Sterility systems

Cytoplasmic nuclear male sterility (CMS) is a maternally inherited trait that is characterized by inability of plants to produce viable and/or functional pollen grains. This trait was first discovered in 1921 in two strains of flax (Linum usitatissimum L.) (Bateson and Gairdner, 1921). Now this CMS has been reported in more than 150 plant species (Kaul, 1988).

CMS results from interaction of sterile (S) cytoplasm with homozygous recessive alleles (fr/fr) of nuclear fertility restorer genes. When a dominant nuclear fertility restorer allele (Fr) is present, male fertility is restored in plant irrespective of a sterile (S) or normal (N) cytoplasm. Because of their value in hybrid seed production, CMS systems have been identified and characterized in many crop species, such as maize (Levings and Pring 1976), sorghum (Chen et al., 1995), rice (Tan et al., 1998), rapeseed (Erickson et al., 1986), wheat (Harold et al., 1993), sunflower (Rieseberg et al., 1994), rye (Dohmen et al., 1994), common bean (Hervieu et al., 1993), tobacco (Nikova et al., 1991), petunia

(Edwardson and Warmke 1967), pearl millet (Sujata et al., 1994) and more recently in pigeonpea.

Efforts have been made in the past to increase the average productivity by developing a number of high yielding pure varieties. Despite releasing more than 100 varieties, the yield levels remained unchanged (Saxena, 2006b). In this scenario, the use of hybrid technology in enhancing productivity has a potential. A stable male sterility system in conjunction with existing natural out-crossing can make it possible. Since the pigeonpea has no male sterility system at the commencement of hybrid program, a deliberate search for male sterile genotypes was made at ICRISAT in the germplasm that led to the identification of male-sterile plants in ICP 1596. In this accession the GMS was associated with translucent anthers and it was controlled by a single recessive gene ms (Reddy et al., 1978).

Later, another source of male sterility, characterized by brown anthers and controlled by non-allelic single recessive gene ms2, was also reported (Saxena et al., 1983). These GMS sources were used for developing hybrids and the world‟s first pigeonpea hybrid ICPH 8 was released by ICRISAT for cultivation in 1991 (Saxena, 1992). Since it was the first commercial hybrid to be available (in any pulse crop), the release of ICPH 8 is considered a milestone in the history of legume breeding. In spite of high yield ICPH 8 did not become popular due to seed production problems. To overcome such constraints, it was found necessary to have a (CMS) system.

The ICRISAT took the initiative using wide-hybridization technology. So far nine different cytoplasmic sources have been identified for the utilization in practical pigeonpea hybrid breeding programs ( Saxena et al., 2010a). These are:

(i) Cajanus sericeus from which CMS with A1cytoplasm was developed (Saxena et al., 1997)

(ii) C. scarabaeoides from which CMS with A2cytoplasm was developed (Tikka et al., 1997; Saxena and Kumar, 2003)

(iii) C. volubilis from which CMS with A3 cytoplasm was developed (Wanjari et al., 2001)

(iv) C. cajanifolius from which the most stable CMS with A4 cytoplasm was developed (Saxena et al., 2005)

(v) Cajanus cajan (L.) Millsp. from which A5 cytoplasm was developed (Mallikarjuna and Saxena, 2005)

(vi) C. lineatus from which A6 cytoplasm was developed (Saxena et al., unpublished)

(vii) Cajanus platycarpus from which A7 cytoplasm was developed (Mallikarjuna et al., 2011b)

(viii) A8 cytoplasm was developed from C. reticulates (Saxena et al., 2013)

(ix) More recently, traits for CMS (A9) were observed in the progeny lines derived from C. lanceolatus (Srikanth et al., in Press).

The identification and utilization of cultivars resistant or tolerant to H. armigera would have a number of advantages, particularly for value crops such as pigeonpea. Notwithstanding the availability of vast germplasm with wide degree of variability for various economic characters within the cultivated types, little progress has been made in evolving varieties hybrids with durable resistance to biotic stresses. Resistant and less susceptible cultivars would provide an equitable, environmentally sound, and sustainable pest management tool.

Though several genotypes (with no or less pod damage) have been identified, these genotypes have not been widely used. The level of tolerance in the cultivated genotypes is low, and some of the genotypes are susceptible to major pigeonpea fungal and viral pathogens. Wild relatives of Cajanus species, especially C. scarabaeoides, have been identified as a potential source of resistance (Pundir and Singh, 1987; Saxena et al., 1990; Shanower et al., 1997). There is also some evidence that the wild species have different mechanisms of resistance against pod borer than in the cultivated types. The genes from the wild relatives can be tapped through wide hybridization for resistance to these pests. However, despite the availability of a wide array of wild sources of resistance, their utility in pigeonpea improvement has not been fully explored. A few isolated reports of utilization of these wild species in pigeonpea breeding include their exploitation as sources of resistance to pod borer, pod wasp and Phytophthora blight

(Reddy et al., 1982; Saxena et al., 1996; Sharma et al., 2000, Mallikarjuna et al., 2011a and Jadhav et al., 2012).

CMS systems in pigeonpea

The CMS systems can arise either through spontaneous mutation, intraspecific, interspecific, or inter-generic crosses. The wide hybridization programs such as interspecific and inter-generic crosses have been found to produce a greater proportion (about 75%) of CMS systems (Kaul, 1988). Reviewing literature on this subject shows that in the dicots most CMS cases have arisen through interspecific crosses, while in monocots it is the inter-generic hybrids that have yielded most CMS sources (Kaul, 1988). Since the expression of CMS requires two different genetic systems (one each in cytoplasm and nucleus) to come together in a single cell, the frequency of spontaneously occurring mutants simultaneously in both the entities (i.e., nucleus and cytoplasm) is quite low. On the contrary in Genomic male sterility (GMS) system, only a single nuclear mutation can lead to the development of male-sterility. Unlike GMS controlling genes, the influence of environment (temperature and/or photoperiod) on CMS controlling nuclear fr and Fr genes is more prominent. This may lead to instability of the expression of male-sterility and its fertility restoration. Such unstable expressions are also sometimes influenced by the genetic background of an individual.

The first attempt to breed a CMS line in pigeonpea was done by crossing a cultivated type (as female) with pollen from two different wild relatives, Atylosia sericea and A. scarabaeoids (Reddy and Faris, 1981). The fertile F1 plants of these two crosses were used as male parent to produce backcrosses with wild species as female parents. The resultant F1BC1 plants were male fertile while their F2 BC1progenies segregated for male- sterility and fertility. The maternally inherited male-sterility in these segregants was found to be closely linked with various floral abnormalities such as petaloid anthers, free stamen or heterostyly. They also reported that these segregants had different degrees of female-sterility and could never be stabilized as pure lines; they could not be used in hybrid breeding programs.

In order to develop CMS through chemical and physical mutagens, a GMS line with ms2 gene, when treated with 0.025% sodium azide or 500mg kg-1 of streptomycin sulphate, showed mutational changes and expressed male-sterility that was maternally inherited (Ariyanayagam et al., 1993). This male-sterility was maintained only by

24 heterozygote sibs that raised doubts about its nature and use in hybrid breeding program. The proportion of male-sterile plants in these mutagenic progenies varied a lot and no good male-sterile line could be derived subsequently, a few CMS systems were developed in pigeonpea and these are briefly described below:

An accession of Cajanus sericeus (A1) was crossed with an advanced breeding line of pigeonpea. The F1 progeny of this cross showed partial male-sterility but in F2 generation a few segregants expressed 100% pollen-sterility (Ariyanayagam et al., 1993). In the subsequent backcross generations these male-sterile plants could not maintain the high levels of male-sterility. In addition, it was also observed that some male-sterile plants reverted back to male-fertility when the local environment, particularly temperatures and photoperiods changed. To stabilize the male-sterile trait, conventional backcrossing and multiple cross genome transfer methodology were implemented (Ariyanayagam et al., 1995). Both these approaches yielded basic proportion of male- sterile segregants, but the backcross derivatives were found to be female-sterile and failed to set any pod. The progenies derived from the genome transfer scheme also produced a few male-sterile segregants which were maintained by other pigeonpea inbred lines. These male-sterile segregants led to the development of male-sterile lines such as CMS 85010A, CMS 88034A and CMS 13091A (Saxena et al., unpublished). From these populations, male-sterile lines that revert back to full male-fertility under low temperature and shorter days and again to full male-sterility under high temperature and longer days were selected (Saxena, 2006a).

In an attempt to develop a stable CMS line, Cajanus scarabaeoids (A2) as female parent crossed with a pigeonpea line ICPL 85030. The F1 plants were partial male-sterile. In the backcross progenies some promising male-sterile plants were identified but no stable CMS line could be bred (Ariyanayagam et al., 1993). The development of a CMS line by crossing a cultivated type with its wild relative C. scarabaeoides as a female parent was also reported. The resultant F1 plant was partial male-sterile and in F2 a number of male-sterile segregants were recovered. Subsequently, a perfect male-sterile maintainer line ICPL 288 was also identified. The fertility restoration of this male-sterile line was also found among fertile F2 segregants (Tikka et al., 1997). This male-sterile source was used in developing experimental hybrids in Gujarat, India. Cajanus scarabaeoides as a female parent was also crossed with four pigeonpea cultivars.

Among F1s, a progeny derived from cross C. scarabaeoides × ICPL 88039 was completely male-sterile. To stabilize this source of male-sterility, backcrosses were made with ICPL 88039 as recurrent parent and all the plants in BC1F1 through BC6 F1 generations were male-sterile (Saxena and Kumar, 2003). They also reported eight fertility restorers and six male-sterility maintainers. This allowed breeding of genetically diverse hybrids for different cropping systems. The fertility restoration in hybrids involving this CMS was not perfect and a large variation (50 to 95%) was observed for pollen fertility. This variation could be due to differential inter-genomic or cytoplasmic- genomic interactions (Saxena, 2008). Differences arising from genes can also yield inconsistent expressions of both male-sterility and fertility restoration (Abdalla and Hermsen, 1972).

A number of male-sterile segregants with maternal inheritance from a cross involving Cajanus volubilis (A3) and a cultivated type were selected (Wanjari et al., 2001). These selections, however, could not be used in any hybrid breeding program due to lack of fertility restoring genotypes.

In an attempt to develop CMS line from Cajanus cajanifolius (A4) as male parent with a GMS line as female parent, the progenies from this cross were male-fertile and could not be used further (Rathnaswamy et al., 1999). However, ICPW 29, an accession of C. cajanifolius (a wild relative of pigeonpea) as female parent crossed with pigeonpea line ICPL 28 (Saxena et al., 2005). The C. cajanifolius resembles the cultivated types in most morphological traits and differs by only a solitary gene (De, 1974). The interspecific

F1 hybrid plants grown in 2001 expressed variable extents of pollen-sterility and one plant with 60% pollen-sterility was backcrossed to ICPL 28. This was followed by six backcrosses to substitute the nuclear genome of wild species with that of the cultivated type. This substitution led to enhanced male-sterility that was fully maintained by its recurrent pigeonpea parent. This male-sterile source is the best among those identified so far and it was designated as ICPA 2039. It was found to be highly stable male-sterile line across environments and years and never showed any morphological deformity (Saxena, 2008; Dalvi et al., 2008a). To develop diverse pigeonpea hybrids this male-sterile source has now been transferred into a number of genetic backgrounds.

A GMS line crossed with C. acutifolius as male parent and all the F1 plants were male-fertile (Rathnaswamy et al., 1999). While using C. acutifolius as a female parent in

26 a cross with pigeonpea accession Cajanus cajan (A5) ICP 1140, only 1.5% pod set was observed. The use of gibberellic acid (@ 50mgL-1) in backcrosses enhanced the pod set to 6%. But the seeds thus obtained were underdeveloped and failed to germinate (Mallikarjuna and Saxena, 2002). To overcome this problem, the developing embryos were rescued and successfully cultured in artificial media. Encouraged with the success of embryo rescue technology, authors again crossed six pigeonpea cultivars as female parent with two accessions (ICPW 15613, ICPW 15605) of C. acutifolius. The F1s involving pigeonpea lines ICPL 85010, ICPL 85030, and ICPL 88014 produced a few male-steriles with some plants exhibiting up to 100% pollen-sterility. The anthers of these male-sterile plants were shrunken and pale yellow in color. When such male-steriles were crossed to their respective wild relative accessions, they maintained their sterility. Most of the cultivated accessions when crossed to these male-steriles restored the male-fertility of the plants. An exception to this was HPL 24, where F1 progeny produced both male-sterile and fertile plants. This suggests the presence of both fr and Fr genes in its nuclear genome (Mallikarjuna and Saxena, 2005). Further backcrossing with this line and selection for pollen-sterility helped in stabilizing the male-sterility. Interestingly, HPL 24 was bred from a cross involving C. sericeus, another wild species (Saxena et al., 2010c), and this suggested that besides C. acutifolius the fr genes may also be present in C. sericeus.

In the 2002 monsoon, a naturally out-crossed partial male-sterile plant was observed in an open pollinated population of Cajanus lineatus (A6) (Saxena, unpublished) and the morphology of this plant was very different from rest of the population. The vegetative cuttings of this plant were raised in a glasshouse and out of five cuttings planted only two survived and the plants were found to be male-sterile. These were crossed with pigeonpea line ICPL 99044 and produced normal pod set. The F1 plants grown in 2004 were partial male-sterile. Back-crosses (BC1F1) were made with ICPL

99044 and out of 20 plants grown five were partial male-sterile. In BC4F1 generation 167 plants were examined for pollen viability and it ranged from 92–100%. The plants showing 100% male-sterility were crossed with four pigeonpea lines in 2008. At present this CMS source is in BC5F1 stage with perfect male-sterility maintenance system available.

Cajanus platycarpus (A7), a wild species in the tertiary gene pool of pigeonpea, is cross incompatible with cultivated types and, therefore, hormone-aided pollinations

27 coupled with embryo rescue techniques were employed to obtain viable F1 and BC1F1 progenies (Mallikarjuna et al., 2011b). In BC2 F1 generation a progeny (BC2-E) with low pollen fertility was selected. Within this progeny two plants with 100% pollen sterility were selected and crossed with a set of pigeonpea cultivars. The examination of their F1s showed that the hybrid involving cultivar ICPL85010 maintained complete male-sterility, whereas cultivars ICPL88014 and ICP14444 restored male-fertility. The detailed studies on this new CMS source are in progress.

Cytological studies

Naithani (1941) was the first to give details about the meiotic division of pigeonpea although it was Roy (1933) who for the first time reported the chromosome number of Cajanus as n-11 which was confirmed by Akinola et al. (1972). Krishnaswamy and Ayyangar (1935) provided information about meiotic pairing chaisma frequencies in pigeonpea and this was followed by the studies of Kumar et al. (1945), Bhattacharjee (1956) and Reddy and De (1982) all of which mentioned normal chromosome pairing and regular formation of eleven bivalents at meiosis. The karyotype description of Naithani (1941) showed no variation for chromosome morphology between varieties. This attempt was followed by that of Diodikar and Thakur (1956) who gave the total complement length as 75.4µm.

These initial attempts were followed by the works of Shrivastava et al. (1973) and Sinha and Kumar (1979) who reported significant varietal variation for arm ratios and chromosome lengths and by those of Sharma and Gupta (1982) and Pundir (1981).

There were two reports on pachytene karyotype analysis (Reddy, 1981a and Dundas et al. 1983). While Reddy (1981a) identified the pachytene chromosomes by the length and centromere positions, Dundas et al. (1983) used chromomere patterns for chromosome identification. The Ciesma C banding technique was attempted for pigeonpea without much success (Lavania and Lavania, 1982).

The satellite chromosome numbers in pigeonpea have been a subject of controversy and from the literature one can see that the number varies from 0 to 2. Sinha and Kumar (1979) and Shrivastava et al. (1973) reported that some cultivars used in their studies had one Satellite (SAT) chromosome while others have none. The karyotypes

28 presented by Sharma and Gupta (1982) do not show any SAT chromosome while those of Pundir (1981) showed two.

Among the species of Cajanus, C. lineatus was used extensively by Kumar et al. (1958), Sikder and De (1967) and Reddy and De (1982) in their studies on meiotic pairing. Kumar et al. (1958), Deodikar and Thakur (1956), Sidker and De (1967), Shrivastava et al. (1973) and Pundir (1981) studied the somatic karyotype of C. lineatus and reported a close relationship between the karyotypes of Cajanus and C. lineatus. Sikder and De (1967), Reddy (1973) and Pundir (1981) studied the karyotypes of C. lineatus, C. sericeus and C. scarabaeoides. The karyotypes of these species were also presented by Reddy (1981 a, b, c) based on pachytene chromsorne analysis. The other Cajanus species for which chromsome numbers were reported are C. platycarpus (Bir and Kumari, 1973, 1977; Pundir and Singh, 1981); C. rugosus (Sanjappa and Satyananda, 1979); C. volubilis (Rao, 1978 and Pundir and Singh, 1978); C.trinervius (Pundir and Singh, 1978); C. albicans (Rao, 1978 and Pundir and Singh, 1978) and C. caianifolius (Pundir, 1981).

The first report on inter-specific hybridization in pigeonpea dates back to 1956 when Deodikar and Thakur crossed C. cajan with C. leneatus. The hybrid was fairly fertile. Kumar et al. (1958) extended the earlier work on to hybrid cytology and found regular bivalent formation in the hybrid. A hybrid between C. cajan and C. scarabaeoides was obtained by Roy and De (1967) who expressed doubts about the generic status of Cajanus. Reddy (1973) analysed pachytene chromosome pairing in C. cajan, C, leneatus, C. scarabaeoides, C. sericeus and their hybrids. These pachytene studies in general revealed a high degree of chromosome homology between cultivated and the three wild species of Cajanus. Studies on the inheritance of a few qualitative traits were also done in this study. Ariyanayagam and Spence (1978) reported hybrids between Cajanus and C. platycarpus while further attempts (Pundir, l98l; Reddy et al., 1980 and Kumar et al., 1985) to cross Cajanus with C. platycarpus failed. Further attempts at wide hybridization by Pundir (1981) involved karyotype comparisons between the cultivated and the wild species and meiotic pairing in the F1 hybrids. These studies revealed a great degree of karyotypic similarities between species. All the studies on cultivated X wild hybridization revealed a close relationship between the species of the two and regular pairing in their hybrids which nevertheless exhibited a fair degree of sterility.

Dundas et al. (1987) studied the meiotic behavior of hybrids of pigeonpea and two Australian native Cajanus species. They found that meiosis in the parental pigeonpea and wild accessions appeared regularly, while the hybrids showed a lower frequency of ring bivalents at Metaphase I. Multivalents, univalents, rod bivalents, chromatin bridges, fragments, laggards, and supernumerary microspores occurred in meiotic cells of hybrids.

Mallikarjuna and Moss (1995) studied the meiotic behavior of F1 hybrids derived from the cross C. platycarpus and C. cajan to check for the cause of pollen sterility. Mallikarjuna and Saxena (2002) found that the interspecific hybrid seed obtained by cross between C. acutifolius and C. cajan were semi shriveled. Very few seeds germinated to give rise to F1 plants. Backcrossing the hybrid plants are obtained after saving the aborting embryos in vitro. BC1 plants showed normal meiotic pairing, but had low pollen fertility. The reason for embryo abortion and low fertility in spite of normal meiosis could be due to effects of wild species cytoplasm. Mallikarjuna and Kalpana (2004) reported two types of CMS plants by studying the meiosis of sterile lines. Type I CMS plants had partially or totally brown shriveled anthers and the process of microsporogenesis was inhibited at the pre-meiotic stage, while Type II CMS plants had pale white shriveled anthers and the breakdown of microsporogenesis was at post meiotic stage after the formation of tetrads which caused male sterility of the plants. Also, Type II CMS system reported in sterile hybrids derived from the C. acutifolius by Mallikarjuna and Saxena (2005).

Mallikarjuna et al. (2011) studied the meiosis of F1 hybrids from the crosses ICPL 85010 X ICPW 29 (C. cajanifolius) and ICPL 85010 X ICPW 31 (C. cajanifolius). Meiocytes showed the formation of 11 bivalents. Twenty percent of the meiocytes showed the formation of 7 bivalents and 2 tetravalents showing two chromosomes which are totally homologous between the parental species. In the cross ICPL 85010 X ICPW 28, 10% of the meiocytes showed the presence of 2 univalents which signifies that one chromosome in each parent did not have a homologous chromosome in the other parent or the divergence of one chromosome in one of the parent. Such an anomaly was present only in 10% of the meiocytes. Tetrads were observed in both the crosses and pollen fertility in the F1 hybrids varied between 48–62%, showing closer relationship between the two species. Mallikarjuna et al. (2011b) also reported Type II CMS system in the male sterile lines derived from the C. platycarpus.

Cytological Studies on Male-Sterile Genotypes

A number of bio-chemical changes are responsible for the development of pollen mother cell (PMC) from the meristematic tissues. Further, it is followed by a series of developmental changes which lead to the mature pollen grains. In the determination of male-sterility in crop plants, the anther wall and in particular the tapetum, plays an important role of producing and transporting critical enzymes, hormones and nutrients that are essential for the growth of PMCs and any abnormality in the anther wall development leads to the production of defective pollen grains. During the process of meiosis any abnormality in the supply of nutrients generally leads to aberrant outputs such as large and more PMCs (Vasil, 1967). The fusion of cells into multi-nuclear syncytia or abnormal vacuolization or degeneration of the tapetal layer leads to the abnormal development and separation of PMCs. The normal development of PMCs in general is arrested either pre-meiotic, during meiosis, or in post-meiotic stages of growth.

The cytological studies on the fertile and sterile siblings showed that the microsporogenesis in the two genotypes was similar up to tetrad formation stage. The differences between the two emerged when the tetrads in the male-sterile plants failed to be released, leading to degeneration of tetrads through vacuolation. The tapetum continued to persist even when the tetrads degenerated. On the contrary, in the fertile plants, tapetum began to degenerate during the formation of tetrad and disappeared during male gametophyte development. In case of male-sterility the callose is synthesized due to the presence of high concentrations of cellular calcium (Worral et al., 1992). Further studies conducted on the persistence of callose and tapetum in the ms1 type of male- sterility concluded the accumulation of callose and persistent tapetum during post-meiotic stages (Ketti et al., 1994). They further deliberated that a gradual reduction in the concentration of polysaccharides and RNA proteins in the tetrads were responsible for disorientation of cytoplasm leading to malnutrition and poor tetrad growth. The degeneration of microspores occurred at the tetrad stage through rupturing of nuclear membrane and resulting in to collapse of the outer wall (Dundas et al., 1981). While reporting a new source of GMS, in the male-sterile plants the PMCs count was almost double than their fertile counterparts (Dundas et al., 1982). The abnormal enlargement of PMCs and their number was associated with the failure of adjacent PMC walls to separate.

The breakdown of microsporogenesis of this male-sterile occurred at prophase I. The delayed and incomplete anther wall development appeared to be responsible for PMC degeneration. Similar observations were also reported in cotton (Murthi and Weaver, 1974). In the male-sterile plants, however, PMC degeneration occurred at young tetrad stage with the rupturing of nuclear membrane and collapse of outer cell walls. The vacuoles developed in the tapetal cells metaphase I and by tetrad stage the entire cell gets vacuolated. In this case, the precocious degeneration of tapetum ending its role as a nutrient source for PMCs could be responsible for tetrad breakdown (Echlin, 1971). Similar results were also reported in Hordeum vulgare (Kaul and Singh, 1966); Sorghum (Overman and Warmke, 1972) and Pennisitum typhoides (Reddy and Reddy, 1974).

In all the three GMS systems the blockages in the microsporogenesis occurred at different stages of development which also determined their anther morphology. Studies showed that if an individual plant carries two male-sterility inducing genes, then the one which expresses first and hinders the normal process of microsporogenesis, determines the phenotype of the anthers and the other genes become redundant as far as their expression is concerned (Saxena et al., 1983; Saxena and Kumar, 2001). Cytological examination of sparse pollen producing flowers revealed that their tetrad formation was normal but soon after this, only a portion of microscopores collapsed. Further, the locules of anthers within individual flowers varied in the proportions of microspore degeneration (Saxena et al., 1981). The cause of this partial breakdown of microsporogenesis could not be ascertained. Ariyanayagam et al. (1995) working with a C. sericeus derived CMS lines, reported that meiosis in the male-sterile plants proceeded normally until the release of microspores and this was followed by vacuolation and degeneration of protoplasm. Cytological investigations with C. acutifolius derived CMS showed that the process of meiosis in the male-sterile plants proceeded normally till the onset of tetrad stage but their further growth was arrested and the tetrads remained inside the tapetum layer. This resulted in the loss of cell contents and collapse of the process of microsporogenesis (Mallikarjuna and Saxena, 2005).

A detailed study identified two different kinds of male-sterile plants in a cross involving a cultivated pigeonpea as female parent and C. acutifolius as male parent (Mallikarjuna and Kalpana, 2004). These two male-sterile variants had different anther morphology. In Type I, the anthers were shriveled with brown color, while in Type II male - steriles, the plants had pale white shriveled anthers. These variants also differed in

32 their microsporogenesis. The PMCs of Type I male-sterile plants remained in prophase stage and subsequent processes of meiosis were arrested. The PMCs enlarged normally and once nucleus grew, further cell division did not take place. In these plants persistence of tapetum was also observed. In Type II plants, the anthers were translucent and microsporogenesis continued up to tetrad stage but the tetrads failed to separate and produce pollen grains. This was followed by collapse of anther development process, a sort of post-meiotic arrest of microspore development. Cytogenetic studies of A4 CMS revealed an early breakdown of tapetum. In these plants the anthers were underdeveloped and the male-sterility expressed at tetrad stage, where the tetrad wall failed to degenerate and resulted in the degeneration of its contents (Dalvi et al., 2008b).

It can be concluded that two primary processes are responsible for producing GMS in pigeonpea. The first process is characterized by the development of brown and shriveled anthers followed by premeiotic breakdown of PMCs. In the other process the anthers are pale white or translucent accompanied by post-meiotic breakdown of PMCs.

It has been widely assumed that the CMS trait is expressed due to impairment of pollen formation processes that result from interaction of the nuclear and the mitochondrial genomes. Pollen maturation requires more amounts of energy (Zhao et al., 2000). This is evident by the dramatic increase in the number of mitochondria in the tapetal tissue and PMCs during pollen development. In sugar beet and wheat, low temperatures cause CMS like microspore disturbances as microspores and tapetum cells are more sensitive than the female reproductive organs and oxidative processes are responsible for this development (Kuranouchi et al., 2000). It is also believed that the mitochondria have a major role to play in the expression of CMS trait. In pigeonpea there is only one report that deals with the assessment of mitochondrial genome of the CMS plants (Sivaramakrishnan et al., 2002).

Assessment of hybrid purity using SSRs

The use of DNA markers particularly Simple Sequence Repeats (SSRs) are useful for a variety of molecular breeding applications because of their codominance, abundance, high genome coverage and multi allelic nature. SSRs are the most suitable markers for hybrid purity assessment as the heterozygosity of the hybrids can be easily determined by the presence of alleles from both the parents used for controlled pollination. SSRs are highly polymorphic and are becoming the marker of choice in both

33 animal and plant species (Condit and Hubell, 1991; Akkaya et al., 1992; Morgante and Oliveri, 1993; Wang et al., 1994). Bohra et al. (2011) reported that a total of 3,072 novel SSR primer pairs were synthesized and tested for length polymorphism on a set of 22 parental genotypes of 13 mapping populations segregating for traits of interest. Utility of developed SSR markers was also demonstrated by identifying a set of 42 markers each for two hybrids (ICPH 2671 and ICPH 2438) for genetic purity assessment in commercial hybrid breeding program. They are already a proven tool for hybrid authentication or hybrid purity assessment and parentage confirmation in many crop species (Bohra et al., 2011). Indeed, molecular markers-based hybrid purity tests have been developed and are in routine use in many crop species such as rice (Yashitola et al., 2002; Sundaram et al., 2008), maize (Asif et al., 2006), cotton (Selvakumar et al., 2010) and safflower (Naresh et al., 2009).

Biochemical Diversity

Proteinase inhibitors (PIs)

The discovery of Proteinase inhibitors (PIs) in plants affecting insect digestive enzymes called attention to the possibility of using these enzymes as targets in the development of new insect control techniques (Ryan, 1990). Proteinase inhibitors are ubiquitous small proteins that are quite common in nature. They are natural, defense- related proteins often present in seeds and induced in certain plant tissues by herbivory or wounding (Koiwa et al., 1997; Browse and Howe, 2008). Proteinase inhibitors are present in multiple forms in numerous tissues of animals and plants as well as in microorganisms. In plants they can be counted among the defensive mechanisms displayed against phytophagous insects and microorganisms. The defensive capacities of plant PIs rely on inhibition of proteases present in insect guts or secreted by microorganisms, causing a reduction in the availability of amino acids necessary for growth and development (De Leo et al., 2002). Protein and peptide inhibitors of various exogenous (from invertebrates, viruses, fungi, and mammals) and endogenous proteinases are widespread in seeds. Proteinase inhibitor II (PIN2), is a serine proteinase inhibitor with trypsin and chymotrypsin inhibitory activities (Bryant et al., 1976; Lawrence and Koundal, 2002), and occurs in many solanaceaous plants, including tomato (Gustafson and Ryan, 1976), potato (Bryant et al., 1976), and tobacco (Pearce et al., 1993). A successful approach was to produce transgenic plants that express, trypsin inhibitors to provide resistance to

34 insects (Hilder et al., 1987). This approach will benefit from knowledge gained on the properties (which differ among insect groups) and midgut distribution (which determine if target enzymes are or not accessible to inhibitors) of insect digestive enzymes.

Plants use proteins as a part of their defense strategies. An interesting class of defense protein is the inhibitors of digestive enzymes that occur in many plants. The two main classes of inhibitors discovered so far are the protease inhibitors and the amylase inhibitors. Among them, protease inhibitors play an important role in defense of plants against herbivorous insects. They act as competitive inhibition of enzymes by binding tightly to the active site of the enzyme. The anti-metabolic activity of the protease inhibitors is due to direct inhibition of the larval proteolysis and utilization of proteins leading to the death of larvae by slow starvation.

Proteinase inhibitors are widely distributed in the plant kingdom, particularly in seeds and tubers, where they often represent several percent of total protein (Liener and Kakade, 1969; Ryan, 1973; Richardson, 1977). They have been most extensively studied in Leguminoseae, Graminae and Solanaceae, presumably because of the large number of species in these families (Richardson, 1977). According to specificity, PIs can be divided into four classes, inhibiting serine, cysteine, metallo or aspartyl proteases. Several non- homologous families of protease inhibitors are recognized among the animal, microorganism, and plant kingdoms (Laskowski et al., 1980). In plants about ten protease inhibitor families have been recognized (Garcia et al., 1987). Members of the serine and cysteine proteipse inhibitors have been more relevant to the area of plant defense than metallo- and aspartyl PIs, since only a few of these latter two families of inhibitors have been found in plants.

Plant protease Inhibitors (PPIs)

Protease inhibitors (PIs) are of common occurrence in the plant kingdom. Plant PIs (PPIs) are generally small proteins that have mainly been described as occurring in storage tissues, such as tubers and seeds, but they have also been found in the aerial parts of plants (De Leo et al., 2002). They are also induced in plants in response to injury or attack by insects or pathogens (Ryan, 1990). In plants, these PIs act as anti-metabolic proteins, which interfere with the digestive process of insects. One of the important defense strategies found in plants in combating predators involves PIs which are particularly effective against phytophagous insects and microorganisms. The defensive

35 capabilities of PPIs relies on inhibition of proteases present in insect gut or secreted by microorganisms, causing a reduction in the availability of amino acids necessary for their growth and development (Lawrence and Koundal, 2002; Chen et al., 2007). The plant protease inhibitors apparently may be involved in several processes, i.e., reveal multifunctional properties. At present, a multitude of biochemical pathways in which protein inhibitors may be involved is of particular interest with respect to the design of transgenic plant species containing proteinase inhibitor genes (Ryan, 1990).

Protease inhibitors have been grouped into families and subfamilies, and into different clans on the basis of sequence relationship and the relationship of protein folds of the inhibitory domains or units. An inhibitor domain is defined as the segment of the amino acid sequence containing a single reactive site after removal of any parts that are not directly involved in the inhibitor activity. On the basis of sequence homologies of their inhibitor domains, PIs have been classified into 48 families (Rawlings et al., 2004). Proteins containing a single inhibitor unit are termed simple inhibitors, and those that contain multiple inhibitor units are termed complex inhibitors. A total of 11 families belong to the latter category and contain between 2 and 15 inhibitory domains. Most of these are homotypic, containing inhibitor units from a single family, some are, however, heterotypic, and contain inhibitor unit from different families (Richardson et al., 2001; Trexler et al., 2001, 2002).

Most of the studies have been carried out on crop plants viz., cereals, legumes, and solanaceous species. Soybean trypsin inhibitor was the first PI isolated and characterized. Since then many PIs have been found to be widely distributed throughout the plant kingdom. Most of plants PIs that have been characterized are from the Gramineae (Poaceae), Leguminosae (Fabaceae), and Solanaceae families (Brzin and Kidric, 1995).

PLANT-PIs are a database developed to facilitate retrieval of information on the distribution and functional properties of protease inhibitors in higher plants. Currently, plant PIs, contain information for 495 inhibitors (plus several iso-inhibitors) identified in 129 different plants (De Leo et al., 2002). PIs are usually found in storage organs, such as seeds and tubers, but their occurrence in the aerial part of plants, as a consequence of several stimuli has also been widely documented (De Leo et al., 2002). Proteinase inhibitors may accumulate to about 1 to 10% of the total proteins in these storage tissues.

An increasing number of PIs is found in non-storage tissues, such as leaves, flowers, and roots (Brzin and Kidric, 1995; Xu et al., 2001; Sin and Chye, 2004). Some PIs also occur in yeast (Matern et al., 1979) and other fungi (Richardson, 1977). A trypsin inhibitor was found localized in the cytosol of green gram (mung) bean cotyledonary cells (Chrispeels and Baumgartner, 1978). Soybean trypsin inhibitor (SBTI) is mainly present in the cell walls, with lesser amounts in protein bodies, the cytoplasm, and the nuclei of cotyledonary and embryonic cells. Soybean Bowman-Brik inhibitor (SBBI) occurs in protein bodies, the nuclei, and to a lesser extent the cytoplasm. In contrast to SBTI, some SBBI has been located in the intercellular space but not in the cell wall (Horisberger and Tacchini-Vonlanthen, 1983). The wound-induced inhibitors accumulate in vacuoles of tomato, wild tomato, and potato leaves. Xu et al. (2004) described the expression of a PIN2 protein from Solanum americanum Mill. in phloem of stems, roots, and leaves, suggesting a novel endogenous role for PIN2 in phloem. Further research showed that both SaPIN2a and SaPIN2b are expressed in floral tissues (Sin and Chye, 2004).

Classification of plant PIs

Proteinase inhibitors are classified based on the active site residues involved in hydrolysis as serine, cysteine, aspartic, and metallo proteinases (Barrett, 1999). Proteinase inhibitors have been grouped into families and subfamilies on the basis of sequence relationship and the relationship of protein folds of the inhibitory domains or units. An inhibitor domain is defined as the segment of the amino acid sequence containing a single reactive site after removal of any parts that are not directly involved in the inhibitor activity. On the basis of sequence homologies of their inhibitor domains, PIs have been classified into 48 families. Proteins containing a single inhibitor unit are termed as simple inhibitors. Proteins that contain multiple inhibitor units are termed complex inhibitors, containing between 2 and15 inhibitory domains. Most of these are homotypic, containing inhibitor units from a single family, some are however heterotypic, and contain inhibitor units from different families. On the basis of tertiary structure, 31 families have been assigned to 26 clans, indicating that a large proportion of families show no relationships in their three dimensional structures. In the past, however PIs have been classified into serine, cysteine, aspartate and metallocarboxy PIs (Laing and McManus, 2002).

Localization of PIs

Proteinase inhibitors are synthesized as pre and pro proteins which undergo further processing for activation in vivo. Furthermore, they have been sub-cellularly localized to various compartments in the plant cell. soybean Kunitz trypsin inhibitor SKTI was reported to be associated with the cell walls and to a lesser extent with protein bodies, cytosol, and nuclei, while the Bowman Birk Inhibitior (BBI) has been localized in protein bodies, nuclei, and cytosol (Horisberger and Tacchini-Vonlanthen, 1983).

Functions of PIs:

 Endogenous functions: Initially, it was thought that many plant PIs do not have endogenous functions against plant proteases but show specificities for animal or microbial enzymes (Reeck et al., 1997). Evidences suggested that infection or injury triggers production of inhibitors occurs through octadeconoid pathway, which catalyses the breakdown of linolenic acid and the formation of jasmonic acid to induce protease inhibitor gene expression. Subsequently, accumulating evidences which supported the developmental regulation and tissue-specific accumulation of plant PIs suggested for their endogenous functions (Lorberth et al., 1992). PIs regulate endogenous proteinase levels before and during seed germination for storage protein digestion and to control protein turnover (Ryan, 1981). Arg-1, a proteinase isolated from Ipomoea batatas was completely inhibited by sweet potato trypsin inhibitor (Hou and Lin, 2002). Proteinase inhibitor II (PIN2) from Solanum americanum not only enhanced resistance to caterpillars when expressed exogenously, but also inhibited endogenous proteases that are expressed during seed development (Chye et al., 2006).

 To store proteins: The PIs serving as storage proteins in plants were identified by the presence of high content of inhibitors in seeds and other storage organs of plants and of their dynamics in the course of seed maturation and germination (Richardson, 1977). Similar to other storage proteins, inhibitors are located in vacuoles and protein bodies (Wingate et al., 1991).

 Protective agents: The possible role of PIs in plant protection was investigated as early as 1947 when Mickel and Standish observed that the larvae of certain insects were unable to develop normally on soybean products. Subsequently the trypsin

inhibitors present in soybean were shown to be toxic to the larvae of flour beetle, Tribolium confusum (Lipke et al., 1954). Following these early studies, there have been many examples of PIs active against certain insect species, proved by both in vitro assays against insect gut proteases (Giri et al., 2003) and in vivo artificial diet bioassays (McManus and Burgess, 1999; Telang et al., 2005). On the other hand, the PI gene expression has been detected in leaves of several species following wounding caused by herbivory. These studies suggest that PIs play a role in protecting plants from insect attack and microbial infection (Ryan, 1990).

 Modulation of Apoptosis: Cysteine proteinase plays an important role in the regulation of programmed cell death leading to hypersensitive (HR) reaction, following pathogen attack. It has been shown that the ectopic expression of cystatin inhibits the induced cysteine proteinase activity, which in turn blocks PCD (Solomon et al., 1999). It was suggested that in plants balancing between the cysteine proteinase and cysteine PI activity regulates the programmed cell death. Leguminaceae because of their important function as high protein source, high amount of proteases are present, which in turn are controlled by proteinase inhibitors present in high amounts in this family. The protease inhibiting proteins especially those inhibiting trypsin and chymotrypsin have been thoroughly studied in Leguminaceae.

Serine protease inhibitors (Serpins)

Serine protease inhibitors are universal in the plant kingdom and have been described in many plant species. Therefore, the number of known and partially characterized inhibitors of serine proteases is enormous (Haq et al., 2004). Serine protease inhibitors have been reported from a variety of plant sources and are the most studied class of protease inhibitors (Mello et al., 2002; Haq and Khan, 2003). Chiche et al., (2004) first introduced the squash inhibitor, a well-established family of highly potent canonical serine protease inhibitors isolated from Cucurbitaceae.

Plant serine PIs (serpins) have been shown to inhibit model trypsin like proteins (Roberts et al., 2003), but there are no obvious targets for these inhibitors in plants which may be involved in inhibiting proteases of plant pathogens (Hejgaard, 2005). Plant serpins have molecular mass of 39 - 43 kDa, with amino acid and nucleotide homology with other well-characterized serpins. The majority of serpins inhibit serine proteases, but

39 serpins that inhibit caspases (Ray et al., 1992; Law et al., 2006) and papain like cysteine proteases (Schick et al., 1998; Irving et al., 2002) have also been reported. Plant serpins exhibit differing and mixed specificities towards proteases (Al-Khunaizi et al., 2002). Barley serpin (Hordeum vulgae) is a potent inhibitor of trypsin and chymotrypsin at overlapping reactive sites (Dahl et al., 1996). The inhibitors from at least four families belonging to serine PIs have been induced sequentially in various plants. These inhibitor families include potato (Solanum tuberosum) and tomato (Lycopersicon esculentum) inhibitors I and II in solanaceous plants (Melville and Ryan, 1972; Bryant et al., 1976; Plunkett et al., 1982; Valueva et al., 2001; Farran et al., 2002), Bowman-Birk inhibitors in alfa alfa (Brown and Ryan, 1984) and a Kunitz inhibitor in poplar trees (Bradshaw et al., 1989).

The serpin family of PIs active against serine proteases also contains inhibitors of cysteine proteases (Heibges et al., 2003; Laskowski et al., 2003). Serine PIs belonging to various families have been reported either in storage organs or in the vegetative cells of a wide variety of plants (Garcia et al., 1987). Two oat (Avena sativa) serpins show specificity for chymotrypsin and / or elastase, and another one has specificity for trypsin and chymotrypsin at overlapping loop sites (Irving et al., 2002). Squash serpin Cmps-1 also inhibits elastase at two overlapping sites (Ligoxygakis et al., 2003). Serpins are irreversible inhibitors. The cleavage of an appropriate peptide bond in the reactive centre loop of the inhibitor triggers a rapid conformational change so that catalysis does not proceed beyond the formation of an acyl-enzyme complex (Huntington et al., 2000).

Bowman Birk inhibitor family

The Bowman Birk inhibitor (BBI) family is one of the most widespread groups of serine PIs, and is particularly abundant in legume seeds. The soybean inhibitor is the most-studied member of this family and is often referred as the classic BBI. BBIs have been classified on the basis of their structural features and inhibitor characteristics. The inhibitors from dicotyledonous plants consist of a single polypeptide chain with the molecular mass of 8 kDa. These are double-headed, with two homologous domains each bearing a separate reactive site for the cognate proteases. These inhibitors interact independently and simultaneously, with two proteases, which may be same or different (Birk, 1985). The first reactive site in these inhibitors is usually specific for trypsin, chymotrypsin and elastase (Qi et al., 2005).The active site configuration in these

40 inhibitors is stabilized by the presence of seven conserved disulfide bonds. BBI from legumes such as Glycine max and pea (Pisum sativum), are proteins that consist of a single polypeptide chain with molecular masses in the range of 6-9 kDa (Clemente and Domoney, 2006). They usually comprise two distinct binding loops and are responsible for the inhibition of two enzyme molecules that may be same or different enzymes (Singh and Appu Rao, 2002).

The BBIs from monocotyledonous plants are of two types. One group consists of a single polypeptide chain with a molecular mass of about 8 kDa. They have a single reactive site. The other group has a molecular mass of 16 kDa with two reactive sites (Prakash et al., 1996). It has been suggested that larger inhibitors have arisen from smaller ones by gene duplication (Odani et al., 1986). In the case of double-headed BBIs, it has been found that the relative affinity of binding of proteases is altered when one site is already occupied. In the same way, the activity of soybean BBIs decreases 100 fold when trypsin is bound at the other site (Gladysheva et al., 1999).The BBI family of protease inhibitors contains a unique disulfide-linked nine-residue loop that adopts a characteristic canonical conformation (Bode and Hubr, 1992). The loop is called protease-binding loop and binds the protease in a substrate-like manner. The BBIs are cysteine-rich proteins with inhibitory activity against proteases that are widely distributed in monocot and dicot species. They have been shown to act as anti-carcinogenic compounds. The soybean derived BBI with a well-characterized ability to inhibit trypsin and chymotrypsin is particularly effective in suppressing carcinogenesis in a variety of in vivo and in vitro systems (Kennedy, 1998).

Various reports indicate the presence of several isoforms (iso-inhibitors) of trypsin and chymotrypsin inhibitors from the seed extracts of red gram (Kollipara et al., 1994; Pichare and Kachole, 1994, 1996; Chougule et al., 2003). Pichare and Kachole (1994) reported the presence of nine trypsin and seven chymotrypsin inhibitors in pigeonpea seed extracts based on the X-ray film contact print technique. Positive correlation existed between the trypsin inhibitor and chymotrypsin inhibitor activities in pigeonpea (Kollipara et al., 1994). Further, the pattern of these trypsin and chymotrypsin inhibitors were known to change during seed germination and seed development (Godbole et al., 1994; Ambekar et al., 1996). Chougule et al., (2003) reported that PIs from some cultivars and wild types of pigeonpea were active against H. armigera gut proteinases.

Biology, nature of damage and management of Helicoverpa armigera

Helicoverpa armigera is a legume pod borer insect that has world-wide agricultural importance. It feeds on a wide range of wild and cultivated host plants. The larvae, particularly the later instars, feed on the reproductive parts of the plant. In India, it is a dominant pest on cotton, pigeonpea, and chickpea. On pigeonpea and chickpea, it usually destroys more than half of the grain yield. The biological characteristics such as high fecundity, extensive polyphagy, strong flying ability and facultative diapauses contribute to the devastating pest status of H. amigera (Fitt, 1989). The ability to feed on various plants enables H. armigera populations to develop continuously during the entire cropping season (Bhatnagar et al., 1982).

The biology of H. armigera is typical of noctuidae. Morphology of the various life stages has been described by Pearson (1958), and Zalucki et al. (1986). Distinguishing features have been described by Dominguez Garcia- Tejero (1957), Hardwick (1965), King (1994) and Mathews (1999). The females lay eggs singly, on the upper surface of the leaves along the midrib, flowers, pods and stems. The number of eggs per female ranges from 387 to 1364 on different host plants (Dhandapani and Subramaniam, 1980). The eggs are white and nearly spherical when freshly laid, and darken with age. Eggs typically hatch in 2-5days. Larval duration varies from 8 to 28days (Singh and Singh, 1975), and there are 5 to 7 larval instars, which vary with temperature and the host plant. Pupation takes place in soil, and the adults emerge in 7 to 10days. One generation can grow in just over 4weeks under favorable conditions. The number of generations varies according to agroclimatic conditions. It passes through four generations in Punjab (Singh and Singh, 1975), seven to eight generations in Andhra Pradesh (Bhatnagar, 1980), and five generations in Uttar Pradesh, India (Tripathi, 1985).

The Lepidopteran borers viz., H. armigera. Exelastis atomosa, Maruca testualalis and dipteran pod fly, Melanagromyra obtusa have been reported as the most damaging at the reproductive phase of the plants (Singh and Singh, 1978). Insecticides have been an important component of the management strategy for pigeonpea pests, especially borers.

The young larvae of H. armigera feed by scraping green tissues and by nibbling various parts of the plant until they find a flower bud or flower; the bud is quickly hollowed out, leaving an empty shell. In pigeonpea and chickpea, the older larvae chew voraciously into the buds, flowers, and pods, leaving characteristic round holes. In canon,

42 older larvae feed on the buds and young bolls. The larvae habitually feed with only the front portion of its body inside the hole thus commonly showing an accumulation of larval faces between the surface and the enclosing bracteoles.

The estimated crop losses due to H. armigera vary in different countries viz., US $ 600million in chickpea and pigeonpea per annum in semi-Arid tropics (ICRISAT, 1992); A$ 16million in 1979 (Alcock and Twine, 1981), A $23.5million (Wilson, 1982), and A $25million annually in Australia (Twine, 1989). In India, losses in pulses, chickpea and pigeonpea were estimated at over $ 300million per annum (Reed and Pawar, 1982).

Helicoverpa armigera is a pest of major importance in most areas, damaging a wide variety of food, fiber, oilseed, fodder and horticultural crops. It is mainly characterized by mobility, polyphagy, high reproductive rate, and diapause, all of which make it well adapted to exploit transient habitats such as man-made agro-ecosystems. The natural control means, chemical or the integrated control methods need to be adopted to minimize the losses due to this pest. The key issue of moth immigration and movement in general is of relevance to the long-term effectiveness of any control strategy aimed to suppress more than one generation. It may have little value if crop infestation is mainly by the immigrant moths of distant origin.

Integrated pest management strategies for Helicoverpa require an integration of different control tactics to implement a threshold based on the relationship between population density and economic loss. It is often difficult to obtain precise data on relationship because many extraneous factors, both environmental and socio-economic, influence it. Tactics that have been evaluated against Helicoverpa include cultural manipulation of crop and its environment, biological control including the use of microbial pesticides, sex pheromones for population monitoring or mating disruption, sterile backcross techniques, chemical control and host plant resistance.

Evaluation for pod borer resistance

Most of the screening for host plant resistance to H. amigera has been carried out at ICRISAT (Lateef and Pimbert, 1990). In general, determinate genotypes show greater susceptibility to pod damage by H. amigera than indeterminate types (Kushwaha and Malik 1987; Lateef and Reed, 1990). One of the reasons for high susceptibility of determinate type genotype to H. armigera may be due to cluster type of flowering making

43 it easier for larvae to move from one pod to another. Within short duration determinate types, ICPL 289 and H 81-95 (Kushwaha and Malik, 1987) have shown less susceptibility to pod borer (Dahiya et al., 2001). Among the medium-duration types, most of genotypes have indeterminate growth habit, and genotypes ICP - 909 - EB, PPE - 45-2, ICP 1811- E3, ICP 1903 - E (ICPL 332), and ICP 10466 - E3 have shown less susceptibility to pod borer (Lateef and Pimbert, 1990). Short duration varieties (150days) are safer from pod borer than extra early varieties (Singh, 1996).

Several workers have reported serious lepidopteran borers damage on determinate clustering and early and medium maturing pigeonpea cultivars (Lateef and Reed, 1980; Reddy et al., 1973; Yadava et al., 1988). Mali and Patil (1993) reported minimum percentage of damage due to H. armigera on variety T-21 (8.98), Sehore – 68 (12.07%) and maximum damage on variety, ICPL-87 (32.77%) in field screening of some pigeonpea varieties against pod borers.

Some of the wild relatives of pigeonpea have shown high levels and biochemical components of resistance to H. armigera (Sharma et al., 2009). Mallikarjuna et al. (2011) reported the damage due to H. armigera, in the wild parent C. platycarpus, was less than 1%. Damage in cultivated parent ICPL 87 was 69%. Damage in F1BC4-A derivatives ranged from 2 to 37% with majority of the lines with less than 15% damage.

The eggs of H. armigera are nearly spherical with a flattened base, and are laid singly. The larva leaves the plant in 3 weeks or less, and bores into the soil to a depth of 1.5 to 2.5 cm, where it pupates. The pupa is 14 to 18 nm long, mahogany, brown, smooth surface, and rounded both anteriorily and posteriorly and has two taperings and parallel spines at the posterior tip. The medium sized brown moths emerge from the soil in about 2 weeks. Adult females are larger and stouter than males. The life cycle will be completed in little more than a month. Female moths live longer than males and can lay more than 1000 eggs causing rapid infestations (Reed et al., 1989). In some cases, more than 3000 eggs per female have been reported, though fecundity in the range of 1000 -2000 is common (Reed, 1965).

In India, three species of Helicoverpa, H. armigera, H. peltigera Schiff and H. assulta Guenee have been recorded, of which H. armigera is the most important. H. armigera passes through four generations in Punjab. The first is on chickpea during March, the second on tomato from end of March to May, third on maize and tomato

44 between July to August (Singh and Singh, 1975). Bhatnagar (1980) reported seven to eight generations of H. armigera in Andhra Pradesh. Oviposition usually starts in early June, with the onset of pre-monsoon showers. Adults possibly emerge from the diapausing pupae and from the larvae on summer crops and weeds. The pre-oviposition period range from 1 to 4days. Oviposition period last 2 to 5days, and post oviposition period is 1 to 2days (Patel et. al., 1968; Singh and Singh, 1975).

The preferred host plants for oviposition by H. armigera were studied by Vijayakumar and Jayaraj (1982) and found to be in descending order as pigeonpea > field bean > chickpea > tomato > cotton > chillies > mungbean > sorghum. Reddy (1973) and Loganathan (1981) reported that pigeonpea was the preferred host for oviposition. The feeding preference descending order was pigeonpea > field bean > cotton > sunflower > sorghum > chickpea > mung bean > urd bean > tomato. The larval period was maximum in tomato and minimum in pigeonpea and ranged from 17 to 20days (Dhandapani and Balasubramanian, 1980). The pupal stage ranged from 10.5 to 13.6days being minimum on pigeonpea and maximum on sorghum, maize and sunflower.

There are several factors associated with the population build up of H. armigera. It is speculated that an increase in irrigation in South India has led to availability of host plants throughout the dry season, and resulted in subsequent increase in pest population (Reed and Pawar, 1982). Helicoverpa armigera undergoes facultative diapause during December to February in North India. As a result, the pupal period lasts for more than 100days. The prolonged pupal period leads to the low population build up during last leg of winter season resulting in the non-availability of larval parasitoids. Jadhav et al., (2012) evaluated over 1200 C. acutifolius derived lines for pod borer damage, which ranged from 0-60%. Around 85% of the lines suffered <10% pod damage, which was significantly lower as compared to the susceptible check, ICPL 85010.

Mechanisms of resistance

The mechanism of resistance needs to be understood for any genetic enhancement programme. An empirical approach was proposed by Painter (1941 and 1951). Painter's proposed mechanism of resistance was grouped into three main categories 1) Non- preference is avoidance of insect by plants and is often projected as a property of the plant. For this reason Kogan and Ontman (1978) proposed to substitute antixenosis for the term 'non-preference'. It is a parallel term to 'antibiosis' and conveys the idea that the plant

45 is avoided as a bad host. 2) Antibiosis includes all adverse effects exerted by the plant on the insect's biology including development, reproduction and survival. 3) Tolerance includes all plant responses resulting in the ability to withstand infection and to support insect populations that would severely damage susceptible plants. Plant physical characters are prime factors to be considered for host plant resistance (Southwood, 1986). Biochemical (Isoprenoids, acetogenins, aromatic derived from shikimic acid and acetate, alkaloids, protease inhibitors and non protein aminoacids and glycosides) and morphological basis of resistance (thickening of cell walls, rapid proliferation of plant tissues, toughness of stem, trichomes effect on feeding and digestion, on oviposition, as a mechanical barrier to locomotion, attachment, association with allele chemical factors, incrustation of minerals in cuticles, surface waxes and anatomical adaptations of organs) were reported .

Antibiosis

Antibiosis, one of the three types of mechanisms of resistance proposed by Painter (1951) is described as those adverse effects on the insect life history when a resistant host plant variety is used as food. The adverse effect on the insect can be in the form of reduced fecundity, decreased size, abnormal shortened life and/or increased mortality. Antibiotic effect (Dodia et al., 1996) of C, scarabaeoides, C. cajanifolius, C. reticulalus, C. sericeus and Fl hybrids of cross (C. scarabaeoides X C. cajan), and the cultivated pigeonpea lines were observed on H. armigera. The results of the study clearly showed that the larval and pupal mass fed on wild pigeonpea flowers and F1 hybrid was significantly lower than those larvae fed on the cultivated pigeonpea. The developmental period of larvae fed on wild pigeonpea flowers was longer than those fed on the cultivated pigeonpea flowers. Similarly, pupal size of larvae fed on the F1s and the wild species was significantly reduced compared to the cultivated pigeonpeas. Growth index and larval fecundity were adversely affected for larvae reared on the wild species and F1s. The studies on the hybrid progenies of a C. cajan x C. scarabaeoides cross by Verulkar et al. (1997) suggested that the antibiosis mechanism of resistance was governed by single dominant gene.

Evaluation of pigeonpea genotypes for Bruchid resistance

Wadnerkar et al. (1978) studied the development of Callosobruchus maculatus Fab on four varieties of Arhar and three varieties of gram. Arhar was more preferred to

46 gram for oviposition and Arhar No. 148 was the most preferred variety. Development was significantly faster in white gram. Prabhat variety of Arhar showed maximum percentage loss in weight, while it was lowest in white gram. Adult emergence did not differ significantly.

Kameswara Rao and Krishnamurthy Rao (1982) reported life cycle of Callosobruchus maculatus between 33 and 42days in the pigeonpea varieties C-53 Hy-2 and PDM-1, while it extended to about 40 to 44days in other varieties, UPAS-120, ST-1, Mukta and BDN-1. Largest number (307) of adult developed in C-53, while only 50 adults developed in UPAS-120. Maximum egg laying was observed on C-53 followed by Hy-2, ST-1, BDN-1, while in PDM-1, UPAS-120 and Mukta significantly low number of eggs were deposited the percentage of grain damage was the highest in C-53 while UPAS-120 and Mukta showed least damage by the bruchid C. maculatus. Dushyant et al. (1991) studied comparative susceptibility of 33 genotypes of pigeonpea (Cajanus cajan) to pulse beetle tested in laboratory conditions. Significant variation was found in the degree of resistance with none being completely immune.

Dogre et al. (1993) evaluated seeds of 24 accessions of pigeonpea (Cajanus cajan (L.) Millsp.) and four other species of Cajanus (formerly Atylosia) for their resistance to infestation by Callosobruchus maculatus (F.). None of the pigeonpea accessions were resistant but resistance was evident in three species of Cajanus. In C. platecarpus most of the larvae failed to enter the hard seed coats but the few which did enter the seeds, developed normally. Adults did not emerge from the seeds of C. scarbaeoides, even though most of the larvae entered the seeds. In C. sericeus, the number of larvae entering the seeds as well as adult emergence was significantly reduced.

Relative susceptibility of 44 pigeonpea varieties to attack by C. chinensis and factors involved in their susceptibility were determined in study by Modi et al. (1994). Based on number of eggs per 50 grains and grain damage, cultivars ICPL-89044, KM-9 and KPAS-120 were found to be less susceptible than the other varieties and susceptibility was affected by size and shape of seed but not by seed color.

Kadoo and Rane (1995) used 12 commonly grown varieties of pigeonpea to study the extent of damage and varietal resistance and susceptibility to C. chinensis. None of the 12 varieties were found to be immune to infestation by C. chinensis. However, there

47 were significant differences in the relative susceptibility of different varieties to bruchid attack.

Jadhav et al. (2012) reported that wild species C. scarabaeoides accession ICPW 130 and C. platycarpus accession ICPW 66 had lowest damage (14% and 16% respectively). The larval/pupal period was prolonged in wild accessions (42 – 55days) compared to 33days on the susceptible control ICPL 85010 indicating the antibiosis mechanism of resistance in the wild species. Though there was high oviposition on the seeds of interspecific derivative of C. platycarpus A 4-10-7-19, it had the least damage (10%). In the rest of the derivatives the damage rating ranged between 10 and 55% compared to 80% damage in the susceptible control, reflecting the potential of utilizing these wild species derivatives in pigeonpea crop improvement to overcome the Bruchid damage.

Validation of determinacy trait expression

In some flowering plants, two types of growth habit/flowering pattern exist. These included indeterminate and determinate types depending on whether the terminal meristems are vegetative or reproductive. In indeterminate genotypes, the terminal meristems at the branch and stem apices remain in a vegetative state during which it controls the production of new nodes with leaves, produces an inflorescence meristems that only generates axillary floral meristems and hence continues to grow in stem length, flower and set pods as long as temperature and moisture permit (Bradley et al., 1997; Tian et al., 2010). In determinate genotypes, the terminal meristems eventually converted from a vegetative to a reproductive state, resulting in the production of a terminal flower and as a result, the vegetative growth ceases at flowering or continues for a short period thereafter (Bernard, 1972; Bradley et al., 1997). Thus, the stem growth habit plays an important role in deciding the plant architecture, which is of major agronomic importance as it determines adaptability of plant to cultivation and potential grain yield (Reinhardt and Kuhlemeier, 2002). Most of the pigeonpea cultivars are indeterminate in nature, i.e., plant continues to produce vegetative growth whenever soil moisture, temperature and other environmental factors are favorable. However, determinate forms of pigeonpea and even sometimes semi-determinate forms also exists (Waldia and Singh, 1987; Gupta and Kapoor, 1991; Mir et al., 2013).

Generally pigeonpea lines have indeterminate (IDT) flowering pattern, where inflorescence develops as axillary racemes from all over the branches and flowering proceeds acropetally from base to apex, both within the raceme and on the branches. However, some genotypes have been reported, where apical buds of the main shoots develop into inflorescences. These genotypes are called determinate (DT) genotypes. The wild relatives and most of the cultivated pigeonpea are having indeterminate growth habit, and therefore, it is believed that determinacy trait has been selected during pigeonpea domestication. Occasionally, flowering pattern of some lines have been found intermediate between IDT and DT, and the lines with this type of flowering pattern are called semi-determinate (SDT) lines (Craufurd et al., 2001). Pigeonpea lines with DT flowering pattern has several advantages over lines with IDT/SDT flowering pattern. For instance, some of the advantages of DT lines include (i) much shorter heights and therefore increased lodging resistance, (ii) more main stem branches/plant, (iii) shorter flowering and reproductive periods and (iv) flowering earlier and requiring a single-pass harvest of pods of nearly the same age, thus allowing mechanized harvesting (Robinson and Wilcox, 1998; Kilgore Norquest and Sneller, 2000).

Determinate genotypes are also available in several grain legumes, for example soybean (Bernard, 1972), common bean (Tulmann and Alberini, 1989), cowpea (Steele et al., 1985) and broad bean (Filippetti, 1986). Determinate growth habit is characterized and described in different crop species. In pigeonpea plants with determinate growth habit, inflorescence is short, the apical bud develops into a flower, the sequence of inflorescence production is basipetal (Sheldrake, 1984) and flowers occur more or less in the same plane (Gupta and Kapoor, 1991).

Molecular genetic studies including QTL mapping and candidate gene sequencing for flowering time/growth habit/determinacy have identified a large number of responsible genes/QTLs in several plant species including Arabidopsis, rice, barley, wheat, maize, pea, common bean, soybean (Alonso-Blanco et al., 2009). Identification of multiple candidate genes including terminal flower 1 (TFL1), terminal flower 2 (TFL2) and polymorphisms has also allowed comparative analyses among some of these species. For instance, in the case of pea, PsTFL1a, a homologue of TFL1 of Arabidopsis, has been found responsible for determinacy trait (Foucher et al., 2003). In the case of common bean, a QTL was identified earlier for growth habit (Koinange et al., 1996, Poncet et al., 2004), and within this QTL region, a candidate gene homologous to TFL1_PvTFL1y_ has

49 been identified (Kwak et al., 2008). The function of this candidate gene has been recently validated for determinacy in common bean through candidate gene sequencing and transformation studies (Repinski et al., 2012).

In Arabidopsis thaliana, primary apical meristems of determinate plants terminated by flowers and plant height reduced (Alvarez et al., 1992). The terminal flower 1 gene (TFL1) in Arabidopsis, a regulatory gene encoding a signaling protein of shoot meristems, is a homolog of the soybean stem growth habit gene Dt1 (Liu et al., 2010). Therefore, terminal flower and reduced height seems to be the characteristics associated with determinate growth habit in different crops. TFL1 is homologous to phosphatidyl ethanolamine binding proteins (PEBPs) that play diverse roles related to signaling pathways controlling growth and differentiation (Benlloch et al., 2007).

Thus, the study of determinacy in pigeonpea is indispensable being an important grain legume crop of the tropics and subtropics. Only a few studies were undertaken long back on inheritance of determinacy, and there is hardly any report available where any gene has been identified for this important adaptive trait in pigeonpea (Waldia and Singh, 1987, Gumber and Singh, 1997). For instance, inheritance of DT, SDT and IDT in pigeonpea was studied by using 15 different cross combinations of DT, SDT and IDT lines (Gupta and Kapoor, 1991). Recently, Mir et al. (2013) investigated determinacy trait in a set of 94 pigeonpea lines including 11 determinate (DT) and 83 indeterminate (IDT) lines which were genotyped using DArT arrays (with 6144 features) and 768 SNP markers using Golden Gate assay. Association analysis on marker genotyping and phenotyping data showed a significant association of determinacy with 19 SNP and 6 DArT markers explaining 8.05to 8.58% and 7.26 to 14.53% phenotypic variation, respectively.

The most apparent difficulty is studying the genetic variability in plant architecture related genes because most of such genes belong to multigenic families and this can lead to errors in comparisons, for example, non-orthologous loci can be incorrectly compared. This difficulty can be overcome if the gene is in a unique copy in the genome, or, at least, if a gene-specific primer pair used for PCR-amplification amplifies a unique sequence. This can be determined by PCR-amplification on genomic DNA from a completely homozygous plant (for example an highly inbred line) and subsequent direct sequencing of the amplicon: if no SNPs occur in the ferogram, then the

50 amplified product is unique and can be compared to other allelic products from genomic DNAs of other lines.

In pigeonpea, several genes have been tested for their candidature for determinacy trait and after careful evaluation, one gene CcTFL1 has been found as most promising/candidate genes for determinacy in pigeonpea (Mir et al., 2014; under preparation). Many earlier studies have also indicated that TFL1 is the key gene associated with determinate growth habit in soybean, pea, common bean (Tian et al., 2010; Foucher et al., 2003, Kwak et al., 2008).

MATERIALS AND METHODS

Materials and Methods

Wide crosses

Seed Source: The Genetic Resources Unit, ICRISAT was the source of seeds for all the C. cajanifolius, C. lanceolatus C. platycarpus and C. volubilis species while the seed of the cultivars of C. cajan was obtained from the pigeonpea breeding sub program at ICRISAT (ICPL 85010, ICPL87119, ICPW29, ICP15639 and ICP 15774). Moreover,

F1 and BC1 hybrids were generated from the cross C. platycarpus x C. cajan at Legumes Cell Biology laboratory of ICRISAT by using embryo rescue technique (Mallikarjuna and

Moss, 1995; Mallikarjuna, 1999). BC2 generation progeny were obtained by normal pod/seed set (Mallikarjuna et al., 2006) and stable F1BC3 lines were obtained. As the progeny lines not screened for any traits of interest before, hence were taken up for the present study.The salient morphological features and economic importance of the species used in this study are presented below.

Hybridization

In C. cajan flower opening begins in the morning at about 7am with the anthesis continuing until late in the afternoon. The flowers remain open for about 20 to 24h and the anthers dehisce before flower opening. The species of wild relatives have a similar floral biology except that they have delayed flower opening in some cases. Anther dehiscence in all the species occurs before flower opening. For hybridization, buds of the appropriate size were opened with the help of forceps and anthers were removed without injuring the stigma. The forceps was dipped in spirit after each emasculation and immediate pollination.

Crossing Methodology

Pollinations were carried out soon after emasculations in the morning before 10am. Care was taken to allow only cross pollinated pistils to remain and grow in an axil, removing all other self-pollinated pistils or immature buds. Application of gibberellic acid (50mg/l), as a cotton swab wrapped around the pistil soon after pollinations, increased pod set. The treatments were given twice, 24 and 48hrs after pollination. Care was taken to avoid physical injury to the buds during hormone application.

Collection of the Data from Interspecific derivatives

Data was collected both at vegetative stage and reproductive stages of the hybrids. Parameters Data collected Days to 50% The difference between date of sowing and the flowering date flower when about 50% of plants in a plot flowered. Days to maturity It was recorded as number of days taken from date of sowing to the date when about 75% of pods in a plot reached maturity. Plant height Height of the plant from ground level to the tip of the plant was measured in centimeters at the time of maturity. Number of Total number of pods bearing branches on the main stem of a primary branches plant was recorded. Pod clusters The total number of pod clusters which had at least two pods per plant-1 bunch was recorded. Pods plant-1 The number of pods born on the sampled plants was counted at maturity. Seed Color Color of seeds was recorded from all the plants after harvesting the pods from the plants.

Cytology

Preparation of Cornoy’s Fluid-I: 1 part of glacial acetic acid, 3 parts of absolute alcohol were taken and 2 drops of ferric chloride was added and mixed well and used when the fluid was fresh.

Preparation of Cornoy’s Fluid-II: 1 part of glacial acetic acid, 3 parts of chloroform and 6 parts of absolute alcohol were mixed and used when the fluid was fresh. Cornoy‟s fluid was a fixative solution, which was used to fix flower buds before undertaking meiosis.

Preparation of Stains:

Acetocaramine:

One gram carmine was dissolved in 45% acetic acid by boiling under a reflux condenser and was filtered. This stain was used to study meiotic preparations, cytology and pollen fertility. Acetocarmine stains DNA bright red. Hence the chromosomes are stained red when stained with acetocarmine. To test the pollen fertility/viability of pollen grains, stained grains were taken as fertile grains and unstained grains were counted as sterile grains.

Alexander’s stain:

95% Alcohol- 10ml, Malachite green- 100mg, Distilled water- 50ml, Glycerol- 25ml, Phenol- 5gm, Chloral hydrate- 5g, Acid fuschin- 50mg, Orange G- 5mg and Glacial acetic acid- 4ml were mixed in the given quantity and the stain was stored in a dark colored bottle. When stain was applied on the non-dehiscent anthers they appeared green and pink in color. The sterile grains stained green and fertile grains stained pink. Staining was hastened by flaming the slide (for loose thin walled pollens) or by immersing the anther for 24-48hrs at 500C in Alexander stain. The stain differentiated both the viable and the nonviable pollens. The differentiation of aborted and nonaborted pollen was found to depend on the concentration of the dyes, thickness of pollen walls and pH of the staining solution.

Malachite green thus used in Alexander stain, stains the pollen walls, as the aborted pollens have only walls it appears green. Acid fuchsin stains the protoplasm and hence it colors the non-aborted pollens from red to deep red depending upon the pH and the concentration of the medium. Orange G improves the differentiation and glacial acetic acid lowers the pH from 2.8 to 2.4 so as to stain the protoplasm of the thin walled pollens red.

Pollen fertility analysis was compared using both Acetocarmine and Alexander stain by applying both the stains for each plant to make sure that the analysis was error free. More or less the same amount of pollen fertility was observed when the two stains were compared.

Bud fixing

For meiotic studies, young buds were collected from the plants at a fixed time (between 7:30am and 8:00am) so the buds were all the correct buds size. However, it was observed that the buds of the correct size produced the same result irrespective of the time the buds were collected. Initially, these buds were placed in a small glass tube that contained Cornoy‟s fluid-II for 48hrs. They were later transferred to glass tubes that contained a mixture of Cornoy‟s fluid–I and 1-2 drops of ferric chloride for 24hrs. After that, these buds were transferred into the Cornoy‟s II solution for 24hrs. The date and time of such buds (collected after this treatment) were noted for reference.

Once this process was complete, the buds were dissected under the magnifier for the anthers and squashed in 1% acetocaramine stain. When the buds were collected, the date and time was noted. The preparations were observed under a microscope after slowly placing the coverslip on the slide.

The buds were then squashed in 4% acetocaramine dye and warmed with an iron needle before putting cover slip. Iron acts as a moderant. In case of over staining of cytoplasm and adequate separation of chromosomes, preparation was treated with 45% acetic acid. The slide was gently warmed and the cover slip put over the preparation to facilitate spreading of the chromosomes.

Dissection of Buds

Buds were dissected for the anthers under the magnifier. After the dissection of the anthers from the buds, they were transferred on to the slide and acetocaramine stain was added to the anthers and squashed with a squasher. These were observed under binocular microscope after placing the cover slip on the slide.

Preservation

To preserve the prepared slide, it was sealed with wax and stored in a petriplate that had a wet tissue paper under the slide. The slide was later stored in a refrigerator.

Pollen fertility

For pollen fertility evaluation, fresh and unopened flowers were collected at 10am. Pollen fertility was estimated by staining with acetocaramine. Uniformly stained round pollen grains were considered as viable grains and all others as sterile ones. The process we followed:

1. A drop of acetocaramine was added to a clean slide. 2. The petals from a bud were removed with the help of forceps and the anthers were dusted in such a way that the pollen grains fall into the stain. 3. A cover slip was placed on the stain containing pollen grains. 4. On observation under the light microscope, fertile pollens appeared to be in dark pink color and the sterile ones were in light pink and oval shaped. 5. Counting was done at 10 x in 10 different regions.

Meiotic analysis

The analysis was done to determine the number of univalents, bivalents, trivalents and tetravalents. From these scores the proportion of the complement of diploid (or) tetraploid plant chromosomes that participated in pairing was ascertained for each cell. The data was expressed as the mean proportion of the complement paired for hybrids of each category.

The different cytological parameters that were analyzed are:  Chromosome configuration at metaphase  Types of frequency of various meiotic anomalies  Number of diads, triads, and tetrads  Pollen fertility and pollen size and  Asynchrony for meiotic division.

Calculations and formulae

The calculations and formulae for metaphase analysis are:

Total number of univalents Mean number of univalents = Number of readings

Total number of bivalents Mean number of bivalents = Number of readings

Total number of bivalents Mean number of trivalents = Number of readings

Total number of tetravalents Mean number of tetravalents = Number of readings

The calculations and formulae to find the % of normal and abnormal cells are:

Total number of normal cells % of normal cells = x100 Number of readings

Total number of Abnormal cells % of abnormal cells = x100 Number of readings

Anaphase Analysis

For anaphase analysis equal distribution and unequal distribution were also be taken into account. The formula to calculate % of normal and abnormal cells:

Total number of normal cells % of normal cells = x100 Number of readings

Total number of Abnormal cells % of Abnormal cells = x100 Number of readings

Tetrad Analysis

The formula to calculate the % of normal and abnormal cells for tetrads:

Total number of tetrads % of normal cells = x100 Total Number of cells (diads + triads + tetrads)

Total number of abnormal cells (diads or triads) % of abnormal cells = x100 Total Number of cells (diads + triads + tetrads)

The formula to calculate the % of normal and abnormal cells for diads and triads:

Total number of tetrads with micronuclei % with micronuclei = x100 Total Number of tetrads

Total number of tetrads without micronuclei % wothout micronuclei = x100 Total Number of tetrads

Microtomy

Anatomical analysis, with the aid of microtomy, will also be carried out, which would provide with an insight into the developmental stages of microsporogenesis from the tetrad stage up to pollen formation. A microtome is a sectioning device that allows preparing samples of material for observation under light microscope. While there are many types of microtome machines, we used the rotary microtome (Micros Rotary- Microtomes RAZOR AND STEELY, the cut thickness is between 1 and 60µm) for our analysis.

The test material can be stored in FAA (Formalin-Alcohol-Acetic acid) for 2- 3days. Replacing water with paraffin hardens the test material. The tissue is then cut at thickness varying from 2 to 50µm thick. The tissue removed from the paraffin and

57 mounted on a microscope slide. It is then stained with appropriate aqueous dye and examined using a light microscope. The different stages of microsporogenesis can be thus analyzed and photographed.

Materials

To find out the mechanism of microsporogenesis, sterile and fertile flower samples derived from the crosses between sterile F1s and other cultivars of pigeonpea.

Methodology

The microtomy procedure (developmental stages of microsporogenesis) is as follows:

1. The different sizes of collected and analysed buds were stored in FAA solution.

2. A series of dehydrating reagents (TBA series) were prepared. Small buds and anthers from large buds were separated and placed in each solution for 24hrs.

3. Similarly, the process of dehydration was done with another series of solutions.

4. After dehydration, the buds and anthers were transferred into separate labeled bottles with cork, along with the last dehydrating solution i.e., xylol. A few shavings of paraffin wax were added. The wax was allowed to dissolve in the last dehydrating solution at room temperature. Paraffin shavings were added 3 to 4 more times and the solution was allowed to dissolve.

5. Next, more wax shavings were added and placed in an incubator (maintained at 56o C) to melt the solution. The melted xylol wax was replaced with pure wax by pouring out the xylol wax partially and adding fresh paraffin shavings. Care was taken not to lose the specimen buds in the transfer. This process was repeated till xylol free wax was obtained. Obtaining xylol-free wax is important as the presence of xylol hinders the clarity of the microscopic preparation.

6. Paper boats were prepared and coated with glycerol, which helped to ease the removal of the solidified wax.

7. Pure wax was melted separately and sufficient quantity was poured into the paper boats. The buds and anthers in the melted wax were transferred into paper boats separately and arranged with a needle. Intermediate heating of the needle aids in easy arrangement of the specimen in the wax block. It was allowed to solidify for minimum 3hrs without disturbing. Air bubbles were avoided as they may lead to gaps in the ribbons while cutting in the microtome.

8. The paper boats were opened and the paraffin block was trimmed into small square blocks according to the size of the specimen. These paraffin blocks were buttressed onto the wooden blocks by heating a little wax and pouring along the sides of the paraffin block. This was done to fortify the paraffin on the wooden block.

• If the paraffin blocks were not trimmed perfectly, the ribbons might not follow a straight line while cutting in the microtome.

• If the paraffin block with the buds or anthers was not fixed firmly, there is a risk of losing the specimen while cutting the ribbons in the microtome.

9. The blocks thus fixed were trimmed into perfect squares using a scalpel blade taking care not to cut the paraffin till the bottom as it might loosen its hold on the wooden block.

10. This wooden block with the specimen was fixed in the microtome machine in the slot provided. The blade was also fixed and tightened appropriately.

11. The position of the blade and the block was adjusted. The required thickness of the ribbons in micrometers can be obtained by adjusting the knob provided. A thickness of 10µm is usually preferred.

12. The handle of the microtome machine was rotated so that the block with the specimen moves forward and cuts on the blade at the bottom and gave out ribbons of required thickness. These ribbons were carefully collected and kept aside.

13. The ribbons thus obtained were cut using a blade into required length and placed on a Haupt‟s adhesive smeared clean slide. The slides were gently heated so that the ribbon gets fixed to the slide.

14. The slides were labelled and the slide was stored in a clean dust free space for 2 to 3 days. Haupt‟s composition: 1gm of gelatin was dissolved in 100ml of warm water (30oC). 2gm of phenol crystals and 15ml glycerin was added to it and mixed.

 These slides were placed in the respective slide holder in the coupling jar; the jar was filled with xylol. Xylol dissolves the wax and provides us with a wax free specimen sections. After the wax dissolves, the xylol was emptied into a bottle labelled „de-wax xylol‟.

15. This process was repeated with remaining dehydrating solution series prepared earlier in the reverse order. The used solutions were collected in separate bottles and were labeled accordingly.

• For staining with Toluidine blue, the slides were moved from the last dehydrating solution to water before staining.

• For staining with Safranine, the slides were moved from the last dehydrating solution to the stain.

• For staining with Fast green, the slides were moved to the dehydrating solution of composition 90% TBA and 10% ethanol and then into stain.

16. After staining, the slides in Safranine and Toluidine blue were washed with the complete dehydrating series for few s in each solution. However, the fast green stained slides were washed only in pure TBA, 1:3 xylene, 1:1 xylene, 3:1 xylene and pure xylene series for 5min min in each solution respectively.

• Fast green FCF composition: 500mg of fast green stain was dissolved in 100ml of 95% ethanol.

• Toluidine blue composition: 0.05gm of toluidine blue stain was dissolved in 100 ml water.

• Safranine composition: 1gm of Safranine was dissolved in 100ml water.

17. Each slide was mounted using D.P.X mountant and allowed to dry.

18. Observations were made and photographs were taken.

Molecular tools

This experiment confirms hybridity in C. lanceolatus and C. volubilis derivatives and validation of growth habit (SDT/IDT or DT) in F2 population C. volubilis derivatives.

Plant material

Leaf samples of the hybrids derived from C. lanceolatus and C. volubilis were collected for hybrid purity assessment. To confirm the determinacy trait in the F2 hybrids derived from C. volubilis are also subjected to PCR by using IDT/SDT or DT trait PCR based allele specific markers.

S. No Plant Sample Cross C. cajanX C. lanceolatus F1 Hybrids

1 ICPL 85010 2 ICP 15639 3 F1 P-1 4 F1 P-2

5 F1 P-4

6 F P-7 1 7 F1 P-8 8 F1 P-9 9 F1 P-10 10 F1 P-11

Cross C. cajanX C. volubilis F2 population 1 ICPL 85010 2 ICPW 15774 3 MN1 4 50 5 50B

6 51 7 51B 8 52 9 52A 10 54 11 54A 12 55 13 57A 14 60 15 61 16 61A 17 64 18 64A

19 65A 20 65

21 66 22 73

23 74 24 74A

DNA Extraction

Genomic DNA was isolated from two to three young leaves of F1 genotypes from C. lanceolatus and F2 progenies of C. volubilis populations by acetyl trimethyl ammonium bromide procedure as mentioned in Cuc et al. (2008).

Sample preparation

• Leaves were harvested from 15days old seedlings.

• Leaf tissue of 70-100mg was placed in 12 x 8 well strip tube with strip cap (Marsh Biomarket, USA) in a 96 deep-well plate together with two 4mm stainless steel grinding balls (Spex CertiPrep, USA).

CTAB extraction

• Each sample was mixed with 450μl of preheated (65oC) extraction buffer (100mM Tris-HCl (pH-8, 1.4M NaCl, 20mM EDTA, CTAB (2-3%w/v) and β- mercaptoethanol) was added to each sample and secured with eight strip caps.

• Samples were processed in a Geno Grinder 2000 (Spex CertiPrep, USA), following manufacturer‟s instructions, at 500strokes/min for 5 times at 2min interval.

• Plate was fitted into locking device and incubated at 65oC for 10min with shaking at periodical intervals.

Solvent extraction

• Each of the sample were mixed with 450μl of chloroform-isoamylalcohol (24:1) by inverting twice.

• Plate was centrifuged at 5500rpm for 10min. The aqueous layer (300μl) is transferred to fresh strip tubes (Marsh Biomarket, USA).

Initial DNA precipitation

• 0.7 vol (210μl) of isopropanol (stored at –20o C) was added to each sample and inverted once to mix.

• Plate was centrifuged at 5000rpm for 15min.

• Supernatant was decanted from each sample and pellet was air dried for 20min.

RNase treatment

200 μl low salt TE [10 mM Tris EDTA (pH-8)] and 3 μl RNase was added to each sample and incubated at 37o C for 30 min.

Solvent extraction

• 200 μl of phenol-chloroform-isoamylalcohol (25:24:1) was added to each sample and inverted twice to mix.

• Plate was centrifuged at 5000rpm for 5min.

• Aqueous layer was transferred to a fresh 96 deep-well plate (Marsh Biomarket, USA).

• 200μl chloroform-isoamylalcohol (24:1) was added to each sample and inverted twice to mix.

• Plate was centrifuged at 5000rpm for 5min.

• Aqueous layer was transferred to a fresh 96 deep-well plate.

• 315μl ethanol-acetate solution [30ml ethanol, 1.5ml 3M NaOAc (pH-5.2)] was then added to each sample and placed in –20o C for 5min.

• Plate was again centrifuged at 5000rpm for 5min.

• Supernatant was decanted from each sample and pellet was washed with 70% ethanol.

• Plate was centrifuged at 6000rpm for 10min.

• Supernatant was again decanted from each sample and samples were air dried for approximately 1h.

• Pellet was resuspended in 100μl low-salt TE and stored at 4oC.

Quantification of DNA

DNA quality was checked and quantified on 0.8% agarose gel with known concentration of uncut lambda DNA standard.

Polymerase Chain Reactions (PCRs)

PCRs for amplification of SSR loci were performed in a 5μl reaction volume

[0.5μl of 10X PCR buffer, 1.0μl of 15mM MgCl2, 0.25μl of 2mM dNTPs, 0.50μl of 2 pM/μl primer anchored with M13-tail (MWG-Biotech AG, Bangalore, India), 0.1 U of Taq polymerase (Bioline, London, UK) and 1.0μl (5ng/μl) of template DNA) in 96-well micro titre plate (ABgene, Rockford, IL, USA) using thermal cycler GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA, USA). A touch down PCR program was used to amplify the DNA fragments: initial denaturation was for 5 min at 95°C followed by 5 cycles of denaturation for 20s at 94°C, annealing for 20s at 60°C (the annealing temperature for each cycle being reduced by 1°C per cycle) and extension for 30s at 72° C. Subsequently, 35 cycles of denaturation at 94°C for 20s followed by annealing for 20s at 56° C and extension for 30s at 72°C and 20min of final extension at 72C. PCR products were checked for amplification on 1.2% agarose gel.

Data recording

The amplification products obtained by using M13 tailed primer pairs together with Liz Gene Scan- 500 labeled internal size standards, were analyzed on 36 cm capillaries with POP7 polymer on ABI 3730 Genetic Analyzer. Fragment analysis data were collected by the data collection software and pre-processed by the GeneMapper software version 4.0 (Applied Biosystems, Foster City, CA, USA). SSR allele data for the population was recorded as “A” (allele from male parent), “B” (allele from female parent) and “H” (alleles from both the parents “Hybrid”) format.

All the successfully amplified primer pairs were screened for polymorphism among parents and hybrids. Only 14 primers yielded scorable amplicons and genotypes were scored as A (allele from female parent), B (allele from male) and H (heterozygote- alleles from both parents). Hybrids which exhibited only „B‟ allele from male parent also considered as a true hybrid as the pollen parent (male parent) was confirmed. Purity index for each marker was calculated using scored data by applying the following formula:

Number of true hybrids containing „H‟ type allele + ‟B‟ type allele Purity index % = x 100 Total Number of hybrid tested

Biochemical Analysis

Materials and methods: Crosses were made between C. cajan and four wild relatives of pigeonpea from secondary (C. cajanifolius and C. lanceolatus) and tertiary (C. platycarpus and C. volubilis) gene pools. This resulted in a wide range of interspecific derivatives derived from the four wide crosses, I. C. cajan (ICPL 87119) x C. cajanifolius (ICP 29), II. C. cajan (ICPL 85010) x C. lanceolatus (ICPW 15639), III. C. platycarpus (ICPW 68) x C. cajan (ICPL 85010) and IV. C. cajan (ICPL 85010) x C. volubilis (ICPW 15774)) at Legume Cell Biology Laboratory, ICRISAT

SNo. Plant identification number SNo. Plant identification number Cross III (ICPW 68 X 1 cv. ICPL 87119 ICPL85010)

2 cv. ICPL 85010 1 F1BC6 A6-1

3 C. cajanifolius (ICPW 29 ) 2 F1BC6 A6-12

4 C. lanceolatus (ICP 15639) 3 F1BC5 B3-2

5 C. platycarpus (ICPW 68) 4 F1BC5 B6

6 C. volubilis (ICP15774) 5 BC3F3 C3-13

Cross I (ICPL 87119 X ICPW29) 6 BC3F3 C7-13

1 ICPL87119XICPW29(18-5) BC1F2 7 BC3F3 C11-3

2 ICPL87119XICPW29(37-1) BC1F2 8 F1BC6 D1-10-1

3 ICPL87119XICPW29(76-2) BC1F2 9 BC3F3 D2-1

4 ICPL87119XICPW29(82-2) BC1F2 10 F1BC6 E5-8-1

5 ICPL87119XICPW29(113-2) BC1F2 11 F1BC6 E8-2 Cross IV (ICPL85010 X ICPW 6 ICPL87119XICPW29(129-1) BC1F2 15774)

7 ICPL87119XICPW29(130-1) BC1F2 1 HYB-3 F2

8 ICPL87119XICPW29(131-1) BC1F2

9 ICPL87119XICPW29(132-1) BC1F2

10 ICPL87119XICPW29(133-1) BC1F2 11 (ICPL87119XICPW29 -3)1-1

12 ICPL87119XICPW29 F1

Cross II (ICPL 85010xICPW 15639)

1 ICPL85010XICPW 15639 P1 F1BC1

2 ICPL85010XICPW 15639 P2 F1BC1

3 ICPL85010XICPW 15639 P3 F2

4 ICPL85010XICPW 15639 P4 F1BC1

5 ICPL85010XICPW 15639 P6 F1BC1

6 ICPL85010XICPW 15639 P7 F1BC1

7 ICPL85010XICPW 15639 P8 F2

8 ICPL85010XICPW 15639 P10 F1BC1

9 ICPL85010 X ICPW 15639 F2 P11

10 ICPL85010 X ICPW 15639 P12 F1BC1

11 ICPL85010 X ICPW 15639 F2 P13

12 ICPL85010 X ICPW 15639 F2 P14

Proteinases and Substrates: Bovine pancreatic trypsin and bovine pancreatic α- chymotrypsin were procured from Sisco Research Laboratory, Mumbai, India. N-- benzoyl-DL arginine-p-nitroanilide (BAPNA), N-Glutaryl-L-pheny-alanine p-nitroanilide

(GLUPHEPA) and Human Pancreatic Trypsin were purchased from Sigma Aldrich (St. Louis, MO).

Extraction of crude protein from seeds

Crude proteins were extracted from decorticated mature dry seeds of both parents and cultivars of pigeonpea according to the procedure described by Prasad et al. (2009). Mature dry decorticated seeds were ground to a fine powder, de-pigmented and defatted with several washes of acetone and hexane, respectively. After the solvents were removed, the air dried seed powder was extracted in 0.05 M Tris- HCl, Ph 8.0 containing 1% PVP (molecular weight of 40,000) in 1:6 (w/v) with mild continuous stirring for overnight at 4o C. The clear supernatant obtained after centrifugation (twice) at 10,000 rpm for 20 min at 4o C, was used as crude protein. The clear supernatant was collected, aliquoted and frozen at -20o C. The protein content was estimated by Folin-Cio-calteau method using bovine serum albumin as a standard (Lowry et al., 1951).

Extraction of Larval Midgut Proteinases

The 4th to 5th instar larvae of H.armigera were collected from the pigeonpea fields of ICRISAT and were narcotized on ice for 15 min and dissected in an insect Ringer solution (0.13M NaCl, 0.5M KCl, 0.1mM CaCl2 and 1mM Phenylmethylsulphonyl fluoride). The midgut was removed and placed on iso-osmotic saline (0.15M NaCl) solution. Gut tissue was homogenized in 0.15M NaCl and centrifuged twice at 12,000rpm for 10min at 4o C. The supernatant was collected and stored frozen at -20o C for further in vitro assays.

Assay of Proteinase Inhibitors

Trypsin or chymotrypsin inhibitory activity was determined by using appropriate volumes of crude extracts that results in 40-60% decrease in corresponding enzyme activity. Assay mixture (1.0ml) consists of PIs in assay buffer, 50mM Tris-HCl, containing 20 mM CaCl2 either at pH 8.2 for trypsin or pH 7.8 for chymotrypsin. 10g of trypsin or 80µg of chymotrypsin was added to the assay mixture and incubated for 15min at 37oC. Residual proteinase activity in the above assay mixture was determined after incubating for 45min at 37oC using 1mM BAPNA (1.0ml) as a substrate (Erlanger et al., 1961) and 1mM GLUPHEPA (1.0 ml) as a substrate for chymotrypsin (Mueller and Weder, 1989).The reaction was terminated by adding 0.2ml of 30% acetic acid. The

68 activity of PIs was expressed as trypsin inhibitor (TI) units/mg protein or chymotrypsin inhibitor (CI) units/mg protein. The molar extinction coefficient (M-1 cm-1) for p- nitroanilide at 410nm is equivalent to 8,800. One PI unit was defined as the amount of inhibitor required to inhibit 50% of the corresponding enzyme (trypsi/chymotrypsin/HGP) activity under the optimal assay conditions.

Effect of PIs on midgut trypsin like proteinases activity of H. armigera

The effect of PIs on the midgut trypsin-like proteinases was determined by incubating them in assay buffer (0.5ml), 50mM glycine-NaOH (pH 10.5) with midgut trypsin like extract of H. armigera. After incubation with 1mM BAPNA at 37°C for 45 min, the reaction was stopped with 30% acetic acid (v/v), and absorbance at 410nm was recorded. All of the assays were done in triplicates along with appropriate controls.

Visualization of Inhibitor Profiles against Trypsin, Chymotrypsin and Insect Midgut Proteinases

Gelatin-PAGE was performed by incorporation of gelatin (0.1%, wt:vol, to final concentration) in to the polyacrylamide gels at the time of casting as described by Felicoli et al. (1997). Electrophoresis was performed at 50V in stacking gel and 100V in resolving gel. Following electrophoresis, the gel was washed thrice with distilled water. Later the gel was incubated in 0.1M Tris-HCl containing trypsin (0.1mg/ml) at pH 8.2 or 0.1M Tris-HCl, containing chymotrypsin (0.2 mg/ml) at pH 7.8 at 4o C for 30min followed by 37o C for 2h. For identification of inhibitor profiles against midgut trypsin like proteinases, the gels were incubated either in 50Mm glycin-NaOH, pH 10.5 containing H. armigera gut extract for 30min at 4o C and subsequently for 2h at 37o C. After hydrolysis of gelatin, the gel was washed with distilled water to remove the excess enzymes and stained with 0.1% Coomassie Brilliant Blue (CBB) R250. The presence of inhibitor activity was identified by the appearance of dark blue bands in a clear background due to complex formation of the unhydrolysed gelatin with stain. Commercially available soybean trypsin chymotrypsin inhibitor (soybean BBI) was used as marker in gelatin- PAGE.

Effect of PIs on Human Pancreatic trypsin (HPT)

The effect of PIs on Human pancreatic trypsin was measured by using the BAPNA as a substrate. The assay mixture (0.2ml) consists of 0.56µg of HPT in 40 µl of

0.05M Tris-HCl (pH 8.2) containing 0.02M CaCl2 and 2mM BAPNA which gives O.D. value 1 at 37oC for 45min. The inhibitory activity of PIs on human pancreatic trypsin was determined by using different concentrations of crude extracts of seeds to the assay mixture (above) and observations are recorded on plate reader (TECAN, Switzerland) at 410nm.

Statistical Analysis

All the experiments were carried out three times each with three replications, and the mean ± SE was reported by using Sigma plot 12.0 (Systat Software Inc.,San Jose, CA). The results are shown as mean ± SE of three independent experiments done in triplicates.

Field screening under natural infestation

The stability of resistance was tested by screening interspecific derivatives along with their susceptible parents (ICPL 85010 and ICPL 87119) under unprotected field conditions. Field trials were carried out at ICRISAT, India. Seeds were sown in two replications in a randomized complete block design on the ridges 75cm apart, each row 2m long for each line (comprising of 20 seeds), crop was thinned to a spacing of 30cm between the plants after 21days of seedling emergence. Standard agronomic practices were followed, with a basal fertilizer (N: P: K) application in the proportion of 100:60:40kg/ha, which was applied in the furrows before planting. In addition, a basal dose of fungicide (metalaxyl 1.0kg/ha) was also applied to control Fusarium wilt at the seedling stage. Subsequently, no other control measures were applied throughout the cropping season. The crop was planted in June at the start of the monsoon season and irrigated at regular intervals between December to mid-February.

Bruchid Analysis

Bruchids were collected with a 'pooter' from the damaged seeds in the pigeonpea field on ICRISAT farm and were allowed to multiply on a popular local variety (ICPL 87119) in the laboratory. After rearing them for two generations, the adults were used for screening wild Cajanus species. Laboratory culture of the beetles was carried out in a cylindrical transparent plastic box (13 x 11cm diameter) with a well ventilated lid.

Seed Material

Seeds of wild (C. lanceolatus) and cultivated (ICPL 85010) parents including some of the F1 hybrids like P1, P2, P4, P5, P6, P7 and P8 were evaluated for bruchid resistance.

Test Procedure

Ten fresh seeds of each accession were placed in glass petri-dish and the dish was placed in a plastic 12-cm wide box and a transparent mesh covering the lid. To facilitate observations and to prevent the movement of seeds when beetles move on them a circular 'Whatman' filter paper was placed in the petri-dish. Four freshly emerged bruchids of both sexes were placed in individual boxes for 5 days. Each box was checked at 48h interval for egg-laying. After 5 days of egg-laying, individual seeds with eggs on them were counted and recorded. At the time of removal of bruchids after 5 days, it was ensured that all the seeds had eggs on them. Most seeds had 2-3 and sometimes more eggs on them. The whole experiment was kept inside a 'Percival' Incubator with 24oC + 2oC with 70% RH and 14:10h (L: D). Observations on the number of eggs laid on 10 seeds was converted to mean % oviposition against control, number of eggs failed to hatch, number of adults emerged and average number of days taken for adult emergence were recorded.

RESULTS AND DISCUSSION

Results and Discussion

1. Species crossability and hybrid morphology RESULTS

Wild relatives of the pigeonpea from the secondary (C. cajanifolius and C. lanceolatus) and tertiary gene pools (C. platycarpus and C. volubilis) were crossed with cultivated pigeonpea. In all the crosses wild relatives were used as the male parent and cultivated pigeonpea as the female parent, except in the case of C. platycarpus which was used as the female parent. Although reciprocal crossings were routinely attempted in all

instances, F1 hybrids were obtained only with cultivated pigeonpea as the male parent. Reciprocal of the cross was not successful. In the unsuccessful crosses bud drop

commenced within two days after pollination. GA3 treatments prolonged ovary development and delayed bud drop by varying periods depending upon the cross combination. i. Crosses utilizing secondary gene pool species A. C. cajan (ICPL 87119) X C. cajanifolius (ICPW 29)  Crossability

Emasculations followed by pollinations were carried out in the morning using C. cajan cultivar ICPL 87119 as the female parent and C. cajanifolius accession as the pollen donor (Fig. 1). Care was taken to allow only cross pollinated pistils to remain and grow in an axil, removing all other self pollinated pistils and/or immature buds.

Application of gibberellic acid (GA3) increased pod set. Mature pods were harvested

upon maturity. F1 hybrids were back-crossed with the recurrent parent and F1BC1 seeds

were obtained. F1BC1 hybrid plants were raised in the glasshouse and mature F2BC1 seeds

were obtained. Seeds of BC1F2 population were planted in the field in mid June 2012 (in

the Kharif season). Ten lines of BC1F2 generation (18-5, 37-1, 76-2, 82-2, 113-2, 129-1,

130-1, 131-1, 132-1and 133-1) and one F1 (ICPL87119 X ICPW29) line were planted in each row.

 Hybrid morphology

Morphology the F1 hybrid was intermediate between the parents with respect to the flowering pattern, pod shape, pod size, seed shape and seed color. A comparison of parents and their hybrids for qualitative and quantitative traits are presented in Table 2. In most cases, F1s were similar to the C. cajanifolius in the leaf color, branching pattern, prominent locules in pods and seed shape. With respect to plant height (185cm) and days to flower (120day), they resembled the female parent, the cultivated variety, i.e. ICPL87119 (Asha). As the plants grew all the branches developed dark green leaves similar to the cultivated parent. Observations such as days to flower, plant height, number of primary, secondary and tertiary branches, branching pattern, growth habit, flower color, pattern of streaks and pod color were recorded for each plant individually and are as described below.

 Days to flower

F1 and BC1F2hybrids flowered in 115 to 125days after sowing similar to the cultivated parent (ICPL87119) which flowered in 120day after sowing. All BC1F3 progeny flowered between 95 to 132days after germination. Some lines of BC1F3 generation, 18-5, 76-2, 82-2, 113-2, 130-1,131-1, 132-1 and 133-1 took 100 to 127days to flower, similar to the cultivated parent. Whereas the remaining two lines 37-1 and 129-1 took 95 to 96days to flower which was earlier than the cultivated parent.

 Plant height

Plant height was one of the important traits considered for plant selection. Lines 18-5 and 113-2 measured 81 to 87.5cm in height as in wild C. lanceolatus. Lines 76-2 and 82-2 measured 65 to 69cm height which were shorter than the cultivated parent. These lines segregated for plant height (Fig. 2). Remaining lines, 129-1, 130-1, 131-1, 132-1 and 133-1 were in range of 103.7cm to 166cm, similar to the cultivated parent which measured 165cm. Lines129-1, 130-1, 131-1, 132-1 and 133-1 exhibited uniformity in plant height as in cultivated parent. Small, dwarf plants were observed in the lines 113-2 and 132-1.

 Number of branches

Total number of the primary (10), secondary (20) and tertiary (30) branches were counted and are detailed in the Table 2. Shorter plants had less number of primary and secondary branches than the taller plants such as line18-5 had 4.5 primary branches, 12 secondary branches and 0.5 tertiary branches. Lines 37-1, 76-2 and 82-2 exhibited 6 to 9.3 primary branches and 6.5 to 14 secondary branches. The line113-2 showed 8.9 primary, 12.7 secondary branches and 4.2 tertiary branches. Line 129-1 showed 9.2 primary, 27 secondary branches and 2.2 tertiary branches. Remaining lines 130-1, 131-1, 132-1 and 133-1 had 11.3 to 14.4 primary, 25.8 to 39 secondary and 1.5 to 8 tertiary branches. Some of the individual plants from few BC1F3lines such as 113-2, 129-1, 130- 1, 131-1, 132-1 and 133-1 showed more secondary and tertiary branches with vigorous growth. Individuals such as 113-2-10 had 62 secondary and 42 tertiary branches; 129-1-6 had 45 secondary and 16 tertiary branches; 130-1-3 had 45 secondary and 18 tertiary branches; 133-1-6 had 34 secondary and 38 tertiary branches. In addition, there are few mutants which did not grow and remained as short plants with 1-4 primary branches

(BC1F3131-1-3,131-1-4 and 132-1-3). F2 lines exhibited 8-14 primary and 9-32 secondary branches with few tertiary branches. From the recorded data it was clear that some of the lines 129-1, 130-1, 131-1, 132-1and 133-1 had more number of branches than the other lines. Also these lines had long spreading branches than the other lines.

 Branching pattern

The branching pattern in pigeonpea depends on genotype and spacing between rows and plants. Few lines such as 18-5, 37-1, 76-2, 82-2 and 113-2 segregated for three types of branching patterns, namely 60% compact, 31% semi-spreading and 9% spreading type, while the remaining lines 129-1, 130-1, 131-1, 132-1 and 133-1 exhibited spreading type of branching pattern.

 Growth habit

Based on the flowering pattern in the interspecific derivatives, the growth habit was of three types. They were determinate, semi-determinate and indeterminate types. Similar to the branching pattern, lines 37-1, 76-2, 82-2 and 113-2 also segregated for indeterminate and semi-determinate and few determinate growth habits. But remaining lines 129-1, 130-1, 131-1, 132-1 and 133-1 exhibited only indeterminate growth habit. All

74 the lines segregated for stem color, being either purple or green. Only individuals from the line 82-2 exhibited purple stem, and the remaining lines were with green stem resembling cultivated pigeonpea.

 Leaf

With respect to the leaf size and shape, BC1F3 lines18-5, 129-1, 130-1, 131-1, 132-1 and 133-1 resembled cultivated parent ICPL 87119. Individual plants of the line 37-1 had both elliptical and small leaves (Fig. 4). But few individuals (from the lines 37-1 and 82-2) 37-1-3, 37-1-4, 37-1-5 and 82-2-7 had small oblong leaves (Fig. 2 B). However lines 37-1, 76-2, 82-2 and 113-2 segregated for medium sized obovate and elliptic leaf shapes. One individual 76-2-3 from the line 76-2 had small oblong leaves (Fig. 4).

 Flowers

Flowers were yellow in 129-1, 130-1, 131-1, 132-1 and 133-1 lines. Some of the individuals 37-1-3, 37-1-4 and 37-1-5 from the line 37-1, 76-2 and 82-2 from had small yellow flowers throughout the flowering duration (Fig. 3 A). Pattern of the streaks on the standard petal was another important trait with respect to the flower morhology. All the individuals in the F2 exhibited sparse streaks on the standard petal. However, all the

BC1F3 lines did not have streaks except few individuals from the lines 37-1, 130-1 and 131-1 had sparse streaks.

 Pods

Pods were a mixture of green and purple color in both the parents (wild and cultivated). BC1 F3 lines (18-5, 37-1, 76-2, 82-2, 113-2, 129-1, 130-1, 131-1, 132-1 and 133-1) had green pods with purple streaks as in the cultivated parent (Fig. 3 C). Some of the individuals in BC1F3 population such as 18-5-3, 18-5-4, 18-5-5 18-5-9, 76-2-6, 82-2- 2, 82-2-7, 113-2-5, 113-2-6, 130-1-2, 130-1-3, 130-1-4 and 130-1-8 had complete purple pods with normal cultivated pigeonpea pod shape. Two individual plants 76-2-3 and 76- 2-5 had green and purple colored, short flattened pods with prominent locules as in the wild (Fig. 3 B) (C. cajanifolius) parent.

B. C. cajan (ICPL 85010) X C. lanceolatus (ICP15639)  Crossability

For the cross C. cajan cv ICPL85010 x C. lanceolatus (ICPW15639) to be successful, it was necessary for the pollinations to be supported with the application of gibberellic acid (GA3) to the pollinated pistils. Crosses were made using C. cajanus as the female parent (Fig. 6A) and C. lanceolatus as the pollen donor (Fig. 6 B). Application of

GA3 in the form of a cotton swab wrapped around the pistil soon after pollination, increased pod set. Out of 86 pollinations, 20 pods were obtained. Pods were harvested 40 to 45day after pollination. From 20 pods, 35 seeds were obtained of which 12 seeds were semi-shriveled. Fourteen F1 hybrid plants were obtained by germinating the seeds. Plants grew normally and flowered profusely. Amongst the 14 F1 hybrid plants, P1, P4, P7, P10 and P12 were male sterile and the rest were fertile plants (Fig. 7). Selfing the fertile F1 hybrids resulted in F2 pods/seeds and backcrossing them with cultivated parent ICPL

85010 resulted in BC1 pods.

Male sterile F1 plants were back-crossed with ICPL 85010 which resulted in a range of BC1 pod set (0 to 30 %) (Table 3). Segregation was observed in BC1 hybrids which had intermediate growth habit, but resembling the female parent, i.e. cultivated pigeonpea, in gross morphology. The F1 male-sterile plants were also crossed to 12 unrelated pigeonpea cultivars (MN1, MN5, MN8, ICPL88039, ICPL85010, ICPL88034, ICPL85030, ICPL16198, ICPL1447, ICP14444, ICP7035 and ICP92016) of which only seven cultivars (MN1, MN5, MN8, ICPL88039, ICPL85010, ICPL88034 and ICPL85030) were able to set the pods (Table 5). Percentage of pod set was more in case of F1 P-4 when crossed with different unrelated cultivars such as ICPL85010,

ICPL88034, ICPL88039, ICP92016, MN1 and MN5 resulted in F1BC1 pod set which ranged between 3.5 to 30.9%. In the remaining sterile plants such as F1P-1 in which the pod set was 1.1 to 20.9%; in P-7 it was 1.4 to 22.7%; in P-10 it was 1.4 to 11.2% and in

P-12 it was 0.6 to 17.5%. Fertile F1 hybrids P2, P3, P5, P6, P8, P9, P11, P13 and P14 were able to set the pods after tripping the partially fertile flowers of the plants and resulted in F2 pods/seeds. Apart from this, sterile F1BC1seeds obtained from the cross

(sterile F1 hybrids X ICPL 85010), when planted in the field became partially fertile and they set F2BC1 seeds (Table 5). But few sterile individuals from lines F1BC1 P1, F1BC1

P4and F1BC1 P7 were backcrossed with cultivated pigeonpea varieties to find out the fertile/sterile maintainers and restorer hybrids.

 Morphology of F1 hybrids

C. lanceolatus was an erect shrub, with thick coriaceous narrow-lanceolate leaflets with short silvery to pale golden brown hair. Pods were oblong with 3-6 seeds. Pod formation was 23% when C. cajan was crossed with C. lanceolatus. Majority of the seeds were normal with exception of few semi-shrunken seeds (34%). Out of 35 seeds only 14 seeds germinated into hybrid plants under in vivo germination conditions. The plants initially grew slowly, but later normal growth was observed. Morphologically the hybrid plants had excessive growth compared to both the parents (Fig. 9 A). F1 hybrids were screened for morphological traits such as height, branching pattern, flower size and shape, pod shape and size, and seed color. Hybrids were tall and measured about 325cm (P10-F1) to 380cm (P13-F1) height which resembled the male parent C. lanceolatus (height 285cm) compared to a height of 185cm in female parent (Fig. 6). Hybrid plants resembled C.lanceolatus (male parent) with the presence of long narrow trifoliate leaves and long secondary branches (Fig. 7 B). Selfing the fertile F1 hybrids resulted in F2 pods/seeds and backcrossing them with cultivated parent ICPL 85010 resulted in BC1 pods. All the F1s were with spreading and indeterminate growth habit (Table 4). The pods of F1 hybrids resembled more of the female parent but had the presence of dense small trichomes on their pod surface making them velvety to touch as seen in C. lanceolatus.

All the hybrids flowered between 98 to 160days from the date of germination. Variation was observed with respect to flower morphology. The flower color was orange yellow in hybrids (Fig. 7 C), whereas the cultivar and C. lanceolatus had yellow flower color. Hybrid flowers had dense streaks on their keel petal, a trait transferred from the male parent (wild) (Fig. 7 C). In the female parent, flowers had no streaks on their keel petal. In comparison to the fertile plants, the anthers of male sterile plants were smaller in size and totally devoid of fertile pollen grains (Fig. 7 D). Pods were flat in C. lanceolatus compared to C. cajan pod (Fig. 8).The locules between the seeds were more prominent in hybrids (Fig. 7 F and H) and wild parent (Fig. 7 G and H) with clear cut demarcations between individual locules. Prominent differences in seed color and seed strophiole was observed between the hybrids and the female parent. Seeds of C. lanceolatus were grayish black and cultivar seeds were beige-brown, while the hybrid seeds were black in color resembling the pollen parent (Fig. 8).

 Morphology of the BC1F1and F2 population

F1BC1 and F2populations were planted in the field and the plants in both the populations segregated for morphological traits such as days to flower, plant height, number of branches, branching pattern, growth habit, leaf size and shape, stem color, flower color and size, pod color and size and seed color and shape. Observations were recorded and tabulated in the table 6 and are as follows.

 Days to flower

Both the populations (F2 and BC1) flowered in 109.8 to 138days after germination. Whereas the cultivated parent took 84days and C. lanceolatus took 178days to flower. Number of days taken for first flowering was an intermediate trait between two parents as in the case of F1 hybrids which took a maximum 160days. But F2 and BC1populations flowered 22days earlier than the F1 hybrids.

 Plant Height

Plant height was the most significant trait in the C. lanceolatus derivatives. C. lanceolatus measured 285cm and C. cajan measured 165cm (Table 4). Plant height in the

F1s was ranged from 235 to 380cm (Fig. 9 A). F1 hybrid plants were backcrossed to the recurrent cultivated parent ICPL 85010. F1 BC1 population (P-1, P- 2, P-4, P-5, P-6, P-7 and P-8) exhibited wide range of difference in the plant height varying from 72.5 to

96.5cm. BC1 plants resembled the female parent (i.e. cultivated pigeonpea) being short statured than C. lanceolatus (Fig. 9 B). In the F2 population (P-2, P-3, P-8 and P-9) too, same kind of height variation was observed which varied from 47.3 to 81.2cm (Table 6).

 Number of branches

Total number of branches was recorded for each individual plant of F2 and BC1 populations. Compared to the F1 hybrids, F2 and F1BC1 lines had less number of primary, secondary and tertiary branches. The individuals in F2 and BC1 populations exhibited 2.2 to 10.7 primary branches, 2 to 14 secondary branches and 3 to 4 tertiary branches. F2 and

BC1 population had shorter and less number of branches than the branches of F1 hybrids.

 Branching pattern

Based on the recorded observations, the F2 and BC1 populations segregated for compact (41.91%) and semi-determinate types (46.3%) of branching. Only 11.76 % of plants exhibited spreading type of branching in F2 and BC1 populations. Few individuals from the F1BC1 lines namely P5-5, P5-7, P6-3, P7-3, P7-4, P7-5, P8-1, P8-2, P8-5 and P8-

7, and two F2 lines namely P9-3 and P9 -4 exhibited complete spreading habit.

 Growth habit/Flowering pattern

Flowering pattern was completely of indeterminate type in the F1 hybrids but in contrast to it F1BC1 lines segregated for determinate, semi-determinate and indeterminate type of growth habit (Fig. 11). The flowering pattern in F1BC1 population involved 64% of semi determinate in 43 individuals (P1-1, P1-2, P1-3, P1-4, P1-5, P1-7, P2-1, P2-2, P2- 3, P2-4, P2-5, P2-6, P4-3, P4-4, P4-5, P4-6, P4-7, P4-8, P5-1, P5-2, P5-4, P6-6, P6-7, P6- 8, P6-9, P6-10, P7-1, P7-2, P7-4, P7-8, P7-9, P7-10, P7-11, P7-12, P8-3, P8-6, P10-1, P10-3, P10-4, P10-5, P10-6, P12-1 and P12-2), 28% of determinate in 18 individuals (P4- 1, P4-2, P5-7, P6-1, P6-2, P6-3, P6-4, P6-5, P7-3, P7-5, P7-6, P7-7, P8-1, P8-2, P8-4, P8- 5, P8-8 and P10-2) and 8% of indeterminate growth habit in 5 individuals (P1-6, P5-3,

P5-5, P5-6 and P8-7). Stem color was generally green in F2 and F1BC1 population as in the cultivated pigeonpea but in some of individuals of F1BC1 P1, P3 and P5 lines it was purple. About 33% of the total population had purple stem color and remaining 77 % had green stem color in F1BC1.

 Leaves

Size and shape of the leaf was another significant character which helped in hybridity confirmation of F1 plants. Wide range of shapes and sizes of the leaves were observed in F1BC1 and F2 populations (Fig. 10). Sizes of the leaves were grouped in to small, medium and large, whereas the shapes of the leaves were grouped into normal, elliptic, obovate, lanceolate, sesame and minute. The tip of the leaves was of two kinds acute and round. In different combinations of these traits, leaves of all the plants exhibited a wide array of leaf shapes. 3 individuals (P1-2, P1-3 and P1-5) from F1BC1P1 and 2 individuals (P4-1 and P4-8) from P4 had elliptic leaf shape with round tips on their trifoliate leaf. Large or medium sized obovate leaves were found on F1BC1P1-3,

F1BC1P2-3, F1BC1P4-8, and F1BC1P8-7 plants in combination with acute or round tips.

Small sesame leaves were observed on the two lines of F1BC1P6 and P8, also few individuals from the F1BC1 lines P1, P4, P10 and P12 (P1-7, P4-6, P4-7, P10-4, P12-2 individuals). Few individuals from F2 lines (P2-2, P2-3, P3-2, P3-3, P8-2 and P9-6) also had small sesame leaves with acute tips. Remaining individuals in F2 and BC1populations segregated for large/normal, medium/normal, large/lanceolate and medium/lanceolate leaves with acute tips.

 Flowers

With respect to the flower size and color, all the flowers were grouped in to small, medium and large flowers either with yellow, orange or pale yellow color (Fig. 12). Plants exhibited a range of flower morphology. Small pale yellow flowers were observed on F1BC1P1-6 and F2 P9-2 individuals. But remaining individuals from the lines

F1BC1P1, P2, P4, P5, P6, P7, P8, P9, P10 and P12 segregated for large/yellow or orange and medium/yellow or orange flowers (Fig. 12 A and 12 B). Flowers in F2 population (F2 P3 and P9) appeared as medium sized with dense streaks on the standard petal. Pattern of the streaks on the standard petal was also an important trait which was inherited from the

C. lanceolatus. Few individuals such as F1BC1P4-10, P8-4, P8-8 and F2 P3-1 had dense streaks on their standard petal. 41% of the individuals from the lines F1BC1 P1, P2, P4,

P6, P12 and F2 P2 exhibited sparse streaks whereas 38.4% of individuals from the lines

F1BC1 P5, P7, P8, P10 and F2 P8 were without streaks (Fig. 12 C and D).

 Pods and seeds

Unlike the pod shape and size in the F1 generation, F2 and F1BC1 population exhibited green pods with purple streaks just like in the cultivated parent ICPL85010

(Fig. 13). But in some of the individuals such as F1BC1P1-2, P2-1, P4-1, P5-2, P7-5 and

F2 P8-2 exhibited complete purple colored pods (Fig. 13 C). In contrast to this, F2 P3-1 exhibited only green pods with dense trichomes on pod surface as in the wild parent (Fig. 13 A). Based on the observations, seed color was grouped in to brown and black. Unlike

F1 hybrids which had all black seeds, 75% of the individuals from the lines of P1, P2, P4,

P5, P6, P7, P8, P9, P10, P11 and P12 from F1BC1 population and F2 P2, P3 and P9 had brown seeds as seen in cultivated pigeonpea parent ICPL85010.

80 ii. Crosses utilizing tertiary gene pool species A. C. platycarpus (ICPW 68) X C. cajan (ICPL 85010)  Crossability

Crosses were made between the female C. platycarpus and the male C. cajan

(Fig. 15 A and B). F1 and BC1 hybrids were possible only through embryo rescue

technique, as they aborted. Normal seed/ pod set were observed from BC2 generation

onwards (Mallikarjuna et al., 2007). Progeny from five different lines F1BC2 A, B, C, D and E from the cross C. platycarpus x C. cajan were grown in a glass house (Fig. 16). A set of 48 individuals/plants were raised from all the five different lines. Ten plants of „A‟ line (Fig. 16 A to D), nine plants of „B‟ line (Fig. 16 E and F), 12 plants of „C‟ line (Fig. 16 G and H), three plants of „D‟ line (Fig. 16 I and J) and 14 plants of „E‟ line (Fig. 16 K to P) were crossed with the recurrent parent ICPL 85010 to advance the generation

(F1BC3). Highly fertile progeny were selfed to raise the BC2F2 generation.

Among the unsuccessful pollinations, bud drop commenced within two days after

pollination. Pod set was 43.2% more in case of F1BC2 E line as they were partially sterile.

The number of pods set ranged from 26 to 132 among the F1BC2 E lines. In F1BC2A pod

set was in a range of 19 to 95 pods in line, 2 to 53 pods in B line, 11 to 46 pods in C line and 36 to 107 pods in D line (Fig. 18). Among the five lines C and D lines were highly

fertile and formed more number of self (F2) pods which ranged from 25 to 156 and 68 to70 pods respectively. Remaining three lines A, B and E lines were partially fertile but

they also resulted in few self (F2) pods in a range of 11 to 91. Total number of cross pods from cross and self pollinations were recorded and are tabulated in the Table 8. Seeds were planted in the field in the 2011 kharif season. Upon repeated backcrosses with

recurrent cultivated parent ICPL 85010, few of the individuals (F1BC4 individuals: A1,

A2, A3, A5, A7, A6, A14, B3, B6, E5 and E8) resulted in to F1 BC5 pods/seeds in a range

of 12 to 23 pods. Some of the fertile individuals (F2BC3 individuals: D1, D2, C3, C7 and

C11) resulted in 12 to 36 BC3F3 pods/seeds. Of these lines, F1BC5A5-7, B3-2, B6,

BC3F3C3-13, C7-13, C7-4, C11-3 and F2BC3D2 lines were selected and planted during

June 2012 for H. armigera screening to compare. Out of all the lines F1BC5 A5-7, B3- 2and B6 lines were male sterile plants and these lines were cross pollinated with different unrelated cultivars such as MN1, MN5, MN8, ICPL 88034, ICPL88034, ICPL 7035,

ICPL92016 ICP14444 and ICPL 85010. Crossing the F1BC5 A5-7 line with nine different

81 cultivars resulted in pod set which ranged from 1 to 8 pods per cross. In the lines of

F1BC5B3-2 and B6 pod set in a range of 1 to 8 and 1 to 3 pods/cross respectively.

Whereas in fertile lines of BC3F3 population, the number of pod set ranged from 21 to 900 pods.

 Morphology

Morphological characters of all the lines were recorded at the vegetative stage to compare with the cultivars used in the study. Test plants from advance generations F1BC3 resembled cultivated pigeonpea with respect to plant height and branching pattern. Morphological observations such as, days to flower, plant height, number of branches, branching pattern, growth habit, flowers, pods and seeds of F1BC3, BC3F2 populations are tabulated in the table 7 and 8, whereas BC3F3 and F1BC5 populations are recorded and tabulated in the Table 9 and are as follows.

 Days to flower

Among the five lines, F1BC3A line flowered in 70 to 90day and F1BC3E line flowered in 64 to71days after sowing as in the C. platycarpus which took 60 to 70days to flower. The two test lines flowered earlier than the F1BC3B line which took 112 to 130 days, F1BC3C line which took 112 to 130day and F1BC3D line which took 88 to 94days to flower (Table. 7). But in the advanced generations (F1BC5 and BC3F3) all the lines flowered in a range of 95 to 114days after sowing as in the cultivated parent ICPL 85010 which took 85days to flower (Table 9).

 Plant height

The F1BC3D line was tall and measured about 175 to 200cm in height compared to the height of 180cm in the cultivated pigeonpea. A few plants from F1BC3A line (A1,

A2, A3 and A4) were 70 to 90cm tall. In the F1BC3 B line B5, B6 and B7 plants had shorter plant height, which measured in a range of 65 to 75 cm. Among the F1BC3B lines,

F1BC3B6 plant was shorter in height which measured 65cm with trailing secondary branches (Fig. 20); the pods were small with black flattened seeds as seen in C. platycarpus. F1BC3C and E lines exhibited medium height ranging from 100 to 160cm (Table 8).

 No. of branches

Total number of branches were counted and recorded from all the individual plants separately. F1BC3 A line showed 9 to 33 primary branches and 2 to 19 secondary branches; B line exhibited 7 to 18 primary and 4 to 21 secondary branches; C line exhibited 4 to 34 primary and 8 to 23 secondary branches; D line exhibited 15 to 20 primary and 52 to 57 secondary branches and E line had 12 to 23 primary and 9 to 43 secondary branches. Tertiary branches ranged from 1 to 5 in all the plants. While in advanced generations such as BC3F2, BC3F3, F1BC4 and F1BC5, number of primary branches, secondary branches and tertiary branches were similar to the cultivated parent (ICPL85010) which had 11 primary branches, 21 secondary branches and 12 tertiary branches. Advanced generation lines, F1BC5A5 had 6.3 primary and 16.3 secondary branches, B6 had 4 primary and 9 secondary branches, BC3F3C3-13 had 7 primary and

4.3 secondary branches, two lines BC3F3 C7-13 and C11-3 had 4.5 to 5primary, 12 to

35.5secondary and 5 to 8 tertiary branches and line BC3F2 D2-1 had 9.3 primary and 8.8 secondary branches (Table 9).

 Branching pattern

Based on the observations made from all the five lines, branching pattern was grouped in to spreading, semi-spreading and compact types. Of all the five lines, F1BC3 D lines had completely spreading type of branches and E lines had completely compact type of branching pattern. However the remaining three lines F1BC3B, A and C were segregated in to spreading semi-spreading and compact type of branching. BC3F3C7-13-1, C7-13-2, C11-3-1 and C11-3-2 showed spreading habit with long branches and vigorous growth (Fig. 22). Other individuals from the lines F1BC5 A, B and BC3F3 D were with semi-spreading branching pattern as in the cultivated parent ICPL85010.

 Growth habit

The flowering pattern of all the individual plants from the lines F1BC3C and D had indeterminate growth habit. However individuals from the lines F1BC3 A, B and E segregated for 25% determinate type, 30% semi determinate type and 45% indeterminate type of growth habit. But the individuals of F1BC5 generation exhibited completely semi- determinate kind of growth habit as in the cultivated parent. Whereas, BC3F3 populations exhibited indeterminate growth habit except in BC3F3 C3 which had semi-determinate

83 habit. Trifoliate leaves were elliptical as in the case of cultivated parent but in a few individuals of the lines F1BC5 A5-7, B6, BC3F3C3-13 and C11-3 had, small sized leaves with regular elliptic leaf shape as in the cultivated parent (Fig. 21). Recorded data was tabulated in Table 9.

 Flowers

Pigeonpea flowers are papilionaceous and closed but 37% of the individuals from the lines F1BC3 A, B, C, D and E had open type of flowers and remaining 63% individuals exhibited papilionaceous type of flowers. Most of the anthers from the lines

F1BC3 A, F1BC3 Band F1 BC3 E were sepalous, transparent and devoid of a regular anther cavity and pollen grains. Some of the anthers had a miniature anther cavity with a few pollen grains. In some of the anthers, a few pollen grains appeared to be fertile, staining pink with acetocarmine. Dehiscence of the anthers did not take place to release the pollen grains. Upon closer study it was observed that anther cavities lacked the line of dehiscence and had thick anther wall, which prevented the dehiscence of anthers and release of pollen grains. Since dehiscence did not take place pollen grains were not released from the anther sacs. Anther sacs were forcefully ruptured by squashing the anthers and self pollinations were carried out. In spite of self pollination none of the plants set seeds. In all the sterile plants style of the stigma was longer than the filaments of anthers (Fig. 17). Flower color was yellow in all the generations such as F1BC3, F1BC4,

F1BC5, BC3F2 and BC3F3 as in the cultivated parent ICPL85010. But in the BC3F3C11-3-1 plant flowers were yellow with dense streaks on their standard petal (Fig. 22 E) as in case of wild parent (C. platycarpus).

 Pods

In case of cross pollinations bud drop commenced two days after pollination in all the test lines. GA3 application resulted few mature pods. Pods were normal green in color with purple streaks (Fig. 20 E). Pods obtained as a result of crosses with ICPL 85010,

F1BC3 B6 were small, green, flattened with prominent locules and black seeds inside (Fig.

20 D). In the next back cross normal green pods with purple streaks were found in F1BC4

B6 plant. But in the next generation (F1BC5B6), one individual B6-1 had black colored pods (Fig. 20 F) and remaining plants had only green pods with purple streaks as in the cultivated parent.

 Seeds

Mature pod locules were not completely filled with mature seeds. Even in the 4-5 chambered pods, 1-2 shrunken or non viable seeds were seen. One to three seeds were found in a pod. All the lines segregated for seed color. Seed color was black in F1BC3 A5,

B6, E2, E5, E6, E7, E11 and E13 (Fig. 19 A and E). It was gray color in F1BC3 C6, C7, E4, E8 and E10 (Fig. 19 L and M). Remaining lines had only brown seed coat as in the cultivated parent. Seed shape was elongate in some lines (F1BC3 C3, C5, C6, C10 and E4) (Fig. 19 F and G). Remaining lines had oval shape as seen in the cultivar parent.

B. C. cajan (ICPL 85010) X C. volubilis (ICP 15774)  Crossability

C. volubilis was used as the male parent and ICPL 85010, an extra short duration pigeonpea variety, as the female parent (Fig. 23). Large number of pollinations resulted in seed abortion although a few pods were obtained. At harvest most of the seeds were in the form of shriveled black specks. Fluorescence microscopy showed that a few pollen grains germinated on the stigma but were slower in growth when compared to the pistils from self pollinations on cv ICPL 85010 and staining with acetocarmine too showed the presence of germinating pollen grains on the stigma (Fig. 23 B). The styles started to wither between 10 to 15day of pollination. Pod set was observed in a few styles but seeds had withered within. Application of GA3 to the pollinated pistils accelerated the growth of the pollen tubes in the style and encouraged the pods to remain longer on the plant. Although most of the pods were shrunken and had withered seeds within, 1.5 % of the pods had bold and had well grown seeds within. As a result of 308 pollinations 10 seeds were obtained. From 10 seeds only one plant germinated and gave rise to F1 plant from which the subsequent generations developed.

Fertile F1 hybrid was selfed and backcrossed to ICPL 85010 which yielded F2 and

BC1 seeds (Fig. 25 A). This hybrid F2 progeny were raised in the kariff season; all the lines had short stature, with determinate flowering branches and extra early flowering. To verify short duration and compact flowering inflorescence trait, more F2 progeny were raised in rabi of 2012 and all the plants had determinate growth habit and compact inflorescence. Selfing the F2 progeny resulted F3 seeds and F3 lines were planted in the field in mid June 2012 (Kharif season). Seeds obtained from the F3 progeny lines were planted in the glass house to raise F4 population from it (Fig. 28).

 Morphology

Cultivated female parent used in the crossing program was a medium sized shrub with normal leaves and male wild parent was a climber with thin membranous leaflets, with large orange yellow flowers and bulbous based hairs on the calyx, and pods with short pubescence. F1 hybrid derived from the cross and its subsequent generations (F2, F3 and F4) were observed for selected morphological traits and observations were recorded and tabulated in Table 10 and 11.

 Days to flower

53 F2 plants along with their parents (ICPL 85010 and ICPW15774) were evaluated for the morphological traits such as days to first flowering, plant height, number of branches per plant, days to 50% pod maturity and number of pods/plant). Male (ICPW 15744)) and female (ICPL 85010) parents flowered in 82days and 142days respectively from date of sowing. The F1 hybrid was a super early short duration plant which flowered in less than 70days after sowing. All the F2 plants flowered between 50 to

70days from the date of sowing (Fig. 24 A). Flowering was significantly earlier in the F2 plants as compared to the cultivated and C. volubilis. The F2 population‟s flowering time was similar to a short duration pigeonpea cultivar MN5 which took 68day to flower.

Eighty five percent of F2population flowered earlier than MN5 and only 15% plants flowered at similar time as that of MN5 (Fig. 27). In the F4 generation, all the individuals flowered between 71 to 96days after sowing.

 Plant height

F1 hybrid derived from this cross C. cajan X C. volubilis was 155cm in height

(Fig. 24 A) and F2 plants between 15 to 40cm in height, while the female parent was

185cm and male parent was 325cm in height (Fig. 24 B). All the F2 plants were significantly shorter than the short duration cultivated pigeonpea MN-5 (45cm) which is a dwarf, being short statured with determinate growth habit. Even in the F4 generation all the individuals continued to exhibit the short plant height which measured in the range of 26 to 56cm.

 Number of Branches

The F1 hybrid was short statured but comparable to the female parent ICPL 85010, with small primary and few secondary branches. None of the F2, F3 and F4 generation plants had tertiary branches. While most of the F2 plants had only primary branches ranging from 1-6 except for few plants such as 50B, 56A and 69A which had 1-2 secondary branches. Few of the F2 plants (60A, 62A and 65A) only had the main stem which turned into a determinate raceme without any primary branches. In contrast the cultivated female parent had 11 primary, 21 secondary and 12 tertiary branches while the wild (male) parent was a creeper with 9 primary, 18 secondary and 9 tertiary branches.

The F4 lines also had 1 to 8 primary and 2 to 6 secondary branches without tertiary branches. A few F5 generation plants were raised and they continued to show the short stature.

 Branching pattern

Based on the morphological observations of F1, F2, F3 and F4 generations, it was clear that all the individuals were erect with compact type of branching pattern.

 Growth habit, flowers and pods

F2, F3 and F4 generation plants segregated for determinate and semi determinate types of flowering pattern. In most of the individuals, the apical meristem terminated in to a terminal inflorescence (Fig. 26 A). Unlike the general pattern of semi determinate growth habit, in these plants it did not appear as semi determinate type of growth habit due to small and short branches. In gross morphology plants resembled the determinate type, but in reality they segregated into semi-determinate and determinate types which were confirmed by utilizing growth habit specific primers in F2 population.

Leaves were normal in size and green in color. F2, F3 and F4 plants flowered profusely in a terminal clusters. Flowers were yellow in color except in three individuals such as F4 P1-3-7, P1-6-12A and P1-6-12B they appeared as pale yellow color (Fig. 26 A). There were no streaks on the standard petal. Pods were green colored appeared as terminal clusters with five to nine pods per cluster (Fig. 26 C). Each pod in cluster had 3 to 4 seeds. Shape and size were like in cultivated parent.

DISCUSSION

 Crossability and morphology o Species crossability

Pigeonpea is an important protein rich food in the vegetarian diet. In most developing countries, smallholder, dry land farmers derive their protein needs from legumes. The resource poor farmers of the regions need a crop, which would not only provide food security but also improve their soil to sustain productivity. In the harsh, semi-arid tropical environments, the choice of the cultivars is limited to a few crops only. Among these, pigeonpea occupies a prime place due to its drought tolerance and multiple uses. Pigeonpea yield has reached a plateau and is susceptible to arrange of diseases caused by virus, fungi and bacteria (Mallikarjuna et al., 2011). Although high degree of morphological variability is seen, the same is not true at the molecular level (Yang et al., 2006). Crop improvement programs are looking for increased genetic diversity by tapping wild relatives from different gene pools. Earlier attempts at interspecific hybridization of pigeonpea were mostly concerned with crosses involving C. lineatus (Deodikar and Thakur, 1956; Kumar et al., 1958; Kumar and Thombre, 1958; Reddy and De, 1983); C. sericeus and C.scarabaeoides (Sikdar and De, 1967: Reddy, 1981; Pundir, 1981 and Ariyanayagam, 1994). Pundir (1981) used C. albicans, C. trinervius and C. cajanifolius. Dundas (1984) used C. reticulatus, C. pluriflorus and C. acutifolius. C. platycarpus was successfully crossed by Mallikarjuna et al. (1995, 2002, 2003, 2011, and 2012). There have been a couple of successfu1 attempts by Mallikarjuna et al. (manuscript under preparation) at crossing R. bracteata and R. sublobata (quaternary gene pool) with C. cajan in intergeneric hybridization. In spite of the success obtained in the utilization of wild relatives from the secondary and tertiary gene pools, there is ample scope to use others, which has not been attempted in the crossing program.

In the present study, two wild relatives (C. cajanifolius and C. lanceolatus) from the secondary gene pool and two from the tertiary gene pool (C. volubilis and C. platycarpus) were used in the interspecific hybridization. Amongst these, C. lanceolatus from secondary gene pool of pigeonpea and C. volubilis were used for the first time in crossability experiments and successfully developed fertile hybrids. Five different progeny lines, F1BC2 A, B, C, D and E of the cross C. platycarpus X C. cajan (Mallikarjuna, 2007) were also included in the present study as they were not studied for

88 any trait and to advance them further by backcrossing with cultivated parent. Varying degree of success has been achieved among all crosses. Out of these four wide crosses of the present study, the degree of species crossability was the highest in crosses involving C. cajanifolius as the pollen parent followed by those involving C. lanceolatus as they are closely related to cultivated pigeonpea. The Australian species of C. lanceolatus have been crossed successfully with cultivated pigeonpea for the first time. Earlier it was used by Sateesh kumar (1985) but the F1s died in the vegetative stage itself. The cross between pigeonpea and C. lanceolatus has generated two categories of progenies. The first one being the fertile progeny with good recombination between the parental genomes leading to fertile plants, a good material for broadening the genetic base of pigeonpea and maybe look for traits of interest when the progenies lines are developed from these. The second one being the development of CMS (A9) utilizing those F1 hybrids with 100% male sterility. Progress has been made to exploit and introgress useful traits including male sterility from C. lanceolatus, a secondary gene pool wild relative of pigeonpea.

Cajanus volubilis Blanco, a native of Asia, belongs to the tertiary gene pool of pigeonpea (Mallikarjuna et al., 2011) and the classification of species into different gene pools is based on their crossability with cultivated pigeonpea. Except for C. platycarpus, none of the other species in the tertiary gene pool have been successfully crossed and fertile progeny lines obtained (Mallikarjuna et al., 2006, 2012). Wanjari et al. (2001) reported crossing C. volubilis as the female parent and recovered male sterile plants, but it was not possible to advance them further to due complete sterility of the progeny lines (Mallikarjuna et al., 2012).

The secondary gene pool has contributed various traits for the improvement of the crop. There is enough evidence to prove that C. cajanifolius is the progenitor species of pigeonpea (Mallikarjuna et al., 2012,). There are many species in the secondary and tertiary gene pool of the genus Cajanus. Many of them have not yet been crossed with pigeonpea.

o Morphology of Interspecific derivatives

Hybrids were intermediate sharing the traits of both the parents involved in the cross. The morphology of the derivatives confirmed the true hybridity between male and female parents. Morphological traits of the F1 hybrid was more skewed towards the female parent and such phenomenon is not new when distant genomes are made to come

89 together through wide hybridization. Cajanus volubilis is a wild relative of pigeonpea placed in its tertiary gene pool (Mallikarjuna et al., 2011; Bohra et al., 2010). Genomic studies have also shown its distant relationship with cultivated Pigeonpea (Pangaluri et al., 2007). Genome wide transcription analysis in synthetic Arabidopsis allotetraploids showed that expression patterns from one genome could be dominant over the other genome (Wang et al., 2006). Pumphery et al. (2009) found that a small percentage of hybrids between wheat and synthetic hexaploids were similar to one of the parents. We report for the first time in pigeonpea that such a phenomenon is taking place in the hybrid between C. cajan (cultivated pigeonpea) and C. volubilis with the morphology of the F2 hybrids skewed towards the female parent. Genetic control in storage proteins has been observed in allopolyploid wheat. Galili and Feldman (1984) showed that inactivation of endosperm-protein is brought about by an inter-genomic suppression. Wheat genome driven control of some agronomic, pest and disease resistance was observed in wheat. Peng et al. (2000) observed that R-gene cluster in the B genome of wheat and high marker clustering in the B genome than the A genome is the result of expression of genome asymmetry. The ability of one genome to suppress the activity of genes in another in newly formed hybrids with different genomes may be to prevent defective organ formation/phenotype. This may be a protective mechanism to obtain viable plants. Whereas the derivatives of C. lanceolatus and C. platycarpus were more similar to the wild parents in the early generations, but after backcrossing with cultivated parent they exhibited similar morphology with cultivated parent. C. cajanifolius derivatives were more or less similar to the cultivated pigeonpea as the parents were indistinguishable in morphology except for pod shape and size and seed strophiole. o Inheritance of qualitative and quantitative characters from parents to progeny  Days to flower

Time from sowing to flowering is determined by three factors: response to photoperiod, which is usually the most important factor; response to temperature; and maturity genes that determine the minimum time to flower (Roberts and Summerfield, 1986; Wallace and Yan, 1998). Medium and early flowering is considered as the desirable. All the derivatives were of medium duration flowered in the medium number of days. Interspecific derivatives derived from C. cajanifolius, C. lanceolatus and C. platycarpus flowered in 100 to 120days after sowing as in the cultivated parent. C.

90 lanceolatus derivatives flowered late in early generations but in the advanced backcross generations they flowered in medium duration. Hence these derivatives were considered as medium duration pigeonpea lines. Whereas C. volubilis derivatives took less number of days to flower compared with the existing extra short duration cultivars MN1, MN5 and MN8. Early flowering pigeonpea germplasm are desirable and development of short statured (dwarfs) early germplasm will have an impact on pigeonpea as a crop. Early flowering pigeonpea will have a place in rice wheat rotation systems replacing rice as availability of water becomes scarce during certain seasons (Vales et al., 2012). Also short duration pigeonpea would fare well as the soil would have only residual moisture. Exploitation of hybrid vigor, restructuring of plant type and early maturity are targets for increasing pigeonpea productivity per unit area and time (Saxena and Sharma, 1981a; Saxena, 2008). Kumawat et al. (2012) have reported first QTL mapping of plant height and maturity time amongst a few more traits. Identification of QTLs for these two traits will play a major role in molecular breeding of pigeonpea.

 Plant height

Among four interspecific derivatives, C. cajanifolius and C. platycarpus derivatives attained normal height as in the cultivated parents. Interestingly, C. lanceolatus derivatives were too tall measured more than 300cm in the F1 generation but later they achieved normal height upon backcrossing. In contrast, C. volubilis derivatives were too short and measured less than 45cm height. These derivatives maintained the dwarfism even in the advanced generations F4 and F5. Mutations in the biosynthesis or signaling pathways of gibberellin (GA) can cause dwarfing phenotypes in plants, and the use of such mutations in plant breeding was a major factor in the success of the green revolution (Muangprom et al., 2005). Dwarfism is a desirable characteristic for many agricultural plants. In grain crops, dwarfism can reduce lodging and increase harvest index, and the breeding of dwarf wheat (Triticum aestivum) and rice (Oryza sativa) cultivars was a major factor in the success of the Green Revolution (Khush, 2001). Dwarfism is often caused by mutations in genes controlling the biosynthesis or signaling pathway of the plant hormone GA (Peng et al., 1999; Hedden, 2003; Sun and Gubler, 2004). The most widely utilized semi-dwarf wheat cultivars in agriculture contain the Rht-B1b or Rht-D1b allele that encodes a mutant form of a DELLA protein, a GA signaling repressor (Peng et al., 1999; Hedden, 2003). We hypothesize that such a phenomenon might have occurred in the dwarfs obtained from the cross C. cajan x C.

91 volubilis. As the F2 plants were dwarfs and continued this trait in the F3 and F4 generation too. Further generations will be tested to verify dwarfism coupled with high pod number. The experiment also shows that novel plant type can be obtained when wild species are crossed with cultivated, which should be exploited for pigeonpea improvement. It would also be interesting to see if more number of novel phenotypes can be recovered if large numbers of F1 hybrids are generated utilizing C. volubilis in the crossing program. Although, lodging is not a problem in pigeonpea, it might increase in the near future with the breeding orientation aimed towards improving the harvest index. The strategy of improving the harvest index by increasing seed yield and reducing biomass may lead to the development of new high yielding varieties with weak stems susceptible to lodging. Hence, development of compact dwarf plant types in pigeonpea and their utilization in breeding programs could alleviate such problem in the near future.

 Branching pattern

Primary, secondary and tertiary branches determine the development of canopy, fruiting node and are related yield. Among all the crosses, C. lanceolatus derivatives had long spreading branches, whereas C. volubilis derivatives had too short branches. Secondary and tertiary branches occurred in all the three kinds of derivatives except in C. volubilis derivatives. Secondary and tertiary branches were completely absent in C. volubilis derivatives with compact plant stature. Although C. volubilis derivatives were small with short primary branches, they resulted in more pods per plant than in the other cultivated varieties viz. MNI, MN5, MN8, ICPL 85010, ICPL 88039, ICPL14444 and ICPL 85030. C. cajanifolius derivatives exhibited spreading habit in most of the lines viz., BC1F3 129-1, 130-1, 131-1, 132-1 and 133-1, but remaining lines exhibited semi- spreading to compact nature of plant. The same trend appeared in C. platycarpus derivatives exhibited spreading type and others with semi-spread to compact types. But after few more advance generations all these derivatives may have stable branching pattern as in the cultivated pigeonpea. Branching pattern may have an impact on yield.

 Growth habit

All most all derivatives were segregated for determinate, semi-determinate and indeterminate growth habits. But in the C. lanceolatus all the F1s were with indeterminate flowering pattern. This trend completely changed in the advance backcross generations where all the kinds of growth habits appeared. In C. cajanifolius and C. platycarpus

92 derivatives indeterminate (IDT) and semi-determinate (SDT) pattern of flowering appeared. But in case of C. volubilis derivatives, more number of determinate (DT) plants identified. As the DT is an important agronomic trait in several important crops including pigeonpea, the molecular expression of this trait was validated by using growth habit specific markers specially developed to amplify the DT/IDT alleles. Genes in the TERMINAL FLOWER1 (TFL1) /CENTRORADIALIS family are important key regulatory genes involved in the control of flowering time and floral architecture in several different plant species (Foucher et al., 2003). In plants, identification and analysis of MADS box genes revealed that they involved in flower development in several plant species, including gymnosperms (reviewed by Ma and De Pamphilis, 2000).The transition from vegetative to reproductive phases at the shoot apical meristem (SAM) is controlled by the interaction of positive and negative regulators, such as LEAFY (LFY), APETALA1 (AP1), and TERMINAL FLOWER1 (TFL1; for review, see Benlloch et al.,2007).

 Variation in leaf shape and size

In the cross C. cajan x C. lanceolatus hybrids an interesting feature was the variation for leaflet shape and texture. Whereas in remaining derivatives leaf shape was more or less similar to the leaf of cultivated pigeonpea. Leaf shapes were minute to lanceolate in C. lanceolatus derivatives. Among all derivatives C. lanceolatus derivatives showed dense trichomes on their leaf surfaces. Since C. lanceolatus is drought resistant variety due to its long narrow leaf shapes (lanceolate), the derivatives of this wild species also might have the same trait (Sateesh Kumar, 1985). But these derivatives were not evaluated for the drought resistance.

 Flower morphology

Pigeonpea has a typical papilionaceous flower. The flower is irregular (zygomorphic) and is made up of five petals, a standard or vexillum, two wing petals, and two petals fused together to form a keel-like structure that encloses the anthers and stigma (Fig. 1 A). Although the structure is most suited for self pollination, in pigeonpea a certain amount of cross pollination does occur with insect visitations (Saxena et al., 1990). In the segregating population from the cross C. platycarpus x C. cajan ICPL

85010, significant variation in flower morphology was observed in F1BC3 progeny. Both open and closed flowers were observed in male sterile derivatives with sepalous anthers. Such chasmogamous flowers encourage cross pollination as the pollinating agents have

93 free access to pollen grains in the anthers and the stigma (Lord, 1981). In case of C. lanceolatus derivatives, male sterile plants showed normal flowers with brown shriveled anthers. Similar kind of observation found in C. acutifolius derivatives (Mallikarjuna et al., 2005). In the past, the natural outcrossing was considered as a negative trait due to its role in the contamination of cultivar purity. A lot of importance is being given to this trait for its potential role in hybrid pigeonpea research and the development of cytoplasmic male sterile systems (CMS) (Tikka et al., 1997; Saxena and Kumar, 2003; Mallikarjuna and Saxena, 2005). In all the CMS systems, cross pollination is essential for hybrid seed set.

Morphology of pigeonpea Flower. (a) stigma. (b) stamen. (c) filament and (d) sepals.

 Pods and Seeds

Pods were normal in case of C. cajanifolius and C. platycarpus derivatives except in few derivatives in which pods with mixed green and purple color. In C. lanceolatus derivatives, mostly purple colored pods with dense trichomes were seen in early generations but in the advanced generations pod color and shape were normal as in

94 cultivated pigeonpea. C. volubilis derivatives had normal pods and seeds. Seeds were brown and black in C. lanceolatus and C. platycarpus derivatives.

2. Cytological analysis

Results

i. C. cajan - C. cajanifolius derivatives

Fixed buds of F1 hybrid were used to study the meiotic behavior of chromosomes in the C. cajan X C. cajanifolius hybrids. A considerable number of microsporocytes exhibited normal (2n=11) bivalent formation at metaphase I (Fig 29 A). This indicated high degree of homology between the parents involved in the cross and no major structural aberrations was evident from the pairing behavior (Fig. 29). Thirty percent of the meiocytes showed the formation of 7 bivalents and 2 tetravalents showing two chromosomes which were homologous between the parental species (Fig. 29 B). The meiocytes showed the presence of 2 univalents which indicated that one chromosome in each parent did not have a homologous chromosome in the other parent or the divergence of one chromosome in one of the parents. Such anomaly was present in 5% of the meiocytes. Anaphase I analysis too showed the presence of 1 to 2 laggards (Fig. 29 D). Even in anaphase II, 4% of the cells showed 2- 5 lagging chromosomes. Only 0.5% of the cells showed three poles in anaphase II instead of regular formation of four poles (Fig. 29 G). Though there were few abnormalities in the interspecific derivatives, 100% tetrads were observed (Fig. 29 H). Pollen fertility in the F1 hybrid was more than 75%, showing closer relationship between the two species (Fig. 29 I). But BC1F3 lines exhibited some variation in the plant morphology such as short plant height with small flowers as well as low pollen fertility. Such individual plants were taken to study the meiosis for identifying the possible chromosomal abnormalities. Flower buds were fixed from 18-5-1, 18-5-2, 37-1-3, 37-1-4, 76-2-1, 76-2-2, 76-2-3, 82-2-1, 82-2-2, 82-2-3, 82-2-5 and 82-2-7 (Table 12). Analysis was carried out by studying the chromosome behavior in about 20 microspore mother cells of an individual plant.

 Metaphase

In early metaphase, chromosomes were sticky and were connected by thin strands. At metaphase I most of the cells were difficult to analyse because of over-contraction and

95 clumping of chromosomes (Fig. 30 A). Metaphase I plate of the sampled lines showed 7 to 11 bivalents which ranged from 9.7 to 10.55. Bivalents appeared as rings or rods (Fig.

30 I and J). In the individuals of F3BC1lines, the mean univalent chromosome configuration was found in the range of 0.15 to 0.8 (1 to 4 univalents) (Fig. 30 A, F and I). Univalents appeared as a result of non-homology between one or two parental chromosomes. Multivalents like trivalents and tetravalents were also present but rare in occurrence. Trivalents and tetravalents ranged from 0.05 to 0.35 (1 to 2 multivalents/cell).

 Anaphase

In anaphase I, univalents were randomly distributed and some univalents were included in the polar groups, while some others became oriented on the equatorial plate. In late anaphase, there were lagging univalents (Fig. 30 E). Depending on the equal/unequal disjunction of chromosomes and presence of univalents in the cells, the pollen mother cells were grouped as normal and abnormal cells. These were analysed based on the normal and abnormal distribution. Even in the anaphase I, 1 to 2 laggards were found away from the polar groups (Fig. 30 G). Among these individuals, 25 to 80% of the cells were found with normal distribution but the remaining 20 to 75% of the cells were with abnormal distribution (Table 12). The number of the lagging univalents (1 to 2) at anaphase I was about half the total number of univalents (2 to 4) found at metaphase I.

 Tetrads

At the telophase II some abnormal disjunctions were found with one or two laggards (Fig. 30 H and K). Normal cytokinesis after telophase II resulted in 100% normal tetrads. However, these tetrads became dimorphic microspores when released from pollen mother cell wall.

 Pollen fertility

Pollen fertility was analyzed based on the number of fertile and sterile spores. Due to the presence of chromosomal abnormalities the percentage of fertile pollen grains decreased. Pollen grains were dimorphic with small as well as large grains. Alexander stained pollen grains appeared as pink as well as green spores. Pink grains were counted as fertile pollen, whereas green spores were considered as sterile pollen. As expected the percentage of the pollen fertility varied and ranged from 52 to 88%. However, in other lines BC1F3 29-1, 131-1, 132-1 and 133-1 it was more and almost 100%. 96 ii. C. cajan - C. lanceolatus derivatives

A study of meiosis of the interspecific hybrids between C. cajan and C. lanceolatus was undertaken. Fourteen plants were studied, five of which were completely male sterile. However, the information presented here is compiled from 14 plants (Fig. 31).

 Metaphase

Pollen mother cells of the hybrids exhibited regular formation of eleven bivalents which were predominantly rings. The frequency of metaphase configuration was presented in Table 13. It is clear from the data that the number of bivalents ranged from 7 to 11 (Fig. 32, 33 and 34). Bivalents appeared either as rings or rods (Fig. 32 A, I, J and Q). A number of univalents were also found in many cells, and the average number of univalents per cell varied greatly, ranging from1 to 5 (Fig. 32 Q and 33 A, K, P). The average number of univalents and bivalents ranged from 0.35 to 1.35 and 9.4 to 10.3 respectively. Meanwhile, multivalents (i.e. trivalents and tetravalents) appeared at a low frequency, which ranged from 0 to 2.

 Anaphase

Meiotic anaphase I showed 25 to 65% of pollen mother cells with normal disjunction (Fig. 32 S) and remaining 35 to 75% with abnormal disjunction (Fig. 32 B, I and R; 33 B, C, G, I, Q and R; 34 D) of chromosomes except in P10 in which 80% of the meiocytes had abnormal disjunction of chromosomes at anaphase I (Fig. 31). Univalents of metaphase continued as laggards in anaphase I (Fig. 34 D). These laggards ranged from 0 to 5 per cell. Normal disjunction of chromosomes indicated the high degree of homology between the genomes of both the parents involved in the cross. Generally in hybrids, the frequency of abnormalities at metaphase I was lower than that at metaphase II. Irregular chromosome segregation was seen in anaphase II and telophase II (Fig. 32 D, E and 34 F). Anaphase II also showed 1 to 4 laggards in pollen sterile individuals such as P1, P4, P7, P10 and P12.

 Tetrads

At the tetrad stage 100% normal tetrads were observed in all hybrids except in P7 in which 6% of tetrads contained micronuclei (Fig. 33 S). The chromosome association

generated several micronuclei in telophase II which remained in the tetrads. In the fertile hybrids these tetrads further developed in to mature pollen grains. Whereas in the sterile lines, development of tetrads was normal, but none of them formed pollen grains. Instead they grouped together and the tetrads did not separate into individual pollen grains (Fig. 32 G). Microspores lost their contents when they were in the pollen mother cell wall and became empty sacs (Fig. 32 G).

 Pollen fertility

Pollen fertility was counted based on the total number of fertile grains in the entire ten anthers of a flower. Pollen fertility was found to vary between 35 to 50% in fertile

hybrids (Fig. 32 P, X; 33 J, O and 34 J). In some of the F1 hybrids (P1, P4, P7, P10 and P12) total male sterility was observed with all the anthers having 100% sterile pollen grains (Fig. 32 H, 33 E and T). An important observation was that male sterility was post meiotic process. Anthers stained with Alexander‟s stain showed partial fertility with pink and green colored pollen grains in partially fertile hybrids P2, P3, P5, P6, P8, P9, P11,

P13 and P14 (Fig. 32 N, O; 33 I, N and 34 I), whereas sterile anthers stained green with

100% pollen sterility in hybrids such asP1, P4, P7, P10 and P12 (Fig. 32 H and 34 E, T). iii. C. platycarpus - C. cajan derivatives

To summarize different stages of meiosis namely metaphase, anaphase and tetrads, a minimum 20 well spread cells were taken into account for studying chromosomal associations. Cytological analysis of the 48 individuals from five advance

generations (F1BC3A, F1BC3B, F1BC3C, F1BC3D and F1BC3E lines) progeny lines showed normal chromosome configuration. Most of the progeny lines showed normal 11 bivalents at metaphase, equal disjunction at two poles in anaphase I, which resulted in normal tetrads.

 Metaphase

The frequency of metaphase configuration is given in the Table 15. Two kinds of bivalents were found in almost all the individuals. The present results indicate that a high

degree of pairing occurred between the two haploid genomes of the parents (BC2F1- female parent and ICPL 85010-male parent) that were not completely homologous. The unpaired chromosomes appeared as univalents. The mean configuration of univalents ranged from 0.05 to 0.9. Among all individuals, 1 to 4 univalents were found in each cell.

But in few individuals such as A7, B1, B2, B4, B7, C3, C5, C6, C7, C9, C10, C11 and E13 univalents were not found, instead they had only bivalents and tetravalents (Fig. 35, 36, 37 and 38). Usually 6 to 11 bivalents per cell were found in all individuals. As mentioned in previous results, bivalents were grouped as ring and rod. The mean configuration of ring and rod bivalents ranged from 4.05 to 9.85 and 1.7 to 9.35 respectively. But in few individuals B5, B6 and D2, only 1 to 2 ring bivalents were found (Fig. 38 and 39). The type of the bivalent, whether it was a rod or a ring, depended on the number of chiasmata present which was conditioned by the degree of homology and perhaps other genetical factors. Trivalents and tetravalents were found in a range of 0.05 to 0.65 (1 to 2 multivalents/cell).

 Anaphase

During anaphase separation of ring and rod bivalents led to the movement of chromosomes towards the poles (Fig. 35, 36, 37 and 38). In the early anaphase stage, chromosomes were less contracted. In anaphase, some univalents were included in the polar groups, while others became oriented on the equatorial plate. Among these individuals 30 to 95% of the cells had shown normal disjunction of chromosomes at polar ends, and 5 to75% of the cells showed abnormal disjunction between two poles. Univalents of the metaphase continued as laggards and were found in the range of 1 to 5.

 Tetrads

Normal tetrads were found in most of the individuals (Fig. 39A) but occasionally some micronuclei were found in few individuals. Tetrads often showed the formations of micronuclei. In few plants such as F1BC3 A1, B1 and E7, 1-4 micronuclei were present in the tetrads (Fig. 39 B to F). Apart from this, in F1BC3B4 and E7, few diads and triads along with normal tetrads were observed in 6 to 10% of the tetrads (Fig. 39 G to K). The presence of univalents in metaphase stage, diads, triads and presence of micronuclei in tetrad stage was in low frequency.

 Pollen fertility

In progeny lines F1BC3A1, F1BC3A2, F1BC3A3, F1BC3B1, F1BC3B2,

F1BC3B6,F1BC3 B7, F1BC3B9, F1BC3E3, F1BC3E4, F1BC3E6, F1BC3E7, F1BC3E9,

F1BC3E11and F1BC3E12 post tetrad stage, microspores lost their contents and turned sterile. Alexander‟s stain was used to check pollen viability of these anthers, which 99 showed green sterile microspores in transparent microsporangium. The epidermis and tapetum of microsporangium was thick and unable to burst open at stomium region. As a result the anthers even after anthesis stage were intact with thick cell wall layers enclosing sterile and few fertile microspores trapped inside the microsporangium.

F1BC3B6, B7 and B9 plants showed lower fertility, which varied from 3.5 to

5.9%. F1BC3 B1 and B2 plants had sterile pollen to begin with and pollen fertility ranged between 5.3 and 5.9%, but later pollen fertility improved and varied between 27% and

44%. Also F1BC3 C lines were highly fertile and their fertility ranged from 75 to 97.4%.

Again F1BC4 B6 was raised in the glass house and its pollen fertility rate was 7.5% only (Fig. 40). Alexander stained anthers showed sterile and fertile grains inside the anther. In sterile plants, anthers were sepalous and irregular in shape (Fig. 41 and 42). iv. C. cajan - C. volubilis derivatives

In the interspecific hybrid from the cross C. cajan X C. volubilis, many of the pollen-mother cells could not be analyzed and appeared to be physiologically disturbed. Often, the bivalents were over-contracted or stretched and clumped together (Fig. 43 G and H).

In the metaphase 7 to 11 bivalents were present with mean chromosomal configuration of 9.95 (Fig. 43 A, B and C). Univalents were present in a range of 1 to 3 with mean configuration of 0.85 (Fig. 43 C and H), whereas multivalents such as trivalents and tetravalents were found in between 0 to 1 per cell. Anaphase I showed only 25% cells with normal distribution of chromosomes and 75% of the cells showed abnormal distribution of chromosomes. Laggards were found in a range of 1 to 4 per cell. Most of the cells showed bridges at anaphase I (Fig. 43 E and F). But anaphase II was normal with equal distribution of chromosomes at four poles. Normal cytokinesis resulted in normal tetrads with four microspores (Fig. 43 I). As a result of this, 100% fertile pollen was found in the hybrid (Fig. 43 K).

However, two individuals such as P1-6-9 and P1-3-8 were with small sized and pale yellow flowers which was a abnormal type in the F4 generation were fixed for meiotic analysis (Fig. 44 and 45). These individuals showed 10 to 12% of the cells with abnormal chromosomal association at metaphase I. Normal chromosomal pairing was also found in more than 85-90% of the cells. Some of the meiocytes from the sample F4

100

P1-6-9 exhibited more than one nucleus (2 to 5 nuclei/meiocyte) (Fig. 44 D, E, F and G). Even in the anaphase I, 20 to 30% of the cells showed abnormal disjunction of the chromosomes. Some of the cells at anaphase II exhibited fragments, bridges and lagging of univalents. Although, some abnormalities were found at metaphase I, anaphase I and anaphase II (Fig. 44 I and J), normal cytokinesis after anaphase II resulted in normal tetrads. Pollen fertility was also found to be normal at more than 80%.

DISCUSSION

Interspecific hybridization provides information on phylogenetic relationships between any two species giving clues with regard to evolutionary patterns. Often generation of such information is based on the cross compatibility, chromosome association and pollen fertility. Such information also helps in developing breeding strategies for introgression of genes from related species in to economically useful species. As it creates genetic variation, it has great potential for plant improvement (Goodman et al., 1987 and Choudhary et al., 2000). Cytological analyses are usually performed to evaluate the meiotic process in experimental hybrids. Species with close genetic affinity produce hybrids with regular chromosome pairing, while the hybrids of those more distantly related species have meiotic irregularities and are sterile (Marfil et al., 2006). Meiotic analysis of chromosome association in the F1 generation shows the homology between the respective pairs of chromosomes.

Meiotic studies in the interspecific hybrids revealed abnormal chromosome pairing with irregular bivalent formation at early metaphase I and unequal separation of chromosomes in anaphase I in hybrids derived from four wide crosses. In the derivatives of C. cajanifolius and C. lanceolatus, 7 to 11 bivalents were common at metaphase I. As in the parents, ring bivalents predominated in these derivatives. BC1F3 population of C. cajanifolius showed 100% pollen fertile lines (129-1, 130-1, 132-1 and 133-1) except few individuals of 18-5, 37-1, 76-2 and 82-2 which exhibited abnormal male meiosis and partial pollen fertility. C. cajanifolius derivatives showed 1 to 4 univalents, whereas C. lanceolatus derivatives showed 1 to 5 univalents per cell. The presence of univalents in the derivatives of C. cajanifolius and C. lanceolatus shows that the genome of C. cajanifolius is more divergent from the C. lanceolatus. More lagging chromosomes (1 to 5 per cell) were found in C. lanceolatus derivatives than in C. cajanifolius derivatives. Diads and triads are the resultant of absence of cell wall formation in telophase II in C.

101 lanceolatus derivative F1 P7. Since the meiotic abnormalities like laggards/lagging chromosomes or fragments were lost in anaphase I, anaphase II was normal with equal distribution of chromosomes at four poles.

Same trend appeared in case of C. platycarpus derivatives where 1 to 4 univalents found in the F1BC3 population and its subsequent generations such as F1BC4 and F1BC5.

Mallikarjuna et al. (2006) reported that meiotic analysis of F1 hybrid (C.platycarpus X C.cajan) showed 5 to 8 univalents, 7 to 9 bivalents per cell and theoretically it had 50% of wild species and 50% of cultivated genome with 8% pollen fertility. The BC1hybrid theoretically had 75% of cultivated genome and 25% of the wild genome with improved pollen fertility which was 26% and BC2 hybrid with 12.5% wild genome showed 38% pollen fertility (Mallikarjuna et al., 2006). In agreement with these results

F1BC3population even with <10% of wild genome also exhibited 1 to 4 univalents and 1 to 5 laggards at anaphase I. But the pollen fertility increased from 40 to 98.3% as result of normal tetrad formation in >40% of PMCs due to equal disjunction at anaphase II and telophase II. While in C. volubilis derivative, 1 to 3 univalents were common in each

PMC as F1 also was a resultant of distant hybridization (C. cajan X C. volubilis). Usually normal tetrads were found in both the derivatives except in few individuals of C. platycarpus derivatives diads and triads were found. In case of tertiary gene pool species, pollen fertility found to be more in C. volubilis derivatives where the cultivated pigeonpea (C. cajan) is used as the female parent, whereas in C. platycarpus derivatives it was less because the cultivated pigeonpea is used as male parent (ICPL85010).

Unequal chromosome separation might be a physiological phenomenon caused by disturbances in the cytochemical balance reaction (Rao and Lakshmi, 1980). The lagging chromosomes are probably the univalents because their frequencies are comparable with each other. Though regular 11 to 11 separation is seen in most of the cells at anaphase I, a certain frequency of bridge formation was encountered which may be due to cryptic structural alterations as a consequence of crossing over with paracentric inversion.

Multivalent chromosome association at metaphase revealed genome affinity between both parental species in the hybrids, suggesting some possibility for gene/DNA homology. Analyses of meiocytes at metaphase I in the interspecific derivatives of C. cajanifolius and C. lanceolatus showed that some chromosome association could have occurred among the parental genomes, but in low frequency. No more than four

102

quadrivalents/tetravalents were found in these derivatives but one tetravalent was the most common multiple chromosome association. Thus, we can assume that gene introgression can occur in the hybrids, but in low frequency. A certain frequency of laggards and anaphase bridges was observed in all derivatives. Such disturbances were usually attributed to the physiologically unbalanced conditions resulting from an interaction of the two parental genotypes in the hybrid.

Chromosome pairing can be used to access genomic relationships between species. This information allows predictions to be made of the likelihood of transferring genes from one species to another. From accumulated results obtained through cytological studies in hybrids, it is evident that cytogenetic analyses are of prime importance in determining which genotypes can continue in the process of cultivar

development and which can be successfully used in breeding programs.

3. Identification of a CMS system A9 from C. lanceolatus RESULTS i. Cytology of sterile lines derived from the cross between sterile F1 hybrid and pigeonpea cultivars

Cytological study was carried out to confirm the chromosome number and meiotic

abnormalities (if any) in fertile and sterile F1BC1 hybrids derived from the crosses

between sterile F1 hybrid and other cultivated pigeonpea cultivars. The results showed normal chromosome pairing in both the fertile as well as sterile hybrids with regular bivalent (n=11) formation in metaphase I (Fig. 46 A) and equal separation of chromosomes in anaphase I with no detectable chromosomal abnormalities.

At metaphase I, precocious separation of bivalents was observed in most cells followed by formation of laggards (Fig. 46 B and C) in a small number of PMCs (pollen

mother cells). Non-synchronization (Fig. 46 B) of genomes of two parents (sterile F1 hybrid and cultivated pigeonpea) at different meiotic stages was observed in 6.5% of the pollen mother cells (PMC) of sterile anthers. However remaining 94.5% of the PMCs showed regular bivalent formation at metaphase I (Fig.46 A). At anaphase I, the separation was normal (11/11) in 90% of the PMCs studied while 10% showed unequal separation due to 1 or 2 laggards (Fig. 46 C). Laggards seen at anaphase may be the univalent because their frequencies were comparable and there were 1-2 laggards

103

whenever observed. Regular 11-11 separation was seen in most of the cells at anaphase I. Normal tetrads were observed after telophase II (Fig. 46 D). Unseparated tetrads were found in the post tetrad stages in agreement with the mechanism of microsporogenesis in

sterile F1 hybrids (Fig. 46 E). Degeneration of microspores after tetrad stage resulted in empty shrunken sacs inside anther lobe observed in sterile anthers (Fig. 46 G). Anther‟s bundle stained with Alexander‟s stain, absorbed only green color (Fig. 46 F and H).

Crossability between F1 sterile hybrids and pigeonpea cultivars detailed pollen grain studies were carried out using acetocarmine and Alexander staining techniques.

Alexander stained anthers of partially fertile F1 hybrids (fertility 35 to 50%) exhibited presence of both pink color stained fertile and green color stained sterile pollen gains.

Whereas anthers from sterile F1s showed 100% green colored pollen grains inside. When

these 100% sterile F1 s were back crossed with different cultivated pigeonpea varieties, they resulted in a high degree of pollen sterility (19 to 100%) and partially fertile lines

developed from all the crosses (Table 14). When the sterile hybrids F1 P-4 and F1 P-7 were back crossed with different cultivars, only crosses involving cv. ICPL 85010 and MN1 resulted in sterile progeny lines along with partial fertile lines. However crosses with other cultivars like ICPL 88039, ICPL 88034, ICPL 92016 and MN5 produced only fertile progeny. In contrast, sterile lines resulted from all the back crosses involving

sterile F1 P-12 and pigeonpea cultivars except with ICPL 88034 in which only fertile progeny were obtained. ii. Anatomy of microsporogenesis in male sterile (P12 X MN1) and fertile plants (P12 X MN1)

The study was undertaken to study the process of microsporogenesis in the male sterile and male fertile lines derived from C. cajan X C. lanceolatus. For this study one

fertile and one sterile individual of F1BC1 generation were selected. These individuals

were generated by crossing the sterile F1P-12 (female parent) with cultivated pigeonpea MN1 (male parent). The progeny resulted from this cross (P-12 X MN 1) included both male sterile as well as male fertile plants. To confirm the sterility of the plant, anthers of sterile individuals were stained with Alexander‟s stain which showed sterile pollen stained completely green after 48hrs of incubation. These flower buds of sterile individual plants were fixed for microtome sectioning to study the anatomy of microsporogenesis of male sterile plants.

104

Longitudinal section (L.S) and Transverse section (T.S) of anthers of similar sizes from both male-fertile (Fig. 47 A and B) and sterile (Fig. 47 J and K) plants showed no differences in the development of sporogenous tissue. In the PMCs (pollen mother cells), anther wall consists of four distinct layers; an epidermis, an endothecial layer, a middle layer and a tapetal layer (Fig. 47 C, D, L and M). The first nuclear divisions of tapetum cells occurred before the meiosis. During the meiosis I, there was no change in epidermis and endothecium. Even at the tetrad stage T.S of anther lobe showed normal tetrads in both fertile and sterile hybrids (Fig. 47 C and L). But in the subsequent stage, pollen grains were normal in fertile plant whereas shrunken in sterile hybrid (Fig. 47 D and M).

In the anther from fertile hybrid, the tapetal layer degenerated, thus nourishing the microspores until the formation of mature pollen grains (Fig. 47 E to I). In the anther from sterile hybrid the process of microsporogenesis was similar up to the stage of differentiation of PMCs into tetrad formation (Fig. 47 L), but it differed thereafter (Fig. 47 M to R). There were no differences between male-sterile (Fig. 47 N and O) and male- fertile (Fig. 47 E and F) anthers for the number of PMCs. A persistent tapetum was observed in male sterile plants leading to the degeneration within their callose walls (Fig. 47 P to R). The sterile anthers shriveled and remained indehiscent (Fig. 47 Q and R). In the case of fertile anthers, the process of microsporogenesis proceeded normally with the development of pollen grains (Fig. 47 H).

Anatomical studies revealed that meiosis in both male fertile and male-sterile plants proceeded normally up to the tetrad stage, and during this period, the tapetum remained intact. The tetrads in male-sterile plants remained enclosed within the tetrad wall due to persistence of tetrad walls and, subsequently, it led to vacuolation and abortion of pollen grains without influencing female fertility of plants. On the contrary, microsporogenesis in male-fertile plants proceeded normally. Although the pollen grains in male-sterile anthers lost their contents, they remained as tetrads, and at the end of microsporogenesis, only empty anther lobes were present with some degenerated mass of sterile pollen.

DISCUSSION

With stagnancy in pigeonpea yield, it has become necessary to look for avenues to increase yield by the utilization of CMS system. Legumes in general were not good

105 candidates until now for the utilization of CMS systems due to cleistogamous nature of flowers that does not permit economical mass pollen. Pigeonpea is however is an exception with natural out crossing upto 70% (Saxena et al., 1990). In one of the CMS system (A7; Mallikarjuna et al., 2011) clasmogamous flowers were observed which is ideal for cross pollination. There are various CMS systems are now available for pigeonpea (Saxena et al., 2010; Mallikarjuna et al., 2012) and some of the systems are being actively utilized to increase pigeonpea yield. For long term viability of a hybrid breeding system, diversification of both genetic as well as cytoplasmic system is very essential. The present source of CMS being reported, which has to be developed further, offers good opportunities in this direction. The fertile plants obtained, are being advanced further to study alien introgression and presence of useful traits (Srikanth et al., 2013).

There are already 8 CMS sources reported (Saxena et al., 2010; Kumar and Saxena, 2013) and amongst which one has been reported on cultivated pigeonpea cytoplasm with the nuclear genome of C. acutifolius (A5 CMS; Mallikarjuna and Saxena,

2005), the present system will be named the A9 CMS system. This system is the second CMS system being reported on cultivated pigeonpea cytoplasm, with nuclear genome of C. lanceolatus.

The results have shown that one of the male sterile lines F1P12 produced only male sterile plants when crossed with cultivar ICPL 85030. The next best one with respect to maintaining male sterility was cultivar MN1. Thus showing that these are good maintainers of male sterility and the rest of the cultivars had different levels of fertility restoring abilities. Especially ICPL 88034 produced only fertile plants and this variety can be classified as a good restorer of this CMS line. In the CMS lines, it was observed that there was breakdown of the microsporogenesis due to persistent tapetum as observed in genetic male sterility reported by Reddy et al. (1978) and non-desolution of tetrad wall as observed by Mallikarjuna and Saxena (2005) in the A5 CMS system of pigeonpea. Mallikarjuna and Kalpana (2004) reported two types of microsporogenesis mechanisms in CMS lines. Type I CMS had partially or totally brown and shriveled anthers and the process of microsporogenesis was inhibited at the pre meiotic stages. Type II CMS had pale white shriveled anthers and the break down in microsporogenesis was at the post meiotic stage after the formation of tetrads caused sterility of plants. The second one appeared to be the case with CMS in the present study. The post meiotic breakdown of microsporogenesis was also reported in Allium cepa (Virnich, 1967). Synthesis of CMS

106

systems by Saxena et al. (2010) and Mallikarjuna et al. (2012) have shown that wild relatives of pigeonpea from both secondary and tertiary gene pool are good sources to develop CMS systems in pigeonpea if concerted efforts are directed to develop CMS systems.

Premeiotic

Post meiotic

Two types of meiotic inhibitions in CMS lines of pigeonpea.

4. Hybridity test RESULTS i. Hybridity test of F1 hybrids derived from C. lanceolatus using SSR markers

In order to identify SSR markers to assess the purity of the eight F1 hybrids (P1, P2, P4, P7, P8, P9, P10 and P11) developed by crossing C. cajan (ICPL 85010) x C. lanceolatus (ICP 15639), a total of 100 SSR primer pairs were screened. Out of these, 100 only 27 primers pairs could produce amplification product in the expected range of sizes (Table 16). Among these amplified primers, a total of 17 were found to be polymorphic among the two parental genotypes and produced clear, scorable and unambiguous polymorphic bands.

The parents and F1 hybrids were carefully observed for their morphological characters in order to identify true hybrids. Various morphological characters were recorded for each of these hybrid plants and are given in Table 4 (from morphology

107 results). Apart from this, 17 polymorphic SSR primers (CcM0008, CcM0035, CcM0047, CcM0057, CcM0710, CcM0974, CcM1991, CcM1459, CcM2012, CcM2071, CcM2176, CcM2228, CcM2505, CcM2639, CcM2672, CcM2707and CcM2855 were used for the identification of the hybrids. As SSR markers are codominant, only one allele was detected in a hybrid when the parents were monomorphic for a particular microsatellite locus and two alleles (one allele from each parent) were present in a hybrid when polymorphism was detected between the male parent and female parent. Out of 17 primers, 14 primers (CcM0008, CcM0047 (Fig. 48), CcM0057 (Fig. 49), CcM0710 (Fig. 50), CcM0974 (Fig. 51), CcM1991 (Fig. 53), CcM2012 (Fig. 54), CcM2071 (Fig. 56), CcM2176 (Fig. 57), CcM2228 (Fig. 58), CcM2505 (Fig. 59), CcM2639 (Fig. 60),CcM2707and CcM2855

(Fig. 61) were polymorphic among parents and F1 hybrids showing the heterozygotic condition of the hybrid/s (Table 17). The amplification products obtained from these 14 primers were described in each hybrid as follows (Fig. 48 to 61).

 In the hybrid F1P1, Primer CcM0008 yielded only male specifc amplicon of 202bp; CcM0047 yielded female and male specifc amplicons of 174bp and 185 bp; CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 284bp (ICP 15639); CcM0710 yielded amplicons of 287bp (ICPL 85010) and 297bp (ICP 15639); CcM0974 yielded amplicons of 175bp (ICPL 85010) and 171bp (ICP 15639); CcM1991 yielded amplicons of 206bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 215bp; CcM2176 yielded male specific amplicon of 271bp; CcM2228 yielded amplicons of 289bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 165bp (ICP 15639); CcM2707 yielded female specific amplicon of 238bp and CcM2855 yielded female specific amplicon of 282bp.

 Hybrid F1P2: Primer CcM0008 yielded only amplicons of 194bp (ICPL 85010) and 202bp (ICP 15639); CcM0047 yielded female and male specifc amplicons of 174bp (ICPL 85010) and 185bp (ICP 15639); CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 283bp (ICP 15639); CcM0710 yielded amplicons of 287bp (ICPL 85010) and 297bp (ICP 15639); CcM0974 yielded amplicons of 175bp (ICPL 85010) and 172bp (ICP 15639); CcM1991 yielded amplicons of

108

206bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 217bp (ICP 15639); CcM2176 yielded male specific amplicon of 271bp (ICP 15639); CcM2228 yielded amplicons of 288bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 216bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded male specific amplicon of 242bp (ICP 15639) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P4: Primer CcM0008 yielded only male specifc amplicon of 202bp (ICP 15639); CcM0047 yielded female and male specifc amplicons of 174bp (ICPL 85010) and 185 bp (ICP 15639); CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 283bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 287bp; CcM0974 yielded amplicons of 179bp (ICPL 85010) and 171bp; CcM1991 yielded amplicons of 205bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 215bp (ICP 15639); CcM2176 yielded female specific amplicon of 272bp; CcM2228 yielded amplicons of 289bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded female specific amplicon of 238bp (ICPL 85010) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P7: Primer CcM0008 yielded only male specific amplicon of 202bp (ICP 15639); CcM0047 yielded female and male specifc amplicons of 174bp (ICPL 85010) and 185bp (ICP 15639); CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 283bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 287bp (ICP 15639); CcM0974 yielded amplicons of 179bp (ICPL 85010) and 171bp (ICP 15639); CcM1991 yielded amplicons of 206bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 215bp (ICP 15639); CcM2176 yielded male specific amplicon of 271bp (ICP 15639); CcM2228 yielded amplicons of 288bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP

109

15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded female specific amplicon of 238bp (ICPL 85010) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P8: Primer CcM0008 yielded female and male specifc amplicons of 194.64bp (ICPL 85010) and 202.41bp (ICP 15639); CcM0047 yielded amplicons of 175bp (ICPL 85010) and 185.60bp (ICP 15639); CcM0057 yielded the amplicons of 291bp (ICPL 85010) and 284bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 287.bp (ICP 15639); CcM0974 yielded amplicons of 171bp (ICPL 85010) and 175bp (ICP 15639); CcM1991 yielded amplicons of 208bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 217bp (ICP 15639); CcM2176 yielded female specific amplicon of 272bp (ICPL 85010); CcM2228 yielded amplicons of 289bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded male specific amplicon of 242bp (ICPL 85010) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P9: Primer CcM0008 yielded female specific amplicon of 194.bp (ICPL 85010); CcM0047 yielded amplicons of 175bp (ICPL 85010) and 185bp (ICP 15639); CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 284bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 287bp (ICP 15639); CcM0974 yielded amplicons of 175bp (ICPL 85010) and 168bp (ICP 15639); CcM1991 yielded amplicons of 208bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 236bp (ICP 15639); CcM2071 yielded only male specific amplicon of 217bp (ICP 15639); CcM2176 yielded female specific amplicon of 272bp (ICPL 85010); CcM2228 yielded amplicons of 289bp (ICPL 85010) and 312bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded male specific amplicon of 242bp (ICPL 85010) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P10: Primer CcM0008 yielded only male specific amplicon of 202bp (ICP 15639); CcM0047 yielded amplicons of 174bp (ICPL 85010) and 185bp (ICP

110

15639); CcM0057 yielded the amplicons of 291bp (ICPL 85010) and 289bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 286bp (ICP 15639); CcM0974 yielded amplicons of 179bp (ICPL 85010) and 168bp (ICP 15639); CcM1991 yielded amplicons of 208bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 237bp (ICP 15639); CcM2071 yielded only male specific amplicon of 215bp (ICP 15639); CcM2176 yielded female specific amplicon of 272bp (ICP 15639); CcM2228 yielded amplicons of 289bp (ICPL 85010) and 312bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded female specific amplicon of 237bp (ICPL 85010) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

 Hybrid F1P11: Primer CcM0008 yielded both male and female specific amplicons of 194bp (ICPL 85010) and 202bp (ICP 15639); CcM0047 yielded amplicons of 175bp (ICPL 85010)and 185bp (ICP 15639); CcM0057 yielded the amplicons of 290bp (ICPL 85010) and 284bp (ICP 15639); CcM0710 yielded amplicons of 297bp (ICPL 85010) and 286bp (ICP 15639); CcM0974 yielded amplicons of 175bp (ICPL 85010) and 167bp (ICP 15639); CcM1991 yielded amplicons of 208bp (ICPL 85010) and 194bp (ICP 15639); CcM2012 yielded amplicons of 240bp (ICPL 85010) and 237bp (ICP 15639); CcM2071 yielded only male specific amplicon of 217bp (ICP 15639); CcM2176 yielded female specific amplicon of 272bp (ICPL 85010); CcM2228 yielded amplicons of 288bp (ICPL 85010) and 311bp (ICP 15639); CcM2505 yielded amplicons of 217bp (ICPL 85010) and 228bp (ICP 15639); CcM2639 yielded amplicons of 145bp (ICPL 85010) and 166bp (ICP 15639); CcM2707 yielded male specific amplicon of 241bp (ICP 15639) and CcM2855 yielded male specific amplicon of 270bp (ICP 15639).

Thus, presence of both female and male parent alleles was observed as a resultant of crossing between two parents (F1 hybrid). It confirmed the crossing and hybridity of the progeny. However, eight F1 hybrids exhibited both the alleles of the parents confirming the heterozygosity condition of the hybrids. The identified SSRs in F1 hybrids showed complementary banding pattern of both parents. It was useful to distinguish the

F1s from their male and female parents. The remaining three primers CcM0035, CcM1459 and CcM2672 were monomorphic among parents and hybrids (Fig. 52). The

111

primer CcM0035 had an amplicon of 275bp in parents and hybrids. In the same way primers CcM1459 and CcM2672 had produced amplicons of 183bp and 243bp respectively among parents and hybrids. From the results, it was clear that each primer was amplified a single loci of expected size in both the parents. Hence these monomorphic primers yielded single amplicon (same as in parents) instead of two in all the hybrids. Moreover, 10 primers (CcM0252, CcM0416, CcM0988, CcM1385, CcM1447, CcM1999, CcM2066, CcM2095, CcM2379 and CcM2451) were amplified only in the female parent but not in the pollen parent. But three of these primers CcM0252, CcM0416 and CcM0988 were polymorphic among hybrids. Since they did not amplify in the pollen parent, these primers were not considered for genotype scoring. Whereas, remaining seven primers CcM1385 (272bp), CcM1447 (281bp), CcM1999 (173bp), CcM2066 (119bp), CcM2095 (246bp), CcM2379 (165bp) and CcM2451 (182bp) were

yielded only female specific alleles among F1 hybrids (Fig. 55). Hence these were also not considered for genotype scoring for hybridity test. Out of (above mentioned) 14 polymorphic primers, seven primers (CcM0008, CcM0035, CcM1459, CcM2176, CcM2672, CcM2707 and CcM2855) amplified in both parents, whereas in hybrids either female or male specific alleles amplified and percentage of male parental genome ranged from (50 to 87.5%). Remaining ten primers (CcM0047, CcM0057, CcM0710, CcM0974, CcM1991, CcM2012, CcM2071, CcM2228,

CcM2505 and CcM2639) yielded both the alleles (male and female alleles) in all the F1 hybrids. Hence these ten primers showed 100% hybrid purity index. Based on genotype scoring data obtained from 14 primers, each hybrid was evaluated for the percentage of male parental genome. In each hybrid, total number of „H‟ alleles and „B‟ alleles were multiplied by total number of alleles. As per the results, it was clear that all the hybrids had 85.71 to 100% (percentage of male parental genome) H and B alleles from the male parent and were confirmed as pure hybrids. The present study showed that SSR markers are quick, effective and results are generally consistent with

morphological analysis in the F1 hybrids. ii. Hybrid purity assessment in C. volubilis F1 hybrids

The F1 hybrid derived from the cross C. cajan (ICPL 85010) X C. volubilis

(ICP15774) was screened with 149 SSR primer pairs to test the hybridity of the F1 hybrid showed morphological traits of the female parent C. cajan (ICPL 85010). Out of 100

112 primer pairs, only 21 primer pairs amplified in parents and F1 hybrid. Almost all the primers yielded only monomorphic alleles among parents and hybrid (Table 19). The marker CcM0008 had resulted in amplifying allele of size 194.88bp in female parent

(ICPL 85010) and 202.52bp in male parent (ICP15774), whereas in the F1 hybrid, the amplicon size was 202.53bp. As the allele size in the F1 hybrid was same as the allele size in the pollen parent, the F1 hybrid was considered as the true hybrid (Fig. 62). Also 21

DNA samples the F2 progeny were tested with 100 primers, of which 26 primers were polymorphic in the parents. But only six primers had produced polymorphism among parents and F2 progeny. Even in the F2 progeny, the primer CcM0008 yielded an allele product 194.82bp in the female parent and 202.53bp in the pollen parent, whereas in 13 samples of F2 progeny (F2 plant numbers-50, 50B, 51, 52, 52A, 57A, 60, 61, 65, 66, 73, 74 and 74A) both the alleles were amplified. This confirmed the heterozygosity condition of the hybrids. Thus, presence of both female and male parent alleles was observed as a resultant of true crossing between two parents. This confirmed the hybridity of the progeny and polymorphism between the two parents. While in other F2 samples (51B, 54, 54A, 55, 56, 57A, 61A and 64) this primer amplified only male specific allele (Fig. 63). DISCUSSION

The use of DNA markers particularly Simple Sequence Repeats (SSRs) are useful for a variety of molecular breeding applications because of their co-dominance, abundance, and high genome coverage and multi allelic nature. SSRs are the most suitable markers for hybrid purity assessment as the heterozygosity of the hybrids can be easily determined by the presence of alleles from both the parents. They are already a proven tool for hybrid authentication or hybrid purity assessment and parentage confirmation in many crop species (Bohra et al., 2011). Indeed, molecular markers-based hybrid purity tests have been developed and are in routine use in many crop species such as rice (Yashitola et al., 2002; Sundaram et al., 2008), maize (Asif et al., 2006), cotton (Selvakumar et al., 2010) and safflower (Naresh et al., 2009). Similar to food crops, in forest trees also inter-specific hybrids are generated routinely through hybridization and the heterotic individuals are selected for clonal propagation and mass multiplication (Nikles and Griffin, 1992; Stanton et al., 2010). The use of microsatellite markers for the identification of true hybrids in the early stages of selection process has shown to be a very useful tool. In species with long breeding cycles, such as Cajanus sps., the conventional process of hybrid identification

113

by morphological traits is slow, especially when looking for pod and seed traits. Molecular Tools may overcome these difficulties and open the way to new strategies for more efficient breeding. The present study utilized the SSR marker techniques for

identification of eight F1 hybrids along with their parental lines, demonstrating that this technique can be successfully applied to distinguish and identify the hybrids from their parental lines. Therefore, the high polymorphic information content (PIC) of SSR had promoted the application of microsatellites as molecular markers in fingerprinting (Ashikawa et al., 1999). The DNA markers such as SSRs could be of excellent choice for hybridity confirmation in pigeonpea, which has good resources of SSRs (Bhora et al.,

2012). Generally, F1 hybrids contain DNA from both the parents and SSR markers identified both male and female parent specific markers allowing differentiation of true hybrids from selfed individuals and out crossed individuals with foreign pollen. SSR

marker information developed in the present study revealed that all the F1s derived from the wild relative C. lanceolatus from the secondary gene pool and C. volubilis, from the tertiary gene pool are indeed true hybrids.

5. Resistance of interspecific derivatives RESULTS i. Biochemical basis of resistance in interspecific derivatives derived from four wide crosses

Among all the interspecific derivatives derived from four different wild species, C. platycarpus x C. cajan derivatives showed high levels of TI units (22.12 to 113.84 TIU/ mg protein) compared with other interspecific derivatives derived from C. cajan x C. cajanifolius, C. cajan x C. lanceolatus and C. cajan x C. volubilis which ranged from 22.5 to 81.7 TIU/mg protein. Interspecific derivatives of the cross C. cajan x C. cajanifolius showed TI activity levels which ranged from 25.7 to 77.5 TIU (Fig. 64A) and 0.79 to 1.19 CIU (Fig. 65A). Few derivatives of this cross, 76-2 (32.5 TIU); 82-1 (44.37 TIU); 113-2 (41. 55 TIU); 129-1 (30. 76 TIU); 130-1 (37.3 TIU/mg proteins); 131-1 (35.13 TIU) and 3-1-1(34.5 TIU) showed more or less TI activity levels compared with cultivar parent ICPL87119 which showed 32.96 TIU (Fig. 64 A),whereas, the derivatives 18-5 (26.1 TIU), 37-1(27.2 TIU) and 132-1 (28.5 TIU/mg proteins) showed TI activity levels closer to the wild Cajanus parent (ICPW 29 showed 25.7TIU/mg proteins).

Moreover two derivatives such as (ICPL87119 x ICPW 29) F1 showed 77.5 TIU, 1.1 CIU

114 and derivative 131-1 showed 69.7 TIU and 2.1 CIU which exhibited two fold increased TI and CI activities compared with cultivated parent C. cajan (ICPL87119) which showed 0.76 CIU proteins (Fig. 64 A). CI activity levels of these derivatives were more or less similar to the wild parent (C. cajanifolius) which exhibited 1CIU/mg protein.

Trypsin inhibitory activity of F1 hybrids (ICPL85010X ICP15639) ranged from 22.5 to 81.76 TIU (Fig. 64B) and chymotrypsin inhibitory activity varied between 0.64 to 1.6 CIU (Fig. 65 B). The TI activity level of the cultivar ICPL85010 which showed 22.6 TIU was more than the wild parent C. lanceolatus (ICP 15639) which showed 12.8 TIU. But all the derivatives of this wide cross exhibited 1 to 2 fold increased TI activity than in cultivated parent (ICPL85010). In two derivatives P2 and P7, the TI activities were 60.9 TIU and 81.7 TIU which were 2.6 to 3.6 folds more than the activity levels of cultivated parent (Fig. 64 B). The CI activity levels of the cultivated parent (ICPL85010) which showed 0.57 CIU and wild Cajanus parent (C. lanceolatus) which exhibited 0.6 CIU were similar. Also, two interspecific derivatives P13 and P14 exhibited similar activity levels compared to both the parents. But in other derivatives exhibited more or less 2 fold increased CI activity units than the parents (Fig. 65 B).

Interspecific derivatives derived from the cross ICPW68 X ICPL 85010 showed highest trypsin inhibitory activity ranged between 33.2 to 113 TI units/mgprotein (Fig. 64 C) against bovine pancreatic trypsin and chymotrypsin inhibitory activity units between 0.7 to 3.04 CIU/ mg protein (Fig. 65 C). Both the parents, (ICPL 85010 which showed 22.6 TIU and ICPW 68 which showed 22.7 TIU) of this cross and one interspecific derivative B3-2 (22.1 TIU) showed similar TI activity levels, whereas the remaining interspecific derivatives exhibited 1.5 to 3.3 fold more inhibitory action against bovine pancreatic trypsin compared to the parents. The C7-13 of this cross showed 113.8 TIU which was the highest TI activity among all the parents and derivatives tested (Fig. 65 C).

F2 hybrid derived from the cross ICPL 85010 X ICP 15774 showed 63 TIU (Fig. 64 D) which was 3 fold increase than the wild parent C. volubilis (ICP 15774) which showed

29.4 TIU. The CI activity of the F2 hybrid was 1.51 CIU which was similar to the wild Cajanus (C. volubilis) which showed 1.6 CIU (Fig. 65 D).

Wild relatives exhibited 14.5 to 76 fold more inhibition of H. armigera mid gut trypsin-like (HGPI) activity compared with cultivated pigeonpea (ICPL 85010) which exhibited 3 units/ mg (Fig. 66). Tertiary gene pool species C. platycarpus showed 228

115 units/mg protein which was highest inhibition of HaGPI than the other wild species such as C. volubilis showed 66.63 units /mg protein (Fig. 66 D), C. cajanifolius showed 27.2 units/mg protein (Fig. 66 A) and C. lanceolatus showed 74.9units/mg protein (Fig. 66 B). With respect to the cultivated pigeonpea, 3 to 35.3 fold more inhibition was observed in the interspecific derivatives derived from four wild relatives of pigeonpea (C. cajanifolius, C. lanceolatus, C. platycarpus and C. volubilis). Among all the interspecific derivatives and cultivars, two derivatives BC3 F3 C7-13 (derived from C. platycarpus) showed 106.2 units/ mg protein and F2 P8 (derived from C. lanceolatus) showed 77.5 units/mg protein which were highest inhibition units of HGPI (Fig. 66B and C). Two derivatives, F1BC1 P1 and F2 of the cross ICPL 85010 X ICP 15639, showed 100% introgression of resistance genes from the wild parent as they exhibited equal inhibitory activity with C. lanceolatus which showed 74.9 units/ mg protein inhibition units against HGPI (Fig. 66 B).

Interspecific derivatives which expressed greater TI, CI and mid gut trypsin like inhibitory activities were further confirmed through activity staining studies. Interspecific derivatives and wild Cajanus species had greater variation in TI, CI and mid gut trypsin like inhibitory activities which was evident through activity profiles (Fig. 67). In gelatin- PAGE under non-denaturing condition, the cultivar ICPL 85010 and C. cajanifolius (ICPW29) exhibited homomorphism in terms of TI and CI isoforms when the gels were incubated with bovine pancreatic trypsin or chymotrypsin, respectively (Fig. 67 A and B). But other wild Cajanus species and interspecific derivatives exhibited variation in number of TI, CI and even against insect midgut trypsin-like proteinases isoforms (Fig. 67 A, B and C) and their banding pattern. All the genotypes (parents and their derivatives) showed two distinct TI, 2 to 5 CI and 3 to 5 HGPI bands on gelatin-PAGE.

However, the interspecific derivative F3 BC3 C7-13 which is derived from C. platycarpus showed TI, CI and HGPI bands with high intensity when compared with other genotypes. Thus, activity staining studies also revealed that PIs from Cajanus species parents and their derivatives had more potential to inhibit the activity of insect midgut trypsin-like proteinases.

Increasing levels of insect mid gut trypsin-like enzyme inhibitors and bovine pancreatic trypsin inhibitors were assayed, in order to determine whether substantial inhibitory effect on human pancreatic trypsin (HPT) may therefore be suspected. This work addressed the above said complex situation and to enable quantitative studies to be

116 carried out, human pancreatic trypsin employed as a model for the digestive juice in the small intestine. Crude protein extracts of cultivars and wild Cajanus species parents along with their interspecific derivatives (cv.ICPL 85010), C. cajanifolius, C. lanceolatus, C. platycarpus and C. volubilis, (ICPL 87119 X ICPW 29)133-1 BC1 F2, (ICPL 87119 X

ICPW29) F1, ICPL (85010 X ICP 15639) P-1 F1 BC1, F2P-8, (ICPW 68 X ICPL 85010)

C7-13 F3BC3, F1BC5B-6, (ICPL 85010 X ICP 15774) HYB -3 F2 which expressed high levels of HGPIs were screened for the human pancreatic trypsin inhibitor levels. Pigeonpea PIs (PPIs) have less inhibitory effect on HPT than the mid gut trypsin like enzyme. HPT was weakly inhibited by crude extracts of pigeonpea PIs. HGPI inhibitor units which ranged from 43.7 to 228 U/mg protein and were more than the human pancreatic trypsin inhibitor (HPTI) units ranged between 11.1 to 53.13U/mg protein in wild and interspecific derivatives. The ratio of HGPI to HPTI units indicated that there is less or no effect on the human pancreatic trypsin but more effect on the insect mid gut trypsin like proteinases (Fig. 68). Human pancreatic trypsin requires more amount of proteinase inhibitors (which expressed in units) for its inhibition i.e. more than the HGPIs. The PIs expressed in the interspecific derivatives can be ruined upon adequate heating. However further studies are necessary to characterize the PIs from wild relatives of pigeonpea to develop strategies for expression of PIs from the wild relatives in the cultivated pigeonpea for resistance to H. armigera.

DISCUSSION

Although it is a known fact that wild relatives of pigeonpea possess considerable insect resistance, the biochemical mechanism involved in the resistance has not been investigated. Interspecific derivatives derived from the wild relatives (C. cajanifolius, C. lanceolatus, C. platycarpus and C. volubilis) of pigeonpea along with their Cajanus species and cultivated parents (C. cajan- ICPL 85010 and ICPL87119) were screened for proteinase inhibitors (PIs). Success has been achieved in introgressing the resistance to H. armigera from wild Cajanus species (Jadhav et al., 2012) and the present work. The PIs were identified in interspecific derivatives and wild relatives of pigeonpea showing higher inhibitory potential against Bovine pancreatic trypsin, chymotrypsin and insect (H. armigera) proteinases. Crude extracts prepared from the interspecific derivatives and parents (cultivated and wild) were examined for inhibitor activity against bovine pancreatic trypsin and chymotrypsin.

117

These results indicated that the hybrids and wild Cajanus parents have shown highest trypsin, chymotrypsin inhibitor activity and H. armigera mid gut trypsin-like proteinase inhibitor (HGPI) activity than the susceptible cultivated parents. In agreement with the previous reports (Chogule et al., 2003, Prasad et al., 2009, Parde et al., 2012) wild parents exhibited strong inhibitory potential against H. armigera mid gut trypsin-like proteinases. Similar observation has been reported in chickpea, where high variation in PIs was recorded in the mature seeds of wild relatives than in the cultivated ones (Patankar et al., 1999). Swathi et al. (2011) identified the presence of trypsin inhibitors conferring resistance to H. armigera in C. platycarpus.

Significant differences between cultivated pigeonpea and its wild relative were evident for seed protein and trypsin inhibitors. Trypsin inhibitors were higher in wild Cajanus species in all pod stages compared with those in cultivated types. No significant chymotrypsin inhibition was evident in enzymatic assays among all the samples of parents and interspecific derivatives.

As the trypsin inhibitors are known to cause reduced digestion and mainly are anti-nutritional (Rackis, 1981) and insecticidal (Johnston et al., 1991), their presence in the pod from juvenile to mature stages could be an important component of the biochemical basis of resistance to pod borer. The levels of these anti-nutritional factors have been determined in pigeonpea and chickpea whole seed (Singh and Jambunathan, 1981). However, examination of these inhibitors in soybean, the most thoroughly studied of all legume species, showed that they are anti-nutritional and that their residual activities, even in processed human foods, is concern to human health (Broadway and Duffey,1986; Liener, 1986), but in pigeonpea, they can be reduced by cooking, germination or fermentation (Singh and Eggum,1984). Interestingly, there are no differences in the electrophoretic forms of these protease inhibitors in the wild and cultivated species. But still these crude proteins were tested against human pancreatic trypsin and no significant inhibition found. Moreover the inhibition levels are more against Helicoverpa mid gut trypsin like enzymes. This was evident through the enzymatic assays with human pancreatic trypsin.

The development of pod borer resistant lines has opened up new perspective in pigeonpea improvement program. Development of pod borer resistant lines will have a

118

major impact on the pigeonpea producers as they need not depend heavily on synthetic chemicals to control this insect and thus, saving farmer‟s resources and protecting the environment. The present study clearly demonstrated that it is possible to introgress pod borer resistance trait from Cajanus species into cultivated pigeonpea. The present study also indicated that it is advantageous to select for low damage in each evaluation to have high levels of resistance to H. armigera. ii. Screening for Podborer resistance RESULTS

Cultivated parents, C. cajanifolius derivatives (F2 and BC1F2 populations),

C.lanceolatus derivatives (F1BC1 and F2 populations) and C. platycarpus derivatives

(F1BC5 and BC3F3 population) were screened in field for podborer resistance (Fig. 69). Total pods per plant were collected from each plant and total number of pods/ plant recorded. Mean configuration of number of healthy as well as damaged pods and percentage of pod damage were tabulated in the Tables 21, 22 and 23.

Helicoverpa armigera larvae fed on the crop in the initial stages of flowering as well as podding. The interspecific derivatives from C. cajanifolius (Table 21), C. lanceolatus (Table 22) and C. platycarpus (Table 23) were evaluated for podborer damage. Among the three derivatives, C. cajanifolius derivatives had lowest percentage of pod damage (1.8 to 14.8%) compared with the other lines. In the C. cajanifolius derivatives, lines 129-1 and 130-1 showed more number of healthy pods with 1.8-12.6%

damage. F2 lines had 6.8-14.8% damage. Among all the C. cajanifolius derivatives highest number of healthy pods found in some of the individual plants. Such as ICPL

87119 X ICPW 29 F2 -2 had 360 pods (6.6% of damage), ICPL 87119 X ICPW 29 F2 -7 had 536 pods (3.24% of damage), 18-5-3 had 530 pods (9.71% of damage), 129-1-1 had 360 pods (3.7% of damage), 130-1-1 had 245 pods(7.55% of damage), 131-1-2 had 245 pods(6.8% of damage, 131-1-7 had 420 pods (6.4% of damage), 132-1-7 had 381 pods (2.8% of damage), and 133-1-5 had 285 healthy pods (4.3% of damage) (Fig. 71).

C. lanceolatus derivatives had lower pod set as they were partially fertile and

early in the (F2) generation. Though, they had lower pod set, percentage of the damage (3.03 to 17.4%) was lower than the cultivar parents. In some of the individual plants such

as BC1F2P1-1 had 51 healthy pods (3.7% of damage) P2-1 had 96 pods (3% of damage),

119

P8-3 had 77 pods (6% of damage), P4-3 had 49 pods (3.9% of damage), P6-3 had 52 pods (5.45% of damage) and P10-3 had 42 healthy pods (6.6% of damage) (Fig. 72).

Percentage of the damage ranged from 2.5 to 22.2% in the C. platycarpus derivatives. In the cultivated varieties percentage of the damage was 47.45% in ICPL

85010 and 44.51% in ICPL87119. Some of the individuals such as BC3F3C7-13-1 had 137 healthy pods (4.3% of damage), C7-13-2 had highest 900 pods (2.9% of damage) (Fig. 70), C11-3-1 had 500 pods (3.6% of damage) and C11-3-2 had 150 healthy pods (2.5% of damage) (Fig. 73). These individuals had more number of healthy pods as compared with C. cajanifolius derivatives.

DISCUSSION

Screening of more than 15,000 accessions of pigeonpea germplasm for resistance to H. armigera has revealed very low levels of resistance to this pest (Sharma, 2005). Development of crop cultivars resistant to this pest has a greater potential for integrated pest management, particularly under subsistence farming conditions in the developing countries (Fitt, 1989). Some of the wild relatives of pigeonpea have shown high levels and biochemical components of resistance to H. armigera (Sharma et al., 2009).

In the present study, it was evident that wild species had self defense mechanisms and successful introgression of these defense genes in to cultivated pigeonpea was possible through wide hybridization. From the Tables 21, 22 and 23, it is clear that the interspecific derivatives exhibited more resistance towards pod borer damage. Cajanus cajanifolius derivatives exhibited low percent damage. The lines C7-13 and C11-3 derived from C. platycarpus showed very less damage with more number of healthy pods. Similar observations were reported in C. acutifolius derivatives where interspecific derivatives exhibited resistance to the H. armigera damage (Jadhav et al., 2012b).Other species such as C.scarabaeoides, C. acutifolious, C. sericeus and C. albicans are some of the wild Cajanus species which showed resistance to pigeonpea pod borer H. armigera (Sujana et al., 2008). Cajanus scarabaeoides, a wild species of Indian origin, has multiple disease resistance (Kulkarni et al., 2003; Upadhyaya, 2006). Pods of C. scarabaeoides have a dense covering of non-glandular and low density of glandular trichomes (Shanower et al., 1997). Since C. scarabaeoides had least damage compared to cultivated

120

pigeonpea, it was concluded that non-glandular trichomes form a preventive layer for insect lodging and feeding on the pod surface.

Pod borer is a major biotic constraint of pigeonpea with low levels of resistance in cultivated germplasm, which completely succumb to the pest under high insect pressure. The development of pod borer resistant lines has opened up new vistas in pigeonpea improvement program. Development of pod borer resistant lines will have a major impact on the pigeonpea producers as they need not depend heavily on synthetic chemicals to control the insect and thus, saving farmer‟s resources and protecting the environment. iii. Screening for Bruchid resistance RESULTS

 Callosobruchus maculatus resistance in the interspecific derivatives derived from C. lanceolatus

In this experiment, C. maculates laid eggs on all the seeds of parent plants and F1 hybrids tested. Invariably, the egg load was high on the susceptible cultivated parent ICPL 85010 and the difference between the resistant and the susceptible lines was easily observed. Oviposition by female bruchids was moderate on C. lanceolatus and its derivatives, whereas the susceptible line ICPL 85010 showed 100 % oviposition. Among

the nine samples, more number of eggs laid and adults emerged from the seeds of F1 P2. In all the hybrids, 3 to 12 adults emerged except in P8 from which only one adult emerged out of 19 eggs laid. Percentage of the eggs hatched ranged from 5.2 to 77% (Table 24). However, in the cultivated parent, ten adults emerged out of 14 eggs. The adults that developed from these wild species were small in size compared to those emerged from the cultivated species, again suggesting the role of physical and chemical factors.

It was observed that 71.4% of the eggs hatched. The larval/pupal period was prolonged in C.lanceolatus (57 to 81days) compared to 35day on the female parent ICPL 85010 (Fig. 74 A and B).The high rate of unhatched eggs in the wild and interspecific derivatives revealed that seed damage was low to moderate level. A similar observation was also noticed in C. lanceolatus (delayed adult emergence) (Fig. 74 C and D). Among all the hybrids, the number of days taken for adult emergence ranged from 41 to 56days.

Adult mortality was 100% in the F1 P8, whereas in the cultivated parent ten healthy adults

121

were found. Wild parent exhibited 40% adult mortality, whereas remaining F1s exhibited 16.7 to 62.5% adult mortality.

DISCUSSION

The present investigation showed that wild relative of pigeonpea used in this experiment is good sources of resistance to bruchids as percent damage was low on this seed sample, with low adult emergence in most lines, with more than 50days for adult emergence and high percent of unhatched eggs. Wild relatives of pigeonpea both from secondary (C. scarabaeoides and C. acutifolius) as well as tertiary genepool (C. platycarpus) are good sources of resistance to bruchid (Jadhav et al., 2012a). The ICRISAT has utilized many of these resources to introgress useful traits such as resistance to pod borers and have succeeded in this endeavor (Mallikarjuna et al., 2011 a and b). Hence, lines derived from the C. lanceolatus can be used to introgress multiple disease resistance.

6. Validation of DT expression in C. volubilis derivatives RESULTS i. Description of the terminal flower Phenotype

In the process of wide hybridization between C. cajan (ICPL 85010) and C.

volubilis (ICP15774), interesting F2‟s were obtained with determinate flowering pattern (Fig. 76). Out of these two parents, cv. C. cajan was a semi-determinate and wild C. volubilis was a indeterminate plant. Days to flowering in the F2 progeny was measured as the number of days between germination and the opening of the first flower. Plant height was recorded after the plant had reached its maximum height and fruits began to shatter.

As the F2 individuals were segregated for determinate (DT) and semi determinate (SDT) growth habits (flowering pattern), the molecular basis of the determinacy trait was validated by using the DT trait specific primers based on the determinacy candidate gene in pigeonpea “TFL1”. ii. Amplification of TFL1 Sequences in F2 population derived from C. volubilis

Two parental and 21 F2„s genomic DNA were tested with the TFL1 primers. One DT cultivar MN1 also tested along with these test samples as a control. Two bands (one common band and one allele specific band) appeared in all the samples including the DT

122 cultivar after gel electrophoresis of the amplified product. The degenerated common primers (TFL1_PCR_CF and TFL1_PCR_CR) amplified to 848 bp specific DNA product among all genotypes. The indeterminate and semi-determinate genotypes exhibited „A‟ allele specific 734bp, whereas determinate genotypes exhibited 167bp „T‟ allele specific fragment. Parents ICPL 85010 and ICP15774 showed two similar bands due to the presence of dominant allele of determinacy gene, the absence of which (dominant alleles) results in the formation of recessive phenotype (determinate) (Fig. 75). Thus it is clear that a 734bp band amplified by CcTFL1 gene is indeterminate type specific band in IDT cultivars, whereas the DT cultivar MN1 also exhibited two t bands, one was a common band (848bp) and other was a DT specific band (167bp) amplified only in DT cultivars. This PCR based marker clearly distinguished IDT lines from the DT lines based on amplification of specific fragments in IDT lines (734bp fragment amplified) and DT lines (167bp fragment amplified) in addition to amplification of common fragment in both DT & IDT lines (848bp fragment). Among F2 progeny, samples 50, 52, 52A, 54, 60, 61A, 64, 64A, 65, 65A, 66, 73, 74 and 74A showed DT specific band and common bands. While in other samples 50B, 51, 51B, 54A, 55, 57A and 61 showed SDT specific bands.

Phenological observations of F2s were reconfirmed with growth habit specific molecular markers.

DISCUSSION The study of marker trait associations for determinacy trait in pigeonpea strongly indicates the involvement of several genomic regions/genes/markers responsible for this important trait in pigeonpea. The identification of four DT/IDT specific PCR based markers explaining phenotypic variation for growth habit may prove useful in molecular breeding programmes for pigeonpea improvement.

The transition from vegetative to reproductive phases at the shoot apical meristem (SAM) is controlled by the interaction of positive and negative regulators, such as LEAFY (LFY), APETALA1 (AP1) and TERMINAL FLOWER 1 (TFL1) (Benlloch et al., 2007). LFY and AP1 are the main promoters of floral meristem identity, and encode transcription factors; both are expressed throughout the young floral meristem from the earliest stages of development, and after the onset of LFY expression, AP1 is expressed in these meristems (Mandel et al., 1992; Weigel et al., 1992; Maizel et al., 2005). The role played by TFL1 in floral initiation is the opposite of that of LFY and AP1. The Arabidopsis tfl1

123 loss-of-function mutants flower early, and the SAM is converted into a terminal flower (Shannon and Meeks-Wagner, 1991; Schultz and Haughn, 1993). The TFL1 gene in Arabidopsis maintains indeterminate growth of the SAM by delaying the upregulation of floral meristem identity genes (Ratcliffe et al., 1998).

TFL1 belongs to a small gene family, one of whose members, FLOWERING LOCUS T (FT), is also a regulator of flowering time. However, FT acts in an opposite manner to TFL1; FT promotes flowering and conversion of the SAM to a flower (Kardailsky et al., 1999; Kobayashi et al., 1999). CENTRORADIALIS (CEN) is an Antirrhinum TFL1 orthologue (Bradley et al., 1996). As with tfl1 mutants, recessive mutations in the CEN gene result in the conversion of the normally indeterminate inflorescence to a determinate condition. However, the time to flowering is not affected in the cen mutants in contrast to the tfl1 mutants (Bradley et al., 1996, 1997). Both TFL1 and CEN are expressed in the subapical region of the shoot meristem; TFL1 is expressed both in vegetative and inflorescence shoot meristems, whereas CEN is only expressed in the inflorescence meristem (Bradley et al., 1996, 1997).

TFL1 gene interactions in Arabidopsis thaliana (http://stringdb.org/newstring_cgi/show_network_section)

124

The homologues of Arabidopsis TFL1/Antirrhium CEN have been isolated and their functions have been characterized in many plant species (reviewed by Benlloch et al., 2007). Pea (Pisum sativum) contains at least three TFL1/CEN homologues (Foucher et al., 2003). No function has been assigned to PsTFL1b, a likely orthologue of CEN, while PsTFL1a and PsTFL1c are most closely related to Arabidopsis TFL1 and probably have functions identical to those conferred by Arabidopsis TFL1 when the functions of these genes are combined. Mutations in PsTFL1a, also known as DETERMINATE (DET), cause the determination of the main apex without affecting flowering time, in a manner similar to that in cen mutants of Antirrhinum. On the other hand, mutations in PsTFL1c, also known as LATE FLOWERING (LF), cause early flowering without affecting determination. Similar to CEN, DET is expressed in the shoot apex only after the floral transition, while LF expression is also observed in the vegetative apex. Therefore, in pea, the two functions of the Arabidopsis TFL1 gene, flowering time and apex determinancy, seem to be controlled by two different genes (Benlloch et al., 2007).

The stem growth habit influences various agronomical traits. Determinate plant lines, for example, generally reach much shorter heights with increased lodging resistance and have lower lowest-pod heights and more main stem branches per plant than do indeterminate cultivars of similar maturities (Bernard, 1972; Foley et al., 1986; Ablett et al., 1989; Ouattara and Weaver, 1994; Robinson and Wilcox, 1998; Kilgore-Norquest and Sneller, 2000). The determinate lines also have shorter flowering and reproductive periods than do the indeterminate lines of similar maturities, although the difference in time to flowering is trivial (Bernard, 1972; Ouattara and Weaver, 1994; Kilgore-Norquest and Sneller, 2000). In agreement with these studies, the determinate short duration C. volubils derivatives are short statured without secondary and tertiary branches. Moreover, they have more pods per inflorescence in lower pod heights than in inderterminate cultivars.

Isolation of candidate genes for determinacy in pigeonpea can facilitate molecular breeding programs for pigeonpea improvement. The validated markers may prove useful in marker-assisted selection (MAS) in breeding programmes for selecting the lines carrying allele for DT trait. Such markers, eventually, may allow the development of ideal DT genotypes for environments with moderate growth (5–6 t/ha) having 30–35% harvest index, initial vigour and tolerance to drought, and water logging (Singh and Oswalt, 1992).

125

TABLES

Table 2. Morphological traits of F3 BC 1 population of the cross C. cajan (ICPL87119) X C. cajanifolius (ICPW29) Streaks Plant No. of branches S. BC1 F3 Generation Branching Stem Leaf size & Flower size on Days to height 0 0 0 Growth habit No 1 2 3 pattern colour shape & colour standard flower (cm) Mean configuration for each branches branches branches petal

line Large, 1 18-5-1 104 81 4.5 12 0.5 Compact Indeterminate Green Large, Normal Orange None Semi Small, 2 37-1-1 95 103.7 8 9.6 0.3 Compact Green Small, Nomal determinate Yellow None Small, 3 76-2-1 111 65 9.3 14 0.8 Compact Indeterminate Green Small, elliptic Yellow None Semi Small, 4 82-2-1 127 69 6 6.5 0 Compact Purple Medium, obovate determinate Yellow None Semi Semi Large, 5 113-2-1 107 82.5 8.9 12.7 4.2 Green Large, Normal spreading determinate Orange None Large, 6 129-1-1 96.4 132.8 9.2 27.2 2.2 Spreading Indeterminate Green Large, Normal Yellow None Medium, 7 130-1-1 105 166 12 39 1.5 Spreading Indeterminate Green Large, Normal Yellow None Large, 8 131-1-1 106 159 14 26 8 Spreading Indeterminate Green Large, Normal Yellow None Large, 9 132-1-1 119 158 14.4 36.8 6 Spreading Indeterminate Green Large, Normal Yellow None Large, 10 133-1-1 112 150.8 11.3 25.8 11 Spreading Indeterminate Green Large, Normal Yellow None Large, 11 (ICPL87119 X ICPW29)-1 F2 127 142.3 8.8 16.6 1.8 Spreading Indeterminate Green Large, Normal Yellow Sparse

126

Table 3. Fertility status of F1 hybrids derived from the cross C. cajan (ICPL 85010) X C. lanceolatus (ICP15639) and number of F2/BC1 seeds obtained from hybrids

No. of % of F1 hybrid F/S F2 No. of pollinations % of BC1 seeds seed Plant No status seeds pods (x ICPL 85010) pod set set P1 sterile 0 320 77 24 210 65.6 P2 fertile 111 375 30 8 73 19.4 P3 fertile 33 231 15 6.4 20 8.6 P4 sterile 0 239 74 30.9 207 86.61 P5 fertile 7 247 19 7.6 31 12.5 P6 fertile 2 286 18 6.2 46 16.08 P7 sterile 0 343 78 22.7 320 93.2 P8 fertile 35 158 21 13.2 48 30.3 P9 fertile 27 186 0 0 0 0 P10 sterile 0 848 80 9.4 200 23.5 P11 fertile 36 15 0 0 0 0 P12 sterile 0 690 70 10.1 185 26.8 P13 fertile 24 30 0 0 0 0 P14 fertile 32 13 0 0 0 0

F:Fertile; S:Sterile

127

Table 4. Morphological traits of F1 hybrids derived from the cross C. cajan (ICPL 85010) X C. lanceolatus (ICP15639) NO. of branches Streaks Days Plant S. height Branching Growth Flower on Seed F1 hybrids to Pod colour No 0 0 0 pattern habit colour standard colour flower 1 branches 2 branches 3 branches (cm) petal Semi- Semi- Green+light 1 ICPL 85010 (C. cajan) 84 165 5 8 5 Yellow None Brown spreading determinate purple 2 ICP 15639 (C. lanceolatus) 178 285 6 14 2 Spreading Indeterminate Pale yellow Sparse Pale Green Black Orange 3 98 298 7 11 0 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P1 yellow Orange 4 112 235 8 13 2 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P2 yellow Orange 5 125 300 11 9 3 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P3 yellow Orange 6 160 345 6 5 0 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P4 yellow Orange 7 145 285 8 9 4 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P5 yellow Orange 8 130 320 6 4 2 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P6 yellow Orange 9 121 335 7 6 3 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P7 yellow Orange 10 112 295 9 8 2 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P8 yellow 11 ICPL 85010 X ICPW15639 P9 155 312 8 12 3 Spreading Indeterminate Pale yellow Sparse Purple Black Orange 12 158 325 9 8 3 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P10 yellow Orange 13 122 370 11 8 0 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P11 yellow Orange 14 135 300 6 18 2 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P12 yellow Orange 15 142 380 7 13 4 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P13 yellow Orange 16 160 330 9 16 0 Spreading Indeterminate Sparse Purple Black ICPL 85010 X ICPW15639 P14 yellow

128

Table 5. Percentage of pod set in sterile F1 hybrids (P-1, P-4, P-7, P-10 and P-12) derived from C. lanceolatus when crossed with different unrelated pigeonpea cultivars (ICPL 85010, ICPL 88039, ICPL 88034, ICPL 85030, MN1, MN5, MN8 and ICPL 92016)

No. of pollinated Sterile hybrids(♀) Pollinator(♂) % of BC1 pod set buds

ICPL 85010 X ICP15639 ICPL 85010 320 20.9 F1 P-1 ICPL 88039 125 11.2 ICPL 88034 78 0 ICPL 85030 170 1.1 MN 1 86 8.1

MN 5 128 6.2 ICPL 85010 X ICP15639 ICPL 85010 239 30.9 F1 P-4 ICPL 88039 25 24 ICPL 88034 25 16 ICPL 92016 25 12 MN 1 35 28.5

MN 5 113 3.5 ICPL 85010 X ICP15639 ICPL 85010 343 22.7 F1 P-7 ICPL 88039 28 14.2 ICPL 88034 30 13.3 MN 1 40 22.5 MN 5 68 1.4 ICPL 85010 X ICP15639 ICPL 85010 710 11.2 F1 P-10 ICPL 88039 136 2.2 ICPL 88034 70 1.4 ICPL 85030 145 2.7 MN 1 159 3.7 MN 5 160 5 MN 8 65 1.5 ICPL 85010 X ICP15639 ICPL 85010 690 10.1 F1 P-12 ICPL 88039 235 1.7 ICPL 88034 175 6.8 ICPL 85030 162 0.6 MN 1 200 6.5 MN 5 150 9.3 MN 8 80 17.5

129

Table 6. Morphological traits of F1 BC1 [(ICPL 85010 X ICP15639) X ICPL 85010] and F2 lines from the cross C. cajan (ICPL 85010) X C. lanceolatus (ICP15639)

Plant No. of branches Streaks Days Plant S. Number Branching Stem Flower size & on to height Growth habit Leaf size & shape No Mean pattern colour colour standard flower (cm) 10 20 30 configuration petal branches branches branches for each line 1 F1 BC1 P1-1 109.8 96.5 8.2 10 0.4 Semi-Spreading Semi determinate Green Medium, lanceolate Medium,Yellow Sparse 2 F2 P2-1 119 81.25 2.2 2.5 0 Compact Determinate Green Small, sesame Large, Yellow Sparse 3 F1 BC1 P2-1 126 86.2 10.7 7.7 0.7 Semi-Spreading Semi determinate Green Medium, lanceolate Large, Yellow Sparse 4 F2 P3-1 138 47.3 10.6 5 0 Compact Determinate Green Medium, lanceolate Medium,Yellow Dense 5 F1 BC1 P4-1 132.8 74.5 6.2 6.7 1 Compact Determinate Green Small, elliptic,round Medium,Yellow Sparse 6 F1 BC1 P5-1 129.7 79.7 8.2 5 0 Semi-Spreading Semi determinate Purple Medium, lanceolate Medium,Yellow None 7 F1 BC1 P6-1 126.4 72.5 6.4 6.1 1 Compact Determinate Green small, sesame Medium,Yellow Sparse 8 F1 BC1 P7-1 126 70.4 4 1.7 0 Semi-Spreading Semi determinate Green Medium, lanceolate Medium,Yellow None 9 F2 P8-1 135 59.8 4.5 4.3 0 Compact Determinate Green Small, sesame Medium,Yellow None 10 F1 BC1 P8-1 124 76.2 7 7.3 3 Semi-Spreading Determinate Green Medium, lanceolate Medium,Yellow None 11 F2 P9-1 116.5 58.3 5 2 0 Compact Determinate Green Medium, lanceolate Medium,Yellow Dense 12 F1 BC1 P10-1 124.3 82.5 5 11 0 Semi-Spreading Semi determinate Green Medium, lanceolate Medium,Yellow None 13 F1 BC1 P12-1 126 90 4 14 0 Semi-Spreading Semi determinate Green small, sesame Medium,Orange Sparse

130

Table 7. Morphological traits of F1 BC3 [(ICPW 68 X ICPL 85010) X ICPL 85010 X ICPL 85010 X ICPL 85010] lines derived from C. platycarpus

Plant S.No Plant number Days to Stem No. of branches height Seed & generation flower colour Branching pattern Growth habit Flower nature Pod colour Seed colour (cm) Primary Secondary Tertiary Shape 1 F1 BC3 A1 70 90 Green 19 6 0 Spreading Determinate Open Green Orange Oval 2 F1 BC3 A2 75 70 Green 20 7 0 Spreading Determinate Closed Green Orange Oval 3 F1 BC3 A3 78 80 Green 19 6 0 Spreading Semi-determinate Closed Green Brown Oval 4 F1 BC3 A4 80 90 Green 20 2 0 Semi-spreading Determinate Open Mixed Brown Oval 5 F1 BC3 A5 72 115 Green 31 19 0 compact Semi-determinate Closed Mixed Black Oval 6 F1 BC3 A6 81 115 Green 9 15 0 Spreading Indeterminate Closed Green Orange Oval 7 F1 BC3 A7 86 130 Green 33 9 0 Spreading Semi-determinate Open+closed Mixed Brown Oval 8 F1 BC3 A8 90 100 Green 23 8 1 Spreading Determinate Closed Green Light brown Oval 9 F1 BC3 A9 88 95 Green 23 7 0 Spreading Indeterminate Open Mixed Brown Oval 10 F1 BC3 A10 81 115 Green 25 2 0 Spreading Determinate Closed Mixed Brown Oval 11 F1BC3 B1 112 120 Green 15 13 0 Spreading Indeterminate Closed Mixed Brown Oval 12 F1BC3 B2 122 130 Green 15 13 0 Spreading Indeterminate Closed Green Orange Oval 13 F1BC3 B3 112 105 Green 9 18 0 Spreading Indeterminate Open Green Brown Oval 14 F1BC3 B4 115 120 Green 17 20 0 Spreading Indeterminate Closed Green Brown Oval 15 F1BC3 B5 118 75 Green 18 21 2 Spreading Determinate Closed Green Brown Oval 16 F1BC3 B6 115 60 Purple 7 12 0 Spreading Indeterminate Closed Purple Black Oval 17 F1BC3 B7 122 70 Green 15 4 0 Semi-spreading Semi-determinate Open Green Brown Oval 18 F1BC3 B8 121 85 Green 13 6 0 Spreading Indeterminate Open+closed Green Brown Oval 19 F1BC3 B9 130 125 Green 7 17 0 Semi-spreading Indeterminate Open Green Orange Oval 20 F1BC3 C1 115 85 Green 19 8 0 Spreading Semi-determinate Open Green Brown Oval 21 F1BC3 C2 112 160 Green 27 23 0 Spreading Indeterminate Open Mixed Brown Oval 22 F1BC3 C3 112 140 Green 32 16 0 Spreading Indeterminate Closed Green Brown Elongate 23 F1BC3 C4 117 145 Green 19 21 0 Spreading Indeterminate Open+closed Green Brown Oval 24 F1BC3 C5 117 145 Green 20 22 0 Spreading Indeterminate Closed Green Brown Elongate 25 F1BC3 C6 118 175 Green 34 14 0 Spreading Indeterminate Closed Green Gray Elongate 131

26 F1BC3 C7 112 145 Green 31 19 0 Spreading Indeterminate Closed Green Gray Oval 27 F1BC3 C8 121 140 Green 24 15 0 Spreading Indeterminate Closed Purple Brown Oval 28 F1BC3 C9 121 105 Green 4 14 0 Spreading Indeterminate Closed Mixed Brown Oval 29 F1BC3 C10 122 105 Green 20 19 2 Compact Indeterminate Closed Green Brown Elongate 30 F1BC3 C11 130 150 Green 17 19 0 Compact Indeterminate Closed Green Brown Oval 31 F1BC3 C12 128 150 Purple 13 22 2 Compact Indeterminate Closed Green Brown Oval 32 F1BC3 D1 92 175 Green 15 57 0 Spreading Indeterminate Closed Green Orange Oval 33 F1BC3 D2 88 180 Green 17 52 5 Spreading Indeterminate Closed Green Brown Oval 34 F1BC3 D3 94 200 Green 20 54 0 Spreading Indeterminate Open Green Brown Oval 35 F1BC3 E1 70 100 Green 17 18 0 Compact Determinate Open Mixed Brown Oval 26 F1BC3 E2 67 95 Green 19 9 0 Compact Determinate Open Mixed Black Oval 37 F1BC3 E3 65 95 Green 17 25 0 Compact Semi-determinate Open Green Brown Oval 38 F1BC3 E4 70 100 Green 12 18 0 Compact Semi-determinate Open+closed Green Gray Elongate 39 F1BC3 E5 67 105 Green 23 25 0 Compact Semi-determinate Open Mixed Black Oval 40 F1BC3 E6 68 140 Green 26 14 0 Compact Semi-determinate Open Mixed Black Oval 41 F1BC3 E7 70 100 Green 31 43 0 Compact Indeterminate Open+closed Green Black Oval 42 F1BC3 E8 71 105 Green 25 22 0 Compact Indeterminate Open+closed Mixed Grey Oval 43 F1BC3 E9 64 100 Green 25 20 0 Compact Semi-determinate Open+closed Mixed Brown Oval 44 F1BC3 E10 65 140 Green 25 32 0 Compact Indeterminate Open Mixed Grey Oval 45 F1BC3 E11 64 115 Green 17 36 1 Compact Semi-determinate Open+closed Mixed Black Oval 46 F1BC3 E12 68 140 Green 25 27 0 Compact Indeterminate Open Mixed Brown Oval 47 F1BC3 E13 67 120 Green 20 29 0 Compact Indeterminate Open Green Black Oval 48 F1BC3 E14 64 130 Green 18 33 0 Compact Indeterminate Open Green Brown Oval

132

Table 8. Number of pods and viable seeds per plant recorded in F1 BC3 [(ICPW 68 X ICPL 85010) X ICPL 85010 X ICPL 85010 X ICPL 85010] lines derived from the C. platycarpus.

Number and nature of cross seeds (F1 BC4 ) Number and nature of self seeds ( F2 BC3 ) S.No No.of cross Plant number No.of self pods (F1 & generation pods(F2 BC3) BC4)

Round Semi-shrunken Fully shrunken Non viable Round Semi shrunken Fully shrunken Non viable 1 F1 BC3 A1 65 93 9 1 35 0 0 0 0 0 2 F1 BC3 A2 61 76 1 0 38 3 3 0 0 2 3 F1 BC3 A3 19 21 0 0 12 1 2 0 0 0 4 F1 BC3 A4 58 91 5 2 43 6 9 0 0 6 5 F1 BC3 A5 66 103 8 0 42 3 4 0 0 2 6 F1 BC3 A6 56 81 2 0 33 6 10 0 0 4 7 F1 BC3 A7 37 39 1 0 48 2 2 0 0 2 8 F1 BC3 A8 86 106 14 0 78 17 24 4 0 5 9 F1 BC3 A9 95 160 10 0 70 15 21 1 0 10 10 F1 BC3 A10 3 2 0 0 3 0 0 0 0 0 11 F1BC3 B1 39 58 7 2 30 8 10 2 0 7 12 F1BC3 B2 33 54 1 0 21 0 0 0 0 0 13 F1BC3 B3 53 84 10 0 28 13 18 2 0 12 14 F1BC3 B4 19 26 0 0 12 55 64 2 0 46 15 F1BC3 B5 4 4 0 0 3 0 0 0 0 0 16 F1BC3 B6 12 9 4 0 2 0 0 0 0 0 17 F1BC3 B7 5 2 3 0 4 0 0 0 0 0 18 F1BC3 B8 7 8 0 0 11 0 0 0 0 0 19 F1BC3 B9 2 2 0 0 0 0 0 0 0 0 20 F1BC3 C1 44 70 3 0 25 53 81 0 2 46 21 F1BC3 C2 46 70 2 0 28 34 37 3 0 20 22 F1BC3 C3 24 54 0 3 15 148 226 30 68 80 23 F1BC3 C4 19 18 1 0 17 27 37 1 0 9 24 F1BC3 C5 11 28 0 0 5 25 30 3 4 18 25 F1BC3 C6 19 44 2 0 10 115 150 30 33 90 26 F1BC3 C7 35 58 1 0 15 156 190 30 30 130

133

27 F1BC3 C8 20 32 2 0 13 25 33 2 0 35 28 F1BC3 C9 49 85 5 1 9 59 82 1 0 25 29 F1BC3 C10 11 13 0 0 9 81 96 19 10 51 30 F1BC3 C11 5 6 0 0 0 140 146 17 6 115 31 F1BC3 C12 0 0 0 0 0 16 19 10 0 3 32 F1BC3 D1 107 210 17 0 45 70 138 1 1 56 33 F1BC3 D2 36 59 3 2 10 68 83 7 12 52 34 F1BC3 D3 65 66 8 0 65 0 0 0 0 0 35 F1BC3 E1 61 87 5 0 60 4 5 1 0 2 26 F1BC3 E2 102 146 16 0 50 1 1 0 0 1 37 F1BC3 E3 41 41 16 0 47 3 2 0 0 2 38 F1BC3 E4 18 23 18 0 8 34 36 8 0 25 39 F1BC3 E5 132 185 30 2 115 20 30 3 0 10 40 F1BC3 E6 37 43 10 0 47 3 3 0 0 2 41 F1BC3 E7 40 51 5 0 39 11 13 1 0 5 42 F1BC3 E8 103 146 14 0 78 91 111 17 0 59 43 F1BC3 E9 26 28 6 0 24 6 8 0 0 6 44 F1BC3 E10 123 248 13 0 55 65 98 4 0 35 45 F1BC3 E11 97 121 15 0 107 45 53 0 0 42 46 F1BC3 E12 54 55 12 0 60 10 11 1 0 9 47 F1BC3 F13 31 40 5 0 44 7 9 0 0 9 48 F1BC3 F14 58 78 9 0 70 20 22 3 0 17 TOTAL 2134 3124 293 1466 1917 203

134

Table 9. Morphological traits of advanced generation lines derived from the cross C. platycarpus (ICPW 68) X C. cajan (ICPL 85010)

No of branches Streaks Plant No. Days Plant Flower Branching Stem Leaf size on S. to height Growth habit size & pattern colour & shape standard No Mean flower (cm) colour 10 20 30 petal configuration branches branches branches for each line Semi Medium, Medium, F BC A5-7-1 111.5 74.1 6.3 16.6 0 Compact Green None 1 1 5 determinate normal Yellow

Semi Small, Medium, 2 F1 BC5 B6-1 102.4 74 4.6 9 0 Compact Purple None determinate normal Yellow

Semi Small, Medium, F3 BC3 C3-13-1 114.8 61.3 7 4.3 0 Spreading Green None 3 determinate normal Yellow

Semi Medium, Medium, F3 BC3 C7-13-1 110 117.5 5 35.5 8 Compact Green None 4 determinate Yellow normal

Small, Large, F3 BC3 C11-3-1 113.5 75 4.5 12 5 Spreading Indeterminate Purple None 5 normal Yellow

Small, Medium, F2 BC3 D2-1 95 71.6 9.3 8.8 0 Compact Determinate Green None 6 normal Yellow

135

Table 10. Morphological traits of F2 population derived from the cross C. cajan (ICPL 85010) X C. volubilis (ICP15774)

Plant Plant identification Days to Plant No. of No. of 20 No of Immature days to 50% Branches with Total Total no. 100 Seed S. No No. flower height(cm) 10branches branches 30branches Pods/plant pod set inflorescence dried pods of seeds weight (g) Female ICPL 85010 82 185 11 21 12 34 117 5 20 60 10.55 parent Male ICPW15774 142 325 9 18 9 12 198 24 8 29 4.6 parent Cultivar MN5 68 45 9 7 0 29 95 8 21 72 10 50 HYB-3-1-1A F2 55 33 5 0 0 32 82 5 35 71 5 50A HYB-3-1-1B F2 55 26 2 0 0 6 84 2 16 22 3.89 50B HYB-3-1-1C F2 58 18 5 0 0 33 84 5 7 15 7.48 51 HYB-3-1-2A F2 58 40 4 2 0 36 78 4 31 46 4.75 51A HYB-3-1-2B F2 58 29 2 1 0 53 78 2 16 26 5.9 51B HYB-3-1-2C F2 64 22 2 0 0 28 81 2 19 26 5.11 52 HYB-3-1-3A F2 64 24 2 0 0 15 84 2 13 18 6.14 52A HYB-3-1-3B F2 64 26 2 0 0 39 85 2 12 27 6.93 53 HYB-3-1-4A F2 60 27 3 2 0 30 80 3 18 42 5.13 53A HYB-3-1-4B F2 64 28 3 1 0 30 80 3 19 29 6.26 54 HYB-3-1-5A F2 64 34 4 0 0 66 83 4 25 65 5.23 54A HYB-3-1-5B F2 64 30 2 0 0 45 82 2 21 42 6.45 55 HYB-3-1-6A F2 68 35 4 0 0 52 86 4 28 63 4.92 55A HYB-3-1-6B F2 68 26 2 0 0 15 85 2 12 30 6.69 56 HYB-3-1-7A F2 59 32 3 0 0 61 79 3 25 48 6.04 56A HYB-3-1-7B F2 69 32 3 2 0 56 83 3 28 52 5.29 57 HYB-3-1-8A F2 54 32 1 0 0 10 85 1 20 34 4.96 57A HYB-3-1-8B F2 64 27 2 0 0 31 86 2 14 29 5.3 58 HYB-3-1-9A F2 57 30 1 0 0 33 82 1 12 25 6.7 58A HYB-3-1-9B F2 58 24 2 0 0 39 82 2 6 11 8.09 74 HYB-3-1-10A F2 54 33 2 0 0 32 83 2 4 10 6.7 74A HYB-3-1-10B F2 58 30 3 0 0 53 79 3 12 11 8.3 59 HYB-3-1-11A F2 58 23 1 0 0 28 86 1 7 21 6.2 60 HYB-3-1-12A F2 69 22 4 0 0 52 78 4 6 12 5.8

136

60A HYB-3-1-12B F2 58 32 0 0 0 12 82 0 20 28 5.64 61 HYB-3-1-13A F2 64 30 2 0 0 32 81 2 17 19 10 61A HYB-3-1-13B F2 60 24 3 0 0 45 82 3 8 20 8.6 62 HYB-3-1-14A F2 60 35 4 0 0 63 83 4 30 63 6.3 62A HYB-3-1-14B F2 58 26 0 0 0 19 82 0 4 13 7.78 63 HYB-3-1-15A F2 58 27 2 0 0 40 85 2 12 27 5.24 63A HYB-3-1-15B F2 60 30 3 0 0 33 86 3 11 26 6.26 64 HYB-3-1-16A F2 51 27 2 0 0 39 86 2 17 35 5.13 64A HYB-3-1-16B F2 57 34 3 0 0 42 81 3 31 55 6.5 65 HYB-3-1-17A F2 58 22 1 0 0 31 82 1 9 19 5.42 65A HYB-3-1-17B F2 53 21 0 0 0 17 85 0 15 30 9.78 66 HYB-3-1-18A F2 68 26 3 0 0 46 83 3 22 41 5.06 66A HYB-3-1-18B F2 68 28 2 0 0 44 83 2 15 30 5.8 67 HYB-3-1-19A F2 66 27 2 0 0 54 82 2 16 31 4.9 67A HYB-3-1-19B F2 66 37 1 0 0 54 82 1 23 39 7.19 68 HYB-3-1-20A F2 66 25 2 1 0 41 83 2 17 29 4.5 68A HYB-3-1-20B F2 66 32 3 0 0 27 81 3 23 44 5.4 69 HYB-3-1-21A F2 66 27 2 0 0 38 86 2 12 27 5.7 69A HYB-3-1-21B F2 66 36 2 1 0 47 79 2 38 58 4.57 70 HYB-3-1-22A F2 66 30 3 0 0 69 81 3 19 43 5.67 70A HYB-3-1-22B F2 68 37 2 0 0 49 82 2 27 65 5.02 71 HYB-3-1-23A F2 68 30 3 0 0 53 82 3 30 60 4.72 71A HYB-3-1-23B F2 68 28 3 0 0 39 83 3 23 58 4.22 71B HYB-3-1-23C F2 68 23 2 0 0 30 83 2 8 16 5.39 72 HYB-3-1-24A F2 68 30 3 0 0 51 84 3 22 39 7.38 72A HYB-3-1-24B F2 68 27 2 0 0 35 82 2 16 39 6.83 73 HYB-3-1-25A F2 68 37 3 0 0 60 82 3 27 52 6.19 73A HYB-3-1-25B F2 69 22 1 2 0 31 83 1 7 17 5.84 73B HYB-3-1-25C F2 69 17 1 0 0 4 98 3 4 8 6.93

137

Table 11. Morphological traitsof F4 population of the cross C. cajan X C. volubilis., S. DET-Semi-determinate and DT- Determinate

F Days to 50% Pod S. 4 Days to No. of branches Branches with Immature population Plant height (cm) Growth habit set from day of No Flower 0 0 inflorescence pods/plant Plant No. 1 branches 2 branches sowing 1 P1-3-1A 75 41 4 0 4 DET 112 61 2 P1-3-1B 79 49 4 0 4 S.DET 112 38 3 P1-3-2A 89 28 6 0 6 DET 107 5 4 P1-3-3A 78 46 6 0 5 S.DET 111 22 5 P1-3-3B 82 29 5 0 5 DET 110 27 6 P1-3-4A 95 31 4 6 4 DET 111 29 7 P1-3-5A 82 38 5 2 4 DET 110 27 8 P1-3-5B 82 38 4 3 4 S.DET 110 18 9 P1-3-6A 76 30 5 2 4 DET 113 53 10 P1-3-6B 76 45 5 6 4 S.DET 113 10 11 P1-3-7A 71 38 7 0 6 DET 113 57 12 P1-3-7B 74 39 4 5 6 DET 111 35 13 P1-3-8A 99 38 6 3 8 S.DET 105 4 14 P1-3-9A 79 32 4 0 4 S.DET 112 5 15 P1-3-9B 96 41 4 4 3 S.DET 103 6 16 P1-3-10A 88 26 3 3 3 S.DET 102 2 17 P1-3-10B 95 31 5 0 4 DET 105 5 18 P1-3-11A 72 40 4 2 5 DET 108 58 19 P1-3-11B 78 53 4 0 4 S.DET 110 74 20 P1-3-12A 75 39 6 0 5 DET 110 16 21 P1-3-13A 82 56 6 3 5 S.DET 125 26 22 P1-6-1A 82 30 4 0 4 DET 128 9 23 P1-6-2A 73 39 5 0 4 DET 105 30 24 P1-6-2B 78 36 8 4 8 DET 109 40 25 P1-6-3A 78 25 4 0 4 DET 116 6 26 P1-6-4A 79 41 7 6 9 DET 114 55 27 P1-6-5A 82 35 5 0 4 DET 123 22

138

28 P1-6-5B 96 29 4 0 3 S.DET 128 5 29 P1-6-6A 79 34 5 2 5 DET 112 16 30 P1-6-6B 82 39 7 4 8 S.DET 113 22 31 P1-6-7A 82 35 5 4 8 DET 115 30 32 P1-6-7B 83 24 4 2 5 DET 119 13 33 P1-6-8A 79 48 7 6 6 S.DET 111 83 34 P1-6-9A 82 46 6 0 6 S.DET 127 8 35 P1-6-10A 78 32 5 2 5 DET 109 33 36 P1-6-10B 82 31 5 0 5 DET 112 37 37 P1-6-11A 91 32 5 2 4 S.DET 121 5 38 P1-6-11B 96 43 5 0 4 S.DET 124 6 39 P1-6-12A 79 37 4 0 3 DET 112 32 40 P1-6-12B 80 38 8 0 5 DET 122 21 41 P1-6-13A 82 30 5 0 4 DET 125 35 42 P1-6-15A 79 30 4 4 4 DET 111 3 43 P1-6-15B 82 37 4 0 4 DET 123 5 44 P1-6-17A 79 46 5 0 4 DET 110 17 45 P1-6-17B 96 35 4 0 3 DET 127 3 46 P1-6-18A 79 40 6 2 5 S.DET 113 59 47 P1-6-18B 82 46 6 3 6 S.DET 125 30 48 P1-6-19A 75 50 5 5 5 DET 110 79 49 P1-6-19B 99 24 7 4 4 S.DET 126 2 50 P1-6-20A 96 52 6 5 10 S.DET 127 12 51 P1-6-21A 72 26 8 0 8 DET 105 73 52 P1-6-21B 89 46 4 0 4 S.DET 127 12 53 P1-6-22A 79 25 4 0 3 S.DET 125 34

139

Table 12. Meiotic analysis of abnormal individual plants of F3BC1 lines derived from the C. cajanifolius indicates more homology between two parental chromosomes involved in the cross. *N.D-Normal distribution, A.D-Abnormal distribution at Anaphase I

Plant No. Metaphase I (mean configuration) Anaphase I( % ) Pollen Fertility

F3BC1 Univalents Bivalents Trivalents Tetravalents *N.D A.D (%) generation 18-5-1 0.4 (1-3) 10.4 (8-11) 0.2 (0-1) 0 (0) 55 45 76 18-5-2 0.8 (1-2) 10.05 (8-11) 0 (0) 0.15 (1-2) 45 55 80 37-1-3 0.15 (1-3) 10.65 (8-11) 0.15 (0-1) 0.05 (0-1) 60 40 86 37-1-4 0.55 (1-2) 9.7 (7-11) 0.25 (0-1) 0.5 (0.1) 60 40 55 76-2-1 0.7 (1-2) 10.25 (8-11) 0 (0) 0.05 (0-1) 65 35 63 76-2-2 0.45 (1-2) 10.3 (9-11) 0.25 (0-1) 0 (0) 65 35 88 76-2-3 0.7 (1-2) 9.95 (8-11) 0 (0) 0.35 (0-2) 30 70 82 82-2-1 0.15 (1-3) 10.5 (9-11) 0.05 (0-1) 0.3 (0-2) 70 30 66 82-2-2 0.5 (1-2) 10 (9-11) 0.3 (0-1) 0.2 (0-1) 55 45 76 82-2-3 0.35 (1-3) 10.55 (9-11) 0 (0) 0.1 (0-1) 80 20 68 82-2-5 0.6 (1-2) 10.1 (7-11) 0.3 (0-1) 0 (0) 65 35 60 82-2-7 1 (1-4) 9.55 (8-11) 0.2 (0-1) 0.25 (0-1) 25 75 52

140

Table 13. Meiotic studies of hybrids derived from the cross C. cajan (ICPL 85010) X C. lanceolatus (ICP 15639) indicates more homology between two parental chromosomes involved in the cross. *N.D-Normal distribution, A.D-Abnormal distribution at Anaphase I

Hybrid Pollen Metaphase I Anaphase I (%) Plant Fertility No. Univalents Bivalents Trivalents Tetravalents *N.D A.D (%) P-1 1.4 (1-4) 9.4 (8-11) 0.2 (0-1) 0 (0) 50 50 Sterile P-2 0.7 (1-4) 10.05 (7-11) 0.1 (0-1) 0.15 (1-2) 65 35 37.8 P-3 1.15 (1-4) 9.65 (8-11) 0.15 (0-1) 0.05 (0-1) 45 55 24.3 P-4 1.05 (1-5) 9.7 (7-11) 0.25 (0-1) 0 (0) 60 40 Sterile P-5 0.7 (1-4) 10.05 (8-11) 0.2 (0-1) 0.05 (0-1) 65 35 40.5 P-6 0.35 (1-4) 10.4 (7-11) 0.25 (0-1) 0 (0) 65 35 40.7 P-7 1 (1-4) 9.75 (8-11) 0.1 (0-1) 0.15 (0-2) 25 75 Sterile P-8 0.35 (1-3) 10.3 (8-11) 0.05 (0-1) 0.3 (0-2) 65 35 56 P-9 0.8 (1-4) 10 (9-11) 0.2 (0-1) 0 (0) 55 45 36 P-10 1.35 (1-4) 9.3 (7-11) 0.25 (0-1) 0.1 (0-1) 20 80 Sterile P-11 0.6 (1-2) 10.1 (7-11) 0.2 (0-1) 0.1 (0-1) 50 50 45 P-12 1.1 (1-4) 9.45 (8-11) 0.3(0-1) 0.15 (0-1) 35 65 Sterile P-13 0.65 (1-4) 9.9 (7-11) 0.15 (0-2) 0.3 (0-2) 50 50 50 P-14 0.75 (1-4) 9.75 (7-11) 0.15 (0-2) 0.35 (0-2) 45 55 48

141

Table 14. Segregation of Male sterility in F1 BC1 [(ICPL 85010 X ICP 15639) X ICPL 85010] plants derived from the crosses between three sterile F1s and pigeonpea cultivars.F:Fertile, S:Sterile

Pollen sterility in F BC plants (%) 1 1 Cross Pollinator Total No. of 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 plants F 100%S (ICPL ICPL 85010 0 1 1 1 1 0 0 1 2 2 9 7 2 85010 X ICP15639) F1 P-4 ICPL 88039 0 1 1 0 1 1 1 2 2 1 10 10 0 ICPL 88034 0 0 0 0 1 0 1 2 0 0 4 4 0 ICPL 92016 0 0 0 0 0 0 0 1 1 0 2 2 0 MN -1 0 0 0 1 3 2 0 0 0 3 9 7 2 MN-5 0 0 0 0 0 0 0 1 0 1 2 2 0 (ICPL ICPL 85010 0 0 2 0 2 1 2 0 0 0 10 8 2 85010 X ICP15639) F1 P-7 ICPL 88039 0 0 1 0 1 0 1 0 0 0 3 3 0 ICPL 88034 0 0 0 0 1 2 1 0 0 0 4 4 0 MN -1 0 0 0 1 0 0 3 1 0 3 8 6 2 MN-5 0 0 0 2 1 0 0 0 0 0 3 3 0 (ICPL 85010 X ICPL 85010 0 0 0 2 5 5 3 2 2 1 20 19 1 ICP15639) F1 P-12 ICPL 88039 0 0 0 0 0 0 2 2 0 1 5 4 1 ICPL 88034 0 0 0 0 3 1 2 2 0 0 8 8 0 ICPL 85030 0 0 0 0 0 0 0 0 0 1 1 0 1 MN -1 0 0 0 1 3 2 2 2 1 7 17 10 7 MN-5 0 0 0 0 3 0 1 4 2 1 11 10 1 MN-8 0 0 0 0 1 2 2 0 0 1 6 5 1

142

Table 15. Meiotic analysis of F1 BC3 [(ICPW 68 X ICPL 85010) X ICPL 85010 X ICPL 85010 X ICPL 85010)] lines derived from C. platycarpus

Hybrid plant Anaphase I Metaphase I (Mean chromosomal configuration) Pollen no. (Distribution %) S.no fertility (ICPW 68 X Bivalents Univalents Trivalents Tetravalents N.D A.D (%) ICPL85010) Ring Rod

1 BC3 F1 A1 0.3 6.55 4.15 0.1 0 70 30 16.3

2 BC3 F1 A2 0.3 7.35 3.35 0.1 0 75 25 46.4

3 BC3 F1 A3 0.6 7.15 2.3 0.1 0.65 30 70 48.5

4 BC3 F1 A4 0.05 9 1.4 0.05 0.25 55 45 49.8

5 BC3 F1 A5 0.45 8.75 1.7 0.15 0.1 70 30 40.4

6 BC3 F1 A6 0.1 9 1.55 0 0.2 80 20 45.02

7 BC3 F1 A7 0 6.55 3.05 0 0.7 75 25 48.7

8 BC3 F1 A8 0.25 8.1 2.1 0.05 0.25 15 75 54.3

9 BC3 F1 A9 0.3 5.45 5.25 0.1 0 80 20 50.9

10 BC3 F1 A10 0.2 5.6 5.15 0.1 0 70 30 45.5

11 BC3 F1 B1 0 4.05 7 0 0.15 85 15 44.5

12 BC3 F1 B2 0 8.25 2.35 0 0.2 75 25 27.1

13 BC3 F1 B3 0.1 8.5 2 0 0.2 65 35 53.3

14 BC3 F1 B4 0 7.85 2.9 0 0.15 65 35 53.8

15 BC3 F1 B5 0.1 2 8.55 0 0.2 70 30 64.3

16 BC3 F1 B6 0.9 1 8.35 0.4 0.35 60 40 11.5

17 BC3 F1 B7 0 5.1 5.7 0 0.1 65 35 17.4

18 BC3 F1 B8 0.15 6.8 3.75 0.05 0.15 70 30 75.3

19 BC3 F1 B9 0.05 7.65 3.25 0.05 0 60 40 5.05

20 BC3 F1C1 0.35 6.8 2.85 0.05 0.35 60 40 85.5

21 BC3 F1C2 0.65 4.95 5.3 0.05 0.15 65 35 78

22 BC3 F1C3 0 6.4 4.4 0 0.1 90 10 84.7

23 BC3 F1C4 0.1 7.8 2.45 0 0.2 75 25 82.6

24 BC3 F1C5 0 8.65 2.25 0 0.05 80 20 96.4

25 BC3 F1C6 0 7.55 3.0 0 0.45 90 10 89.6

26 BC3 F1C7 0 8.7 1.8 0 0.25 80 20 98.3

27 BC3 F1C8 0.1 7.6 2.6 0 0.4 95 5 83.13

28 BC3 F1C9 0 7.55 2.75 0 0.35 85 15 75

29 BC3 F1C10 0 8.15 2.85 0 0 90 10 83

143

30 BC3 F1C11 0 7.65 3.15 0 0.1 65 35 82.8

31 BC3 F1C12 0.2 7.25 3.45 0 0.1 95 5 97.4

32 BC3 F1 D1 0.15 6.75 3.4 0.05 0.65 85 15 74.1

33 BC3 F1 D2 0.05 0.45 9.35 0.25 0.4 95 5 94.1

34 BC3 F1 D3 0.3 5.7 3.65 0 0.75 50 50 54.8

35 BC3 F1 E1 0.05 9.85 0.85 0.05 0.1 75 25 40.1

26 BC3 F1 E2 0.25 8.6 2.05 0.05 0.05 70 30 56.2

37 BC3 F1 E3 0.2 5.9 4.65 0.1 0.15 80 20 74.5

38 BC3 F1 E4 0.3 7.9 2.6 0.1 0.1 75 25 42.6

39 BC3 F1 E5 0.35 6.6 3.35 0.15 0.05 70 30 94.2

40 BC3 F1 E6 0.2 7.65 3 0.1 0.05 75 25 45.7

41 BC3 F1 E7 0.4 7.35 3.15 0.05 0.1 65 35 30.4

42 BC3 F1 E8 0.35 6.6 4 0.05 0.05 70 30 87.4

43 BC3 F1 E9 0.9 5.35 3.85 0.6 0.3 75 25 34.2

44 BC3 F1 E10 0.3 6.85 3.9 0.1 0 60 40 72.8

45 BC3 F1 E11 0.3 8.6 2 0.1 0.05 70 30 47.46

46 BC3 F1 E12 0.3 7.15 3 0.1 0 70 30 41.6

47 BC3 F1 E13 0 5.7 4.95 0.1 0.1 75 25 71.9

48 BC3 F1 E14 0.4 8.95 1.7 0.1 0 75 25 71

144

Table 16. Polymorphism status of SSR markers tested on eight F1 hybrids derived from the cross C. cajan (ICPL85010) X C. lanceolatus (ICP15639) and two parental genotypes. *NA- Not amplified.

SSR S. No PARENT1 PARENT 2 Allele size (bp) in F1 Hybrids Markers ICPL 85010 ICP15639 amplified P-1 P-2 P-4 P-7 P-8 P-9 P-10 P-11 1 CcM0008 195.93 202.46 202.28 NA 194.33 203.19 202.02 NA 202.12 NA 194.64 203.41 194.71 NA 202.06 NA 194.57 203.27

2 CcM0035 275.95 275.07 275.25 275.91 275.07 275.66 275.07 275.12 275.07 275.17

3 CcM0047 174.91 185.36 174.92 185.45 174.84 185.46 174.94 185.55 175.06 185.53 175.09 185.60 175.06 185.51 175.02 185.65 175.00 185.61

4 CcM0057 290.81 283.78 284.01 290.9 283.49 290.63 283.8 290.79 283.69 290.75 284.08 291.19 284.04 290.95 284.32 291.52 284.1 291.51

5 CcM0252 246.55 NA 246.94 255.37 242.70 253.28 246.84 257.07 246.74 257.17 242.45 253.04 241.98 253.44 246.77 257.77 243.35 253.53

6 CcM0416 147.33 NA 147.15 154.62 147.34 155.61 147.64 154.66 147.37 154.32 148.96 154.54 147.16 154.87 147.12 154.75 147.37 154.74

7 CcM0710 297.06 286.99 287.09 297.21 287.16 297.18 287.05 297.05 287.09 297.21 287.12 297.06 287.08 297.14 286.80 297.01 286.73 297.09

8 CcM0974 175.31 169.23 179.20 171.90 175.19 172.08 179.32 171.85 179.22 171.47 175.34 167.79 175.44 168.19 179.69 168.23 175.58 167.89

9 CcM0988 247.70 NA 245.78 252.04 240.94 248.03 245.14 252.14 244.93 253.87 240.43 246.82 240.56 250.18 244.71 254.01 240.94 250.27

10 CcM1385 272.26 NA 272.05 NA 271.99 NA 271.88 NA 272.06 NA 271.98 NA 271.96 NA 272.15 NA 272.17 NA

11 CcM1447 281.89 NA 281.66 NA 281.65 NA 281.71 NA 284.67 NA 281.76 NA 281.94 NA 281.58 NA 281.71 NA

12 CcM1459 183.52 183.40 183.14 183.10 183.40 183.46 183.46 183.19 183.83 183.76

13 CcM1991 208.52 194.55 194.34 206.3 194.47 206.51 194.98 205.92 194.45 206.18 194.14 208.02 194.34 208.29 194.51 208.53 194.38 208.41

14 CcM1999 173.53 NA 173.24 NA 173.44 NA 173.22 NA 173.52 NA 173.47 NA 173.44 NA 173.41 NA 173.17 NA

15 CcM2012 240.26 236.40 236.31 240.02 236.29 239.85 236.24 239.95 236.21 239.98 236.27 239.81 236.34 240.02 237.33 240.92 237.09 240.80

16 CcM2066 119.23 NA 121.01 NA 119.14 NA 121.24 NA 121.16 NA 119.42 NA 119.50 NA 120.85 NA 118.97 NA

17 CcM2071 199.32 215.98 215.87 NA 217.93 NA 215.99 NA 215.91 NA 217.87 NA 217.93 NA 215.77 NA 217.78 NA

18 CcM2095 246.17 NA 242.22 NA 242.24 NA 242.07 NA 242.09 NA 242.30 NA 242.39 NA 242.07 NA 242.16 NA

19 CcM2176 273.09 271.80 271.93 NA 271.89 NA 272.05 NA 271.64 NA 272.29 NA 272.76 NA 272.03 NA 272.88 NA

20 CcM2228 289.07 311.98 289.03 311.19 288.82 311.21 289.22 311.75 288.99 311.89 289.15 311.89 288.90 312.08 289.14 312.31 288.88 311.91

21 CcM2379 165.11 NA 165.45 NA 168.20 NA 168.12 NA 167.51 NA 165.47 NA 165.45 NA 165.80 NA 167.86 NA

22 CcM2451 183.37 NA 183.44 NA 183.30 NA 183.30 NA 183.30 NA 183.53 NA 183.73 NA 183.50 NA 183.60 NA

23 CcM2505 217.00 228.60 217.16 228.70 216.99 228.76 217.06 228.60 217.23 228.69 217.36 228.75 217.14 228.68 217.16 228.76 217.10 228.60

24 CcM2639 145.78 165.98 145.44 166.33 145.65 166.37 145.67 166.21 145.66 166.48 145.74 166.19 145.82 166.43 145.40 166.06 145.49 165.97

25 CcM2672 243.72 243.98 243.66 243.66 243.55 243.83 243.69 243.83 243.80 244.01

26 CcM2707 239.49 242.73 238.24 NA 242.70 NA 238.20 NA 238.24 NA 242.67 NA 242.76 NA 237.63 NA 241.44 NA

27 CcM2855 282.46 270.56 282.46 NA 270.03 NA 270.03 NA 270.07 NA 270.15 NA 270.03 NA 270.21 NA 270.22 NA

145

Table 17. Genotype scoring of 14 polymorphic primers among eight F1 hybrids & two parental genotypes of the cross C. cajan X C. lanceolatus and percentage of hybrid purity index for each marker

Genotype scoring in F1 hybrids Hybrid purity index SSR Markers ICPL 85010 ICP15639 S. No for each marker amplified PARENT1 PARENT2 P-1 P-2 P-4 P-7 P-8 P-9 P-10 P-11 (%)

1 CcM0008 A B B H B B H A B H 87.5 2 CcM0047 A B H H H H H H H H 100 3 CcM0057 A B H H H H H H H H 100 4 CcM0710 A B H H H H H H H H 100 5 CcM0974 A B H H H H H H H H 100 6 CcM1991 A B H H H H H H H H 100 7 CcM2012 A B H H H H H H H H 100 8 CcM2071 A B B B B B B B B B 100 9 CcM2176 A B B B B B B A B A 75 10 CcM2228 A B H H H H H H H H 100 11 CcM2505 A B H H H H H H H H 100 12 CcM2639 A B H H H H H H H H 100 13 CcM2707 A B A B A A B B A B 50 14 CcM2855 A B A B B B B B B B 87.5

146

Table 18. Percentage of hybrid purity in F1 individuals of the cross C. cajan X C. lanceolatus.

Total number of alleles F hybrids Allele type 1 P-1 P-2 P-4 P-7 P-8 P-9 P-10 P-11 A allele type 2 0 1 1 0 2 1 1 B allele type 3 4 3 4 4 3 4 3 H' allele type 9 10 10 9 11 9 9 10 % of hybrid 85.71 100 92.85 92.85 100 85.71 92.85 92.85 purity

Table 19. Polymorphism status of the Primers screened to test the F1 hybrid derived from the cross C. cajan (ICPL 85010) X C. volubilis (ICP 15774).

SSR Allele size S. No Marker ICPL85010 ICP15774 F1 (85010 x 15774) 1 CCM0008 194.88 202.52 202.53 2 CCM0047 167.84 168.19 167.96 3 CCM0095 223.67 224.18 224.3 4 CCM0181 288 Not amplified Not amplified 5 CCM0188 194.54 196.92 Not amplified 6 CCM0282 178.72 171.64 178.79 7 CCM0444 207.61 207.78 Not amplified 8 CCM0445 247.51 247.66 Not amplified 9 CCM0473 255.36 255.44 253.79 10 CCM0516 206.27 Not amplified 205.35 11 CCM0594 204.66 198.22 203.02 12 CCM0603 245.87 245.86 245.88 13 CCM0790 268.26 269.8 268.8 14 CCM0858 186.68 188.61 187.03 15 CCM0885 148.33 146.96 148.15 16 CCM0887 186.22 176.41 Not amplified 17 CCM0960 257.46 Not amplified 256 18 CCM1331 246.27 246.79 246.21 19 CCM1411 158.11 158.11 Not amplified 20 CCM1454 293.68 283.24 Not amplified 21 CCM0717 118.26 117.05 117.13

147

Table 20. Percentage of pod damage by Helicoverpa armigera in the infested field correlated with the Helicoverpa mid gut trypsin inhibitors

Helicoverpa Healthy Total % of Plant Identification Number damage pods Pods/plant Damage (ICPW 68 X ICPL 85010) derivatives F2 BC5 B6-2 3 36 39 7.69 F4 BC3 C7-13-1 7 137 144 4.86 F4 BC3 C7-13-2 27 900 927 2.91 F4 BC3 C11-3-1 19 500 519 3.66 F4 BC3 C11-3-2 4 150 154 2.59 (ICPL 87119 X ICPW29) derivatives F2 (ICPL 87119 X ICPW29)-2 50 360 410 12.19 F2 (ICPL 87119 X ICPW29)-7 18 536 554 3.24 F3 BC1 129-1-1 26 360 386 6.73 F3 BC1 129-1-2 15 260 275 5.45 F3 BC1 129-1-4 3 163 166 1.8 F3 BC1 131-1-2 6 460 466 1.28 F3 BC1 133-1-1 35 197 232 15.08 (ICPL 85010 X ICPW15639) derivatives F2 BC1 P1-1 2 51 53 3.77 F2 BC1 P1-2 4 36 40 10 F2 BC1 P2-1 3 96 99 3.03 F2 BC1 P2-2 2 20 22 9.09 F2 BC1 P8-3 5 77 82 6.09 F2 BC1 P10-3 3 42 45 6.66

Susceptible Cultivar parents ICPL 85010 56 62 118 47.45 ICPL 87119 73 91 164 44.51

148

Table 21. Percentage of pod damage in F3 BC1 lines of C. cajanifolius

Plant identification Number Number of HA Total % of healthy pods damage pods/ plant damage S. no ICPW 87119 X ICPW 29 1 (ICPW 87119 X ICPW 29) F2-1 58 10 68 14.7 2 (ICPW 87119 X ICPW 29) F2-2 360 50 410 6.6 3 (ICPW 87119 X ICPW 29) F2-4 62 9 71 12.6 4 (ICPW 87119 X ICPW 29) F2-5 185 19 204 9.3 5 (ICPW 87119 X ICPW 29) F2-6 23 4 27 14.8 6 F3 BC1 18-5-3 530 57 587 9.71 7 F3 BC1 18-5-4 7 1 8 12.5 8 F3 BC1 18-5-9 11 2 13 15.3 9 F3 BC1 37-1-1 54 9 63 14.2 10 F3 BC1 76-2-1 8 1 9 11.1 11 F3 BC1 76-2-2 45 6 51 11.7 12 F3 BC1 76-2-4 6 0 6 0 13 F3 BC1 76-2-5 54 9 63 14.2 14 F3 BC1 113-2-1 35 6 41 14.6 15 F3 BC1 113-2-4 144 12 156 7.6 16 F3 BC1 113-2-5 28 0 28 0 17 F3 BC1 113-2-6 3 0 3 0 18 F3 BC1 113-2-10 180 21 201 10.4 19 F3 BC1 129-1-1 360 26 386 3.7 20 F3 BC1 129-1-2 260 15 275 4.6 21 F3 BC1 129-1-3 36 2 38 5.26 22 F3 BC1 129-1-4 163 3 166 1.8 23 F3 BC1 129-1-5 132 6 138 4.34 24 F3 BC1 129-1-6 246 12 258 4.65 25 F3 BC1 129-1-7 32 2 34 5.88

26 F3 BC1 130-1-1 245 20 265 7.54

27 F3 BC1 130-1-2 185 10 195 5.12

28 F3 BC1 130-1-3 210 11 221 4.97

29 F3 BC1 130-1-4 208 30 238 12.6

30 F3 BC1 131-1-1 207 21 228 9.21

31 F3 BC1 131-1-2 245 18 263 6.84

32 F3 BC1 131-1-3 168 11 179 6.14

33 F3 BC1 131-1-4 137 9 146 6.16

34 F3 BC1 131-1-7 420 29 449 6.45

35 F3 BC1 132-1-1 57 4 61 6.55

149

36 F3 BC1 132-1-2 235 11 246 4.47 37 F3 BC1 132-1-3 41 5 46 10.86 38 F3 BC1 132-1-4 209 8 217 3.68 39 F3 BC1 132-1-6 76 3 79 3.79 40 F3 BC1 132-1-7 381 11 392 2.8 41 F3 BC1 133-1-1 235 8 243 3.29 42 F3 BC1 133-1-3 44 0 44 0 43 F3 BC1 133-1-4 27 1 28 3.57 44 F3 BC1 133-1-5 285 13 298 4.36 45 F3 BC1 133-1-6 180 9 189 4.76

1 (ICPW 87119 X ICPW 29) F2-1 60 5 65 7.69 2 (ICPW 87119 X ICPW 29) F2-3 15 0 15 0 3 (ICPW 87119 X ICPW 29) F2-4 280 11 291 3.78 4 (ICPW 87119 X ICPW 29) F2-5 44 3 47 6.38 5 (ICPW 87119 X ICPW 29) F2-7 536 18 554 3.24 6 (ICPW 87119 X ICPW 29) F2-8 142 18 160 11.25 7 (ICPW 87119 X ICPW 29) F2-9 22 0 22 0 8 F3 BC1 18-5-2 36 5 41 12.19 9 F3 BC1 18-5-3 32 3 35 8.57 10 F3 BC1 18-5-4 13 0 13 0 11 F3 BC1 18-5-5 14 1 15 6.66 12 F3 BC1 37-1-1 22 2 24 8.33 13 F3 BC1 37-1-5 5 0 5 0 14 F3 BC1 37-1-6 30 5 35 14.28 15 F3 BC1 76-2-3 48 4 52 7.69 16 F3 BC1 82-2-2 102 11 113 9.73 17 F3 BC1 82-2-6 13 2 15 13.33 18 F3 BC1 82-2-7 11 0 11 0 19 F3 BC1 129-1-1 410 21 431 4.87 20 F3 BC1 129-1-2 120 10 130 7.69 21 F3 BC1 129-1-3 195 9 204 4.41 22 F3 BC1 129-1-4 115 7 122 5.73 23 F3 BC1 129-1-5 83 5 88 5.68 24 F3 BC1 129-1-6 95 9 104 8.65 25 F3 BC1 129-1-7 145 12 157 7.64 26 F3 BC1 129-1-8 56 6 62 9.67 27 F3 BC1 129-1-9 290 16 306 5.22 28 F3 BC1 129-1-10 176 11 187 5.88

29 F3 BC1 130-1-1 56 3 59 5.08

150

30 F3 BC1 130-1-4 59 4 63 6.34

31 F3 BC1 130-1-5 38 0 38 0

32 F3 BC1 130-1-6 104 9 113 7.96

33 F3 BC1 130-1-7 41 3 44 6.81

34 F3 BC1 130-1-8 10 0 10 0

35 F3 BC1 130-1-9 26 1 27 3.7

36 F3 BC1 130-1-11 320 18 338 5.32

37 F3 BC1 131-1-2 460 6 466 1.28

38 F3 BC1 131-1-3 48 2 50 4

39 F3 BC1 131-1-4 170 9 179 5.02

40 F3 BC1 131-1-5 22 2 24 8.33

41 F3 BC1 131-1-6 349 18 367 4.9

42 F3 BC1 132-1-1 246 22 268 8.2

43 F3 BC1 132-1-2 6 0 6 0 44 F3 BC1 132-1-3 186 25 211 11.84 45 F3 BC1 132-1-5 214 20 234 8.54 46 F3 BC1 133-1-1 320 14 334 4.19 47 F3 BC1 133-1-2 72 6 78 7.69 48 F3 BC1 133-1-3 286 11 297 3.7 49 F3 BC1 133-1-4 78 3 81 3.7 50 F3 BC1 133-1-5 170 16 186 8.6 51 F3 BC1 133-1-6 186 12 198 6.06 52 F3 BC1 133-1-7 87 5 92 5.43 53 F3 BC1 133-1-8 360 10 370 2.7

151

Table 22. Percentage of pod damage in interspecific derivatives of C. lanceolatus

Number of HA S. No PLANT NUMBER healthy % of pod damage Total pods podspods damage

(ICPL 85010 X ICPW 15639) 1 F2 BC1 P1-1 51 2 53 3.77 2 F2 BC1 P1-2 36 4 40 10 3 F2 BC1 P1-3 15 1 16 6.25 4 F2 BC1 P1-4 7 0 7 0 5 F2 BC1 P1-7 10 1 11 9.09 6 F2 BC1 P2-2 8 2 10 20 7 F2 BC1 P2-3 14 3 17 17.64 8 F2 BC1 P2-6 4 0 4 0 9 F2 BC1 P5-3 3 0 3 0 10 F2 BC1 P5-5 7 0 7 0 11 F2 BC1 P5-7 5 0 5 0 12 F2 BC1 P6-3 8 0 8 0 13 F2 BC1 P6-7 4 0 4 0 14 F2 BC1 P7-11 4 0 4 0 15 F2 BC1 P8-1 6 0 6 0 16 F2 BC1 P8-3 77 5 82 6.09 17 F2 BC1 P8-7 7 0 7 0 18 F2 BC1 P1-2 6 0 6 0 19 F2 BC1 P1-4 10 0 10 0 20 F2 BC1 P1-5 22 1 23 4.34 21 F2 BC1 P1-6 15 1 16 6.25 22 F2 BC1 P2-1 96 3 99 3.03 23 F2 BC1 P2-2 20 2 22 9.09 24 F2 BC1 P2-3 13 1 14 7.14 25 F2 BC1 P4-1 4 0 4 0 26 F2 BC1 P4-3 49 2 51 3.9 27 F2 BC1 P4-4 3 0 3 0 28 F2 BC1 P4-6 5 0 5 0 29 F2 BC1 P4-7 34 2 36 5.55 30 F2 BC1 P6-1 37 2 39 5.12 31 F2 BC1 P6-3 52 3 55 5.45 32 F2 BC1 P7-1 4 0 4 0 33 F2 BC1 P7-4 4 0 4 0 34 F2 BC1 P7-6 21 1 22 4.54 35 F2 BC1 P8-1 14 0 14 0

152

36 F2 BC1 P8-4 6 0 6 0 37 F2 BC1 P10-1 32 1 33 3.03 38 F2 BC1 P10-2 12 0 12 0 39 F2 BC1 P10-3 42 3 45 6.66 40 F2 BC1 P10-6 25 2 27 7.4 41 F2 BC1 P12-1 8 0 8 0 42 F2 BC1 P12-2 18 1 19 5.26

Table 23. Percentage of pod damage in advanced generation lines of C. platycarpus

Number of HA Total %pod S. No Plant number healthy pods damage pods/plant damage 1 F1 BC5 A6-12 4 0 6 0 2 F1 BC5 A6-1 11 1 12 8.33 3 F1 BC5 D1-10-1 43 5 48 10.4 4 F1 BC5 D1-10-2 6 0 8 0 5 F1 BC5 D1-10-3 22 4 26 15.3 6 F3 BC3 D2 21 6 27 22.2 7 F1 BC5 E5-8-1 5 0 5 0 8 F1 BC5 E5-8-2 40 6 46 13.04 9 F1 BC5 E8-2 121 11 136 8.08 10 F2 BC5 B6-2 36 3 39 7.69 11 F3 BC3 C7-13-1 137 7 144 4.86 12 F3 BC3 C7-13-2 900 27 927 2.91 13 F3 BC3 C11-3-1 500 19 519 3.66 14 F3 BC3 C11-3-2 150 4 154 2.59

153

Table 24. Bruchid damage in the F1 hybrids of the cross C. cajan X C. lanceolatus and their parents (ICPL 85010 and ICP 15639)

No. of days Plant No. of No. of No. of % of Adult No. of taken for Number adults eggs adults hatched mortality seeds emergence (F s) released laid emerged eggs % 1 (range) P1 10 8 16 8 50 44-54 62.5 P2 10 8 38 12 31.5 41-56 16.7 P4 10 8 14 4 22.2 44-56 33.3 P5 10 8 25 5 20 47-53 60.0 P6 10 8 13 6 46 41-51 33.3 P7 10 8 9 7 77 44-54 42.9 P8 10 8 19 1 5.2 56 100 ICPW 15639 12 10 29 10 34.4 57-81 40.0 ICPL 85010 10 6 14 10 71.4 35-38 0

154

FIGURES

A B

Figure 1: (A) Male parent ICPW 29 (C. cajanifolius), (B) female parent ICPL 87119 (C. cajan).

155

Figure 2: (A) Variation in plant morphology in progeny lines of F3 BC176-2. (B) Variation in the leaf size and shape of F3 BC137-1 and (C) Uniformity in the morphology of progeny lines of F3 BC1129-1.

156

A B C A

Figure 3: (A) Small sized flowers of 82-2-7 F3 BC1. (B) Purple and green mixed small pods with prominent locules of 76-2-3 and (C) Green and purple coloured pods in the different plants of

130-1 F3 BC1 (single row) line.

Figure 4: Morphology of leaves in the F3 BC1 population of the cross C. cajan X C. cajanifolius. 157

B A

Figure 5: (A) F3 BC1 plantation of C. cajanifolius derivatives (B) interspecific derivatives in the field.

158

C D

Figure 6: Female parent (ICPL 85010), Male parent (ICP 15639) and tallest F1 hybrid P-13 measuring about 380cm. Inflorescence of the BC1 F1 hybrid with flowers resembling the wild parent and pods resembling the cultivar parent.

159

B C D

E G F

Figure 7: Morphological observations of hybrids (ICPL 85010 X ICPW15639) and parents (cultivar- ICPL 85010 and ICPW15639). (A) Comparison of hybrids (middle) with cultivar (female-left) and wild (male-right) parents. (B) Morphology of the leaves (left-cultivar, right- wild and hybrid -middle). (C) Morphology of flowers (left-wild, right- cultivar and hybrid - middle). (D) Anthers of cultivar (left), sterile hybrid (middle), partial fertile hybrid (middle) and fertile hybrid (right) and plants. (E) Pod formation in C .cajan. (F) Hybrid pod formation. (G) Pod formation in C. lanceolatus.

160

A B

Figure 8: (A) Morphology of pods and seeds in parents (cv. ICPL 85010- left and ICP 15639 - right), (B) hybrid (middle) plants.

Figure 9: Variation in plant height of C. lanceolatus derivatives.

161

Figure 10: Variation in the leaf size and shape of the progeny lines derived from the cross C. cajan X C. lanceolatus.

A B C

Figure 11: Receme morphology in F1 BC1 population of C. cajan X C. lanceolatus. (A) determinate growth habit. (B) semi-determinate growth habit. (C) indeterminate growth habit.

162

A B

C D

Figure 12: F1 BC1 plants segregated for flower colour and pattern of streaks on standard petal. (A) yellow flowers. (B) orange flowers with sparse streaks. (C) dense streaks and (D) flowers without streaks.

A B D

Figure 13: Variation in the pod colour and shape in the F2 and F1 BC1 pods. (A) green pods resembling the wild parent with dense hairs on pods of F2 P3-1. (B) green and purple mixed pods in F1 BC1 P-12-1. (C) complete purple pods in F1 BC1 P4-1. (D) normal green pods resembling cultivated parent in F1 BC1 P2-5.

163

Figure 14: F1 BC1 [(ICPL 85010 X ICP15639) X ICPL 85010] and F2 progeny plantation in the field.

 Derivatives of C. platycarpus.

A B

Figure15: Female parent C. platycarpus ICPW 68 (left) and male parent ICPL 85010 (right)

164

A B C D

E F G H

I J K L

M N O P

Figure 16: Morphology of the F1BC3 plants derived from C. platycarpus. (A-D) F1 BC3 A lines. (E and F) F1 BC3 B lines. (G and H) F1 BC3 C lines. (I and J) F1 BC3 D lines and (K-P) F1 BC3 E lines.

165

Figure 17: Variation in Anther bundles and pollen load in F1BC3 lines derived from C. platycarpus.

166

A B C D

E F G H

I J K L

M N O P

Figiure 18: Pods obtained by cross and self pollinations of F1BC3 lines derived from C. platycarpus. (A-D) pods from F1 BC3 A line. (E-H) pods from F1 BC3 B. (I-K) pods from F1 BC3 C. (D) pods from F1 BC3 D and (M-P) pods from F1 BC3 E line.

167

A B C

D E F

G H I

J K L

M N O

Figure 19: Seeds obtained from F1 BC3 lines derived from C. platycarpus were compared with cultivated parent ICPL85010. (A) Black seeds obtained from plant A5. (B) brown seeds of A8. (C) light brown seeds obtained from A9. (D) brown seeds of B2. (E) black seeds obtained from B6. (F) brown seeds with elongate shape of C3. (G) Gray coloured seeds of C6. (H) brown seeds with elongate shape of C7. (I) brown seeds of D1. (J, K and N) black seeds of E line. (L and M) gray seeds of E line. (O) nonviable seeds obtained from all cross pods.

168

A B C

D E F

Figure 20: Morphology of F1 BC3 B6 plant and their pods. (A) F1BC3 B6 plant with trailing habit. (B) F1 BC4 B6 with erect habit. (C) F1 BC5 B6-1 plant with normal growth habit. (D) small flattened pods of F1 BC3 B6. (E) green and purple mixed pods of F1 BC4 B6. (F) F1 BC5 B6-1 with black/purple pods.

Figure 21: Variation in leaf let shape and size in the C. platycarpus derivatives.

169

B C

D E

Figure 22: Morphology of F1 BC5 lines of the cross C. platycarpus X C. cajan in the field. (A) medium sized sterile F1 BC5 A5-7 line crossed with different cultivars. (B) sterile F1 BC5 B3-2 line. (C) F3 BC3 C7-13-1 with spreading habit. (D) F3 BC3 C11-3-1 plant with spreading habit and more pods. (E) flowers of F3 BC3 C11-3-1 with dense streaks on the standard petal.

170

A B

Figure 23: Morphology of parents involved in the wide cross. Female parent ICPL 85010 (left),

F1 hybrid (middle) and male parent ICP 15774 (right) (A), Stigma of cultivated parent with germinated pollen of wild parent (B and C)

171

A B

C D

Figure 24: Frequency polygon graphs showing Days to first flowering (A), Plant height (B),

Days to 50% pod maturity (C) and Number of pods/plant (D). F2 plants data with line plot, Female (ICPL 85010) and male (ICPW15774) parents along with one extra short duration pigeonpea MN- 5 used as a check represented in coloured bars

172

A B C

Figure 25: Morphology of hybrids in C. volubilis derivatives. (A) F1 hybrid. (B) F2 hybrid plant and (C) F4 hybrid plant.

C A B

Figure 26: Colour and pattern flowering in F4 population of C. cajan X C. volubilis. (A) pale yellow coloured flowers in determinate inflorescence. (B) yellow coloured flowers (C) pods per inflorescence.

173

A B

C D

Figure 27: Comparisons of F2 and F4 hybrid plants derived from the C. volubilis with different cultivars. (A) ICPL 85010 and F2 hybrid plants, both were 60 days old plants. (B) 90 days old plants, ICPL 85010 (left) and F4 hybrid plant (right). (C) F4 hybrid and determinate short duration cultivar MN5 and (D) F4 hybrid and another determinate short duration cultivar MN8.

174

Figure 28: F4 plantation of the C. volubilis derivatives in the glass house.

175

A B C

D E F

G H I

Figure 29: Meiotic analysis of F1 hybrid derived from C. cajan (ICPL 87119 ) X C. cajanifolius (ICPW 29). (A) Meiosity in diakinesis stage showing bivalents. (B) 7 bivalents and two tetravalents. (C) 11 bivalents at metaphase I. (D) anaphase I with 2 laggards. (E) unequal distribution of chromosomes at anaphase I. (F) anaphase II showing lagging chromosomes. (G) anaphase II showing only three poles. (H) tetrads and (I) fertile and sterile pollen grains.

176

B A C

D E

F G H

I J K

Figure 30: Meiosis in BC1 F3 73-1-3 (A-E), 76-2-1(F-H) and 82-2-2 (I-K). (A) metaphase with two univalents. (B) anaphase I. (C) metaphase II with two metaplates. (D) meiocyte at diplotene, abnormal anaphase I and anaphase II. (E) abnormal anaphase I with lagging chromosomes and anaphase II with more number of chromosomes. (F, I and J) metaphase I. (G) lagging chromosomes in anaphase I. (H) anaphase II and (K) anaphase II with laggards.

177

11 10 9 8 7 6 5 Tetravalents Trivalents 4 Bivalents 3 Univalents 2 1

Mean chromosomal2n=11configuration Mean 0 P-1 P-2 P-3 P-4 P-5 P-6 P-7 P-8 P-9 P-10 P-11 P-12 P-13 P-14 (ICPL 85010 X ICP 15639 ) F1 Hybrids

100%

90%

80%

70%

60% abnormal distribution of 50% chromosomes at 40% anaphase I Percentage% 30% Normal 20% distribution of chromosomes at 10% anaphse I 0% P-1 P-2 P-3 P-4 P-5 P-6 P-7 P-8 P-9 P-10 P-11 P-12 P-13 P-14 ICPL 85010 X ICP 15639 F1 HYBRIDS

Figure 31: Summary charts (stacked columns) for the meiotic analysis of C. lanceolatus F1 hybrids. Metaphase analysis (1st bar chart) and anaphase analysis (2nd bar chart)

178

A B C D

E F G H

I J K L

M N O P

Q R S T

U V W X

Figure 32: Meiotic analysis in F1 hybrids derived from C. lanceolatus. (A, I, J and Q) metaphase I with ring and rod bivalents. (B, I and R) anaphase with lagging chromosomes. (C, K and S) anaphase I. (D and E) early anaphase II and anaphase II with laggards. (F, M and U) normal tetrads. (G) unseparated microspores. (H) sterile anther. (N and O) partial fertile and sterile anthers from P2. (V) separated microspores. (W) partial fertile anther in P3 and (X) Alexander’s stained pollen grains.

179

A B C D E

F G H I J

K L M N O

P Q R S T

Figure 33: Meiosis in F1 hybrids P4, P5, P6 and P7. (A, F, K and P) metaphase I with rong and rod bivalents. (B, C, G, I, Q and R) anaphase with lagging chromosomes. (D and M) normal tetrads. (G) unseparated microspores. (E and T) sterile anthers. (I and N) partial fertile and sterile anthers from P5 and P6. (S) tetrads with micronuclei. (J and O) Alexander’s stained pollen grains.

A B C D E

F G H I J

Figure 34: Meiosis in fertile hybrid F1 P8. (A) diakinesis showing nucleus in the meiocyte. (B) metaphase I. (C) early anaphase I. (D) lagging chromosomes at anaphase I. (F) early anaphase II. (G) laggards in anaphase II. (H) tetrads with micronuclei. (I) partially fertile anther and (J) acetocarmine stained pollen grains. 180

A B C D E

F G H I

J K L M

N O P Q

R S T U

V W X Y Z

Figure 35: Meiotic analysis in BC3 F1 A lines derived from C. platycarpus. (A) metaphase with 11 bivalents and (B) anaphase I in A1. (C) metaphase showing 11 rig bivalents, (D) early anaphase I. (E) anaphase I showing two laggards. (F) metaphase showing univalents. (G) early anaphase I. (H) anaphase I. (I) anaphase II. (J) diakinesis. (K) metaphase. (L) anaphase I showing two laggards. (M) telophase I with two poles. (N, R) metaphase with bivalents. (O, S) anaphase I. (P, T) telophase I. (Q, U) telophase II. (V) metaphase with univalents. (W) early anaphase I. (X) anaphase I. (Y) metaphase with bivalents. (Z) anaphase I with unequal distribution.

181

A B C D E

F G H I

I K L M N

Figure 36: Meiotic analysis in BC3 F1 B lines derived from C. platycarpus. (A) metaphase with 11 bivalents. (B) anaphase I . (C) metaphase showing 11 rig bivalents, (D) early anaphase I & (E) anaphase I showing four laggards. (F) metaphase showing11 bivalents. (G, H) early anaphase I. (I) anaphase I showing three laggards. (J) metaphase showing ring and rod bivalents. (K) anaphase I showing one laggard. (L) metaphase with bivalents. (M) early anaphase I. (N) anaphase I with unequal distribution.

182

A B C D E F

G H I J K

L M N O P

Q R S T U V

B C F A D E

Figure 37: Meiotic analysis in BC3 F1 C and D lines derived from C. platycarpus. (A) Diakinesis in C1. (B) metaphase with bivalents. (C) anaphase I, (D) diakinesis in C2. (E) metaphase with ring and rod bivalents. (F) anaphase I with unequal distribution. (G) metaphase in C4. (H) early anaphase I. (I) anaphase I. (J) metaphase in C5. (K) anaphase I. (L) diakinesis in C9. (M) metaphase with bivalents. (N) early anaphase I. (O) anaphase I with laggards. (P) normal anaphase I. (Q) metaphase in C10. (R) anaphase I with bridges and fragments. (S) telophase I. (T) metaphase in C12. (U) early anaphase I. (V) anaphase I with unequal distribution. (A) metaphase in D1 (B) anaphase I with laggards. (C) metaphase showing bivalents in D2. (D) normal anaphase I. (E) metaphase with ring and rod bivalents in D3. (F) anaphase I with unequal distribution.

183

184

Figure 38: Meiotic analysis in BC3 F1 E lines derived from C. platycarpus. (A) metaphase in E1. (B) anaphase I, (C) metaphase in E2. (D) early anaphase I. (E) anaphase I with one laggard. (F) metaphase in E3. (G) anaphase I. (H) metaphase in E4. (I) anapahase I with unequal distribution. (J) normal anaphase I. (K) metaphase in E5. (L) anaphase I. (M) metaphase in E6. (N) early anaphase I. (O) normal anaphase I. (P) metaphase in E7. (Q) early anaphase I. (R) anaphase I. (S) early metaphase in E8. (T) metaphase I. (U) anaphase I with unequal distribution. (V, W) metaphase in E9. (X, Y) early anaphase I. (Z) anaphase I with laggards and normal anaphase I. (AA) metaphase I in E10. (AB) early anaphase I. (AC) anaphase I with equal distribution. (AD, AE) metaphase in E12. (AF) anaphase I with laggards. (AG, AH, AI) metaphase in E13. (AJ) anaphase I. (AK) anaphase II. (AL) telophase II. (AM, AN & AO) metaphase I in E14. (AP) early anaphase I. (AQ) anaphase I with laggards. (AR) normal anaphase I.

185

A B C

D E

F G

H I

K J

Figure 39: Tetrad analysis in BC3 F1 population derived from C. platycarpus. (A and B) tetrads in BC3 F1 A1. (A) normal terads each with 4 micronuclei. (B) tetrad with one micronuclei. (C to F) tetrads in BC3 F1 B1 (C) hexad showing 6 micronuclei. (D) tetrads with one and two micronuclei. (E) diad with two micronuclei. (F) normal tetrad and tetrad with one micronuclei.

(G to I) tetrads in BC3 F1 B4. (G) diads and tetrads. (H) diads, triads and tetrads. (I) triads. (J and K) tetrads in BC3 F1 E7. (J) diads and tetrads (K) tetrad with one micronuclei

186

Figure 40: Pollen fertility in F1 BC3 A, F1 BC3 B, F1 BC3 C, F1 BC3 D and F1 BC3 E lines derived from C. platycarpus

187

Figure 41: Anther morphology in F1 BC3 A, F1 BC3 B and F1 BC3 C lines derived from C. platycarpus.

188

Figure 42: Anther morphology in cv.ICPL 85010, F1 BC3 D and F1 BC3 E lines derived from C. platycarpus.

189

A B C

E F D

H J G

K Figure 43: Meiotic behavior in F1 hybrid derived from C.cajanx C. volubilis. (A) early metaphase. (B) metaphaseI. (C) metaphase with univalents. (D) anaphase I (E) bridges/fragments at anaphase I. (F) lagging fragments. (G) univalents. (H) univalents moved towards poles earlier than bivalents early at anaphase I. (I) tetrads and (K) pollen.

190

A B C

D E F

G H I

Figure 44: Abnormal meiosis in F4 P1-6-9 derived from C. volubilis. (A) meiocyte at diakinesis stageshowing tetravalents and bivalents. (B and C) metaphase with ring bivalents. (D, E and F) multinucleate condition in meiocytes. (G) multinucleate meiocyte with chrosomal fragments. (H) anaphase I with fragments and lagging chromosomes and (I) anaphase II with 2-3 nuclei at each pole and anaphase I with laggards.

191

B A

C D E

Figure 45: Meiosis in F4P1-3-8 derived from C. volubilis. (A) diakinesis and metaphase I with bivalents. (B) metaphase I with tetravalents and univalents. (C) 11 bivalents at early metaphase I. (D) chromosome fragments at early anaphase I and E late anaphase I showing nuclei at two poles.

A B C D

F G H E Fig 46: Meiosis in (ICPL 85010 X ICP15639) P12 F1 X MN1 male sterile hybrid. (A) Normal metaphase with 11 pairs if chromosome. (B) Metaphase with 2 univalents, 8 ring and 2 rod bivalents. (C) Anaphse I with one laggard. (D) Normal tetrads. (E) Intact microspores after tetrad stage. (F) Alexander stained sterile (green colour) anther at premature anthesis stage. (G) degenerated microspores inside anther wall at anthesis stage and (H) Sterile anther bundle devoid of functional pollen.

192

A B K

C D L

N O

F E

G P

H Q

I R

Figure 47: Microtomy sections of flower buds from Fertile ((ICPL 85010 X ICP15639) P-12 X MN1) and sterile ((ICPL 85010 X ICP15639) P-12 X MN1) plants. (A to I) fertile anthers sections. (A) longitudinal section (L.S) of fertile bud at premature anthesis stage. (B). transverse section (T.S) of fertile anther. (C) sporangium with tetrads. (D) anther wall layers showing epidermis, tapetu and inner wall layer. (E) tapetum around PMC. (F) disintegrating tapetal layer at anthesis stage. (G) L.S of anther showing mature pollen. (H) T.S of anther showing fertile pollen inside microsporangium. (I) The sacs of anther opened in staminate stage. (J to R) sterile anther sections. (J) L.S of sterile bud at premature anthesis stage. (K) T.S of sterile anther. (L) sterile anther lobe showing normal tetrads. (M) sterile anther lobe showing thick tapetal layer. (N, O) showing thick prominent tapetum even at late tetrad stage. (P) L.S of sterile anther at staminate stage showing intact tetrads. (Q) T.S of sterile anther. (R) the sterile anther shriveled and indehiscent at pistillate stage

193

Figure 48: Primer CcM0047, Electropherogram of SSRs obtained with software Genemapper.

Female parent ICPL 85010, Male parent ICP15639 and F1 hybrids

194

Figure 49: Primer CcM 0057, Electropherogram of SSRs obtained with software Genemapper.