Volume 2 (1), March 2001

Edited by Colin Hanbury, CLIMA, Australia

Jointly supported by Lathyrus Lathyrism Newsletter 2 (2001)

CONTENTS

Page Editor's Comment 1 Colin Hanbury -Australia Articles Opinion and Future Directions 2 Do we need more research on neurolathyrism? S.L.N. Rao -India 5 Vapniarca revisited: Lessons from an inhuman human Fernand Lambein, Delphin Diasolua Ngudi, experience. Yu-Haey Kuo -Belgium

Food Processing 8 Fermentation of teff (Eragrostis tef), grass-pea (Lathyrus Y. Yigzaw, L. Gorton, G. Akalu, T. Solomon sativus), and their mixtures: Aspects of nutrition and food –Sweden and Ethiopia safety.

Neurology 11 Similarities between Tropical Spastic Paraparesis (TSP) Vladimir Zanonivic -Colombia and neurolathyrism

Plant Genetic Resources, Evaluation and Breeding 15 Electrophoretic phenotypes of different enzymes in some E. Alba, G. B. Polignano, D. De Carlo, A. entries of Lathyrus sativus L. Mincione -Italy 21 Autogamy and allogamy in genus Lathyrus. N. Ben Brahim, D. Combes, M. Marrakchi – Tunisia and France 27 Evaluation of selected traits in grasspea (Lathyrus sativus Michaela Benková, Mária Žáková –Slovak L.) genetic resources. Republic 31 Morphological characterisation of Spanish genetic L. De la Rosa, I. Martín -Spain resources of Lathyrus sativus L. 35 Protoplast, cell and tissue cultures for the biotechnological S. Ochatt, P. Durieu, L. Jacas, C. Pontécaille breeding of grass pea (Lathyrus sativus L.) -France 39 Progress in isolation and purification of Lathyrus sativus M. Anisur Rahman, M. Matiur Rahman, M. breeding lines Akhtaruzzaman Sarkar -Bangladesh 41 Mutants of grasspea (Lathyrus sativus L.) obtained after Wojciech Rybinski -Poland use of chemomutagens- Abstract 42 Development of low ODAP somaclones of Lathyrus I.M. Santha, S.L. Mehta -India sativus.- Abstract. 43 Isozymes variability of grasspea (Lathyrus sativus L.) in Wuletaw Tadesse, Endashaw Bekele - Ethiopia. Ethiopia Lathyrus Lathyrism Newsletter 2 (2001)

Agronomy 47 Effect of foliar spray of potassium nitrate and calcium R.K. Sarkar, G.C. Malik -India nitrate on grasspea (Lathyrus sativus L.) grown in rice fallows.

Animal Feeding 49 The nutritional value of Lathyrus cicera and Lupinus Colin White, Colin Hanbury, Kadambot angustifolius grain for sheep- Summary Siddique -Australia

51 Recent Publications

53 Items of Interest

The Lathyrus Lathyrism Newsletter can be obtained on-line at http://go.to/lathyrus OR http://www.clima.uwa.edu.au/lathyrus

All research articles are provided there in pdf format.

Jointly supported by:

Centre for Legumes in Mediterranean Agriculture (CLIMA), University of Western Australia, 35 Stirling Highway Crawley 6009, Australia http://www.clima.uwa.edu.au

and

Third World Medical Research Foundation (TWMRF), PO Box 9171, Portland, Oregon 97207, USA http://www.twmrf.org Lathyrus Lathyrism Newsletter 2 (2001)

EDITOR’S COMMENT

Welcome to the second edition of the revived Any comments on improving the newsletter in the Lathyrus Lathyrism Newsletter. Thank you to all the future or any other suggestions are very welcome, authors who have contributed their work. This contributions from readers are imperative to volume will be in two parts, this edition is the first maintaining the direction and momentum of the [Vol. 2(1)], the second part [Vol. 2(2)] will follow in newsletter. Most research submissions should be September/October 2001. If you wish to contribute an approximately 1500 words and can include a small article for the coming edition it will be still possible number of tables or figures, photographs of the to include some that arrive by 15 September 2001. I author and of research are encouraged as they are will however be away from 23 July until 14 easily included in the on-line version. September. Please also note that my Email address Introduction/Methods/Results and Discussion are the will be changing shortly, although I hope to continue preferred layout for research summaries, although with both old and new addresses for some time. I do this can be altered as necessary. Abstracts are also not know the new Email address at this time. welcome, if they have been published elsewhere then full acknowledgment will be made. Electronic copies One of the aims of the newsletter has been to increase are preferred but not essential. cooperation across disciplines and distance and this seems to be occurring already. I for one have a greater understanding of what has been recently achieved in areas other than my own. Authors Colin Hanbury working in all disciplines are encouraged to submit work relevant to Lathyrus and lathyrism.

Please see the section “Items of Interest” (page 53) on the Lathyrus Email group. If you have Email access this will provide you with ready Editor contact details: communication to so far 40 other researchers Dr Colin Hanbury worldwide, all interested people are welcome to join. Centre for Legumes in Mediterranean Agriculture (CLIMA) Thanks is due to the Centre for Legumes in University of Western Australia Mediterranean Agriculture (CLIMA) and the Third 35 Stirling Highway, Nedlands 6009 World Medical Research Foundation (TWMRF) for Australia supporting the newsletter. Email: [email protected]

______

1 Lathyrus Lathyrism Newsletter 2 (2001)

Do we need more research on than ODAP (6). Even if one were to consider agonistic neurolathyrism? activity of ODAP at glutamate receptors as a plausible mechanism of toxicity, it simply cannot explain the S.L.N. Rao extreme species differences in susceptibility to ODAP. Thus can the adult rat and the BALB/c mice which are Department of Biochemistry resistant to ODAP possess a different disposition of Osmania University, Hyderabad-500 007, India glutamate receptor subpopulations than the susceptible species the chick? Vexed by this inconsistency, we Email: [email protected] came up with the most surprising finding that the C57BL6J black mice are easily susceptible to ODAP It is nearly four decades since the discovery of ODAP while the BALB/c white mice are resistant to it but as the major neurotoxic constituent of Lathyrus sativus would become susceptible if pretreated with tyrosine seeds. However, several basic questions relating to (3). This, further led us to establish the first ever neurolathyrism still remain unanswered. We are stereospecific inhibition of an enzyme activity by unable to convince several governments to ban the ODAP, namely, that of tyrosine aminotransferase. cultivation of the pulse and to take on a more serious Over the years we have been so enamoured by the approach to introducing the low toxin varieties. "structural relatedness" of ODAP with glutamate that Meanwhile, farmers continue to grow the traditional no one dreamt that it may inhibit the aromatic amino varieties and the consumption of the pulse continues. acid transferase. Our molecular modelling studies with We still do not have a straightforward method to ODAP (to be published) show that ODAP is indeed produce a convincing animal model to study the conformationally cognate more with tyrosine and less disease. We do not even have an acceptable with glutamate (Fig. 1). We have shown that ODAP mechanism for the toxicity of ODAP. While we are administration results in a significant increase in brain concerned that the disease affects the poorest of the dopa and dopamine levels, only in black mice, as a poor, we do not know how the majority of the fallout of TAT inhibition and not in white mice. Lathyrus consuming population escape from the Further studies are needed to pinpoint the toxic disease. The precise enzymatic pathway and its metabolites of dopa and dopamine that may be the real characterisation for the biosynthesis of ODAP are still culprits of neurotoxicity as a result of oxidative needed before genetic engineering principles can be damage (3). applied to produce genetically modified ODAP free Lathyrus. It is rather intriguing how these seemingly One of the more puzzling features of neurolathyrism simple questions relating to a human disease of relates to how and why a large majority of the nutritional origin remain unsolved. I am certain that Lathyrus consuming subjects escape from the disease, there are reasons that most Lathyrus researchers would which is quite evident even in the recent Ethiopian agree upon. epidemic. To my knowledge, no one has attempted to explain this feature although some newer thoughts ODAP is deceptively a dicarboxylic amino acid appear to be emerging in evaluating the incidence of chemically related to glutamate (aspartate). However, the disease. Since most laboratory animal studies, to date, no direct effect of ODAP on any of the including those with monkeys, show that orally enzymes associated with the metabolism of either administered ODAP is excreted in the urine largely glutamate or aspartate has been demonstrated and also unchanged, it has generally been assumed that ODAP it does not have any substrate activity with any of would have a similar fate in humans. I would like to these enzymes. A number of studies have shown that share some of our recent yet unpublished findings on ODAP acts as an agonist at the AMPA preferring this aspect. In a survey of a Lathyrus consuming glutamate receptors (5,7). But in spite of our best population, the overnight urine samples collected efforts, we have been unable to demonstrate any following a Lathyrus meal showed very little urinary significant specific binding of 3H ODAP to synaptic excretion of ODAP (less than 2-5%), this was membranes of the chick or rat (1). This crucial evidence confirmed on several occasions. We have confirmed thus fails to establish antagonistic property for ODAP this finding in controlled studies with human at glutamate receptors. Moreover, we have now shown volunteers wherein the ODAP excretion was less than that even free diaminopropionate (DAP) and 1% of the dietarily consumed ODAP. This suggests carboxymethyl DAP (CMDAP) another synthetic that, as opposed to the findings in experimental glutamate/ODAP analogue which are non-neurotoxic animals, humans have an ability to (acute study) exhibit a far greater excitotoxic potential metabolise/detoxify orally ingested ODAP. While we

2 Lathyrus Lathyrism Newsletter 2 (2001)

Fig. 1: Energy minimised structures of tyrosine, ODAP and glutamate.

are yet to establish the pathway in humans, our studies References with chicks, rats and mice indicate that orally 1. Jain RK, Jumaid MA, Rao SLN. (1998). administered 14C ODAP does undergo some oxidation, Receptor interactions of ODAP, the Lathyrus albeit to a limited extent, resulting in the appearance sativus putative excitotoxin with synaptic 14 of CO2 in the expired air. It is likely that in humans a membranes. Neurochem. Res. 23, 1191-1196. similar pathway may operate to a greater extent and 2. Jyothi P, Pratap Rudra M, Rao SLN. (1998). In vivo metabolism of ODAP the Lathyrus lead to near complete metabolism/detoxification of sativus neurotoxin in experimental animals. ODAP (2). This finding opens up the interesting Nat. Toxins 6, 189-195. possibility that this pathway could be deficient in 3. Kalivendi SV, Pratap Rudra MP, Rao SLN. certain individuals or under certain conditions such (1997). Inhibition of tyrosine amino that these individuals may run the risk of the disease. transferase by ODAP the Lathyrus sativus Identifying such individuals and the metabolic/ neurotoxin. J. Neurochem. 68, 2477-2484. detoxification pathway should receive high priority in 4. Lambein F. (2000). Homeopathy, longevity future investigations. and Lathyrus sativus toxicity. Lathyrus Lathyrism Newsletter 1, 4-5. I am certain that readers of Lathyrus Lathyrism 5. Nunn PB, Seelig M, Zagoren JC, Spencer PS. Newsletter would agree with me that further research (1987). Uncommon plant amino acids linked on these lines would go a long way in clearing up to human motor system diseases. Brain Res. some of the grey areas of Lathyrus toxicity. Further, as 410, 375-379. pointed out by Dr.Fernand Lambein (4) in the previous 6. Omelchenko IA, Jain RK, Jumaid MA, Rao issue of this newsletter, there is a need for a SLN, Allen CN. (1999). Neurotoxic potential coordinated effort on the part of different Lathyrus of three structural analogues of ODAP. researchers in solving the mystery associated with this Neurochem. Res. 24, 791-797. pulse and the disease. 7. Ross SM, Roy DN, Spencer PS. (1989). Beta-N-oxalylamino-L-alanine on glutamate receptors. J. Neurochem. 53, 710-715.

3 Lathyrus Lathyrism Newsletter 2 (2001)

and diarrhoea that can easily be prevented by a better Following commentary by Fernand Lambein, balanced diet. Imposing a ban on grass pea in areas University of Gent, Belgium where it is an essential part of the survival economy (Email: [email protected]) may only solve the problem on paper while furthering the neglect of the victims and of research. On the Professor SLN Rao was at the cradle of Lathyrus other hand, lifting unconditionally an existing ban research with the publication of the structure and might only benefit grain traders and give totally wrong chemical properties of ODAP in 1964 (3). The ever signals to the consumers if no proper nutritional popular "Rao-method" for the quantitative analysis of education is in place. Until now, researchers did not ODAP is still an important tool for breeders and for voice a single and convincing opinion on how to other Lathyrus researchers (2). As Prof. Rao continues prevent lathyrism. with surprisingly innovative and thorough research, he can rightly be considered the godfather of Lathyrus During the last decade, Prof. Rao and others have researchers. Those who were present at the festivities added some exciting new features to the increasingly and the scientific symposium at the occasion of his complex biochemistry and physiology of this simple retirement witnessed the signs of extraordinary molecule ODAP, in the producing plant as well as in sympathy and respect he received from colleagues and the consuming human or animal. The remarkable students (4). We wholeheartedly welcome SLN Rao as variability of the plant and the remarkable variability contributor to the renovated Lathyrus Lathyrism in the susceptibility of species and individuals to this Newsletter. molecule make it increasingly difficult to draw the simple conclusions that are needed to educate the As a man of dialogue, he will understand that among people at risk and so prevent lathyrism. his students and friends different opinions can emerge. Or perhaps we use only different words to express the Undoubtedly, the environment affects the level of same conviction. Banning the sale of Lathyrus but not ODAP and other secondary metabolites in the plant. the consumption has been a controversial topic and Can the environment also affect human susceptibility? probably will remain as such, while propositions to lift Under the term environment we also need to consider the ban has achieved nothing but raising blood our internal environment or intestinal flora, which is pressures. In Europe, it took at least two generations much more variable than our genetic blueprint and for the ban on Lathyrus sativus to take effect (edicts in may be prone to dietary or environmental effects. We 1671, 1705 and 1714 for the same Wurtemberg area) very much hope that Prof. Rao can continue to unveil (1). Yet 300 years later consumption still persists and the intricacies of ODAP metabolism leading to the bloomed during the Second World War. After crop oxidative damage and probably apoptosis of some failures due to drought in Eastern Poland and specific neuronal cells. Only from a complete and Northwest China, grass pea was reintroduced recently. probably very complex picture, we will be able to Grass pea and lathyrism become an acute problem in derive the simple advice that can prevent lathyrism periods of drought-triggered famine, when other food and protect the consumers. grains are not available and grass pea seed forms a survival food from which up to 95% of the people References suffer no ill consequences. Considering this one 1. Loudon's Encyclopaedia of Plants (Mrs Loudon, should not be surprised that a ban on grass pea has no Ed.) Longman, Brown, Green and Longmans, immediate effect. London 1855, p 620. 2. Rao SLN (1978) A sensitive and specific As Prof. Rao points out the lathyrism victims are the colorimetric method for the determination of α,β- poorest of the poor. The depressing socio-economic diaminopropionic acid and the Lathyrus sativus environment where lathyrism thrived among neurotoxin, Anal. Biochem. 86, 386-395. subsistence farmers also promotes a low socio- 3. Rao SLN, Adiga PR, Sarma NPS (1964) The political status of the victims, which may be isolation and characterization of β-N-oxalyl-α,β- responsible for the under-financing of much diaminopropionic acids: a neurotoxin from the Lathyrus/lathyrism research. It is cynical that so much seeds of Lathyrus sativus. Biochem. 3, 432-436. research money is spent on diseases such as cancer 4. Subramanyam C, Singh SS, Eds (1997) Molecular and cardio-vascular diseases of which a sizable mechanisms of toxicity. Proceedings of an portion can be prevented by a poorer diet. Yet so little international meeting. Department of money is spared for diseases such as lathyrism, konzo Biochemistry, Hyderabad, India. 105 pp.

4 Lathyrus Lathyrism Newsletter 2 (2001)

Vapniarca revisited: Lessons from an prisoners, noted many details on the diet and inhuman human experience. neurological symptoms of the prisoners and the lathyrism victims. This was published in the German Fernand Lambein, Delphin Diasolua Ngudi, language in 1947. Additional information on the camp Yu-Haey Kuo conditions can be found on the web:

Lab. Physiological Chemistry, Faculty of Medicine http://motlc.wiesenthal.com/pages/t081/t08153.html and Health Sciences, Ghent University, Kluyskensstraat 27, B-9000 Ghent, Belgium. What is termed in the website as "horse fodder" consisted of 200 gram boiled grass pea seed and 200 Email: (1) [email protected] gram barley bread containing 20 % chopped straw. This was given to 150 inmates who were moved from Bukovina to Vapniarca 3 to 6 months before the arrival of 1200 inmates on 16 September 1942. Of these, 680 came from other camps, while the others were recent In the previous issue of this Newsletter, a contribution arrests. At that time, the diet of all inmates was on "Homeopathy, longevity, and Lathyrus sativus changed to 400 gram boiled grass pea seed per day and toxicity"(5), indicated that the problem of the economic the same 200 gram barley bread. value of this crop and the medical consequences of its over-consumption involved nutritional aspects that need In November 1942 one of the earlier inmates, who had better consideration. To better understand the essence been in the camp for five months (three months on a of the problem, and to further this discussion we diet of 200 gram grass pea and 160 gram barley, and revisited some older literature focusing on observations two months on a diet of 400 gram grass pea and 160 or statements concerning human nutrition. gram barley), developed symptoms of lathyrism. Three more cases developed in December 1942, and on In the old Loudon's "Encyclopaedia of Plants" re-edited December 29, the first of the group arrivals of in 1855(7), the well-known edict of the Duke of September 16 developed the symptoms. One week later Wurtenberg of 1671 forbidding the use of grass pea in on January 5, already 19 lathyrism cases were apparent. bread is mentioned. It also adds: "and this not being On January 23, the consumption of grass pea was observed, was enforced by two other edicts under his halted. New cases continued to become apparent until successor Leopold in 1705 and 1714", underlining the March 1, with a peak on January 20, 1943. Together, popularity of this food in Europe three hundred years 60% of the inmates had developed various levels of ago. And further a nutritional observation: "Mixed with neurolathyrism, the highest incidence ever reported in wheat flour in half the quantity, it makes a very good the literature. A number of these patients are now under bread, that appears to be harmless. But bread made with the care of Dr Dan Cohn in Tel Aviv(9). this flour only has brought on a most surprising rigidity of the limbs in those who have used it for a continuance; Like Loudon before, Kessler also mentioned the effect insomuch that the exterior muscles could not by any of the environment on the toxicity of the seed. means be reduced, or have their natural action restored". Information on the atmospheric conditions of the time Loudon also made allusion to the effect of the might give some information on the potential content of environment and the soil on the variable toxicity of the ODAP in the grass pea seed. We can conservatively seed. estimate that the content was about 0.25 % (seeds recently harvested in Poland have a range of 0.03 to The most remarkable account comes from the 0.33 % ODAP, the lower values apparently for selected description of the epidemic in the German prisoners of lines; A. Winiarska, pers. comm.). This means that a war camp in Vapniarca, a town in Ukraine occupied by daily intake of 500 mg of ODAP per day under severe Germany during the second world war. There the conditions of malnourishment and physical exhaustion highest incidence of neurolathyrism on record did not cause apparent signs and symptoms of developed between December 1942 and the end of neurolathyrism in those inmates during a period of 3 to February 1943(3). Dr A. Kessler, who was among the 5 months. After September 16, 1942, when the diet was

5 Lathyrus Lathyrism Newsletter 2 (2001) changed to 400 gram of grass pea seed and the same When comparing with the dietary aspects of konzo, a amount of the same bread, the inmates absorbed about 1 disease with similar socio-economic background as gram of ODAP per person per day. From that moment lathyrism and exactly the same final stage of spastic the latency period of 2 to 3 months of continued intake paraparesis of the legs(1), we can make a nutritional was still needed for apparent cases of neurolathyrism to comparison. Cassava, the dietary cause of konzo is occur. quantitatively a much poorer source of protein than grass pea seed, containing only 1 to 1.5 %. For the From this dramatic episode in history we may deduce detoxification of the cyanide intake (from the cassava), that a threshold level of ODAP intake may be close to sulphur atoms derived from sulphur amino acids are 500 milligram per day per adult person. This has to be needed and are no longer available to feed the taken with extreme caution, as some of those earlier biochemical pathways protecting against oxidative inmates worked outside during the day, and some had stress. Konzo is prevalent only in areas where little relatives nearby who could visit. In both these cases, other food is available and especially in periods of food small amounts of additional unspecified food could be scarcity when treatment to reduce cyanide by soaking is received and consumed, changing the balance of the insufficient. In areas where 25% of corn is added to the dietary intake, while the total intake of ODAP was daily diet of cassava, konzo is not prevalent. The amino probably unaffected. Our assumption that the seeds acid score of cassava is only 14.1 and increases to 48.7 consumed before and after September 16 were the same with addition of 25% corn. Still then, the daily also has no proof. An additional factor is the requirement for essential amino acids is not met when preparation of the seeds: if the water after boiling was 500 g of this mixture is consumed per day. Less than discarded this would give an important reduction of the 40% of the daily requirement of cysteine plus toxin ingested. methionine is supplied by this diet; the lysine supplied is only 65% of the daily requirement and tryptophan We calculated the amino acid score, that gives a only 28%. This mixed diet has a higher amino acid measure of the balance of essential amino acids and for score, but is still not well balanced. Although both which the FAO gives the standard. For the diets as cassava and grass pea have an obvious deficiency in described, and not considering the unspecified additions essential amino acids, the rate of susceptibility to konzo mentioned above, we find an important decrease from between males and females is different from 61% in the earlier diet to less than 50% of the optimal neurolathyrism. Points of similarity may be found ratio in the diet after September 16, 1942. The amino further in the biochemical pathways and physiological acid score for pure grass pea seeds is around 20 %. The effects, where metabolites involved in the biosynthesis minimal requirement for essential amino acids was met of nitric oxide and its modulation can be derived from in both diets, except for sulphur amino acids. The diet the diet. Spencer and co-workers have studied the before 16 September had only 798 mg of cysteine plus effects of sulphur amino acid depletion on the methionine, when an adult male needs 1100 mg. The metabolism of cyanide(10). diet after 16 September had about 1067 mg of cysteine plus methionine. The important variation in the diet on While there is an obvious point to learn, that we should 16 September may have been in the ratio of ODAP to respect old literature, we should not overemphasise the essential amino acids. An important point is that the conclusions from such a one-sided approach of the quantitative daily requirement for essential amino acids problem. As mentioned in the contribution in the is considerably lower for women than for men. Perhaps previous Newsletter issue (1), many aspects need further there is possibly a simple explanation for the difference research, such as i) the micro-nutrients essential for our in susceptibility between males and females on the basis defence against oxidative stress(6), ii) the potential of this intake of essential amino acids. The bread depletion of these micro-nutrients through chelation mentioned in Loudon's Encyclopaedia containing half with ODAP, iii) the potential presence of minor toxins grass pea and half wheat and causing no deleterious that may have a synergistic physiological effect together effects, should have had an amino acid score and a ratio with ODAP, iv) the neurological and physiological of ODAP to essential amino acids close to the diet effects of ODAP other than the well established before 16 September. excitation of the AMPA receptors(2,4,8) and v) the effects

6 Lathyrus Lathyrism Newsletter 2 (2001) of ODAP and other metabolites such as homoarginine methylisoxazole-4-propionate on the excitatory on nitric oxide metabolism and apoptosis. amino acid and taurine levels in the brain of freely moving rats. Neurochem. Intern. 36, 523-530. The threshold level of safe intake of ODAP as deducted 5. Lambein F. (2000). Homeopathy, longevity and from the Kessler report is probably the only such Lathyrus sativus toxicity. Lathyrus Lathyrism information in the literature. Also this needs a careful Newsletter 1, 4-5. approach as the level of ODAP in the grass pea seed of 6. Lambein F, Haque R, Khan JK, Kebede N, Kuo YH. 1942 is unknown. Perhaps the archives of the (1994). From soil to Brain: zinc deficiency increases Vapniarca camp, or the survivors of this tragedy may the neurotoxicity of Lathyrus sativus and may affect shed further light on this inhuman human experience. the susceptibility for the motorneurone disease neurolathyrism. Toxicon 32, 461-466. 7. Loudon's Encyclopaedia of Plants (Mrs Loudon, Ed.) Longman, Brown, Green and Longmans, References London 1855, p 620. 8. Kusama T, Kusama-Eguchi K, Ikegami F, 1. Banea-Mayambu JP, Tylleskar T, Gitebo N, et al. Yamamoto A, Kuo YH, Lambein F, Watanabe K. (1997). Geographical and seasonal association (2000). Effects of β-ODAP, the Lathyrus sativus between linamarin and cyanide exposure from neurotoxin, and related compounds on cloned cassava and the upper motor neurone disease konzo glutamate receptors and transporters expressed in in former Zaire. Trop. Med Int Health 2, 1143-1151 Xenopus oocytes. Research Comm. Pharmacol 2. Jain RK, Junaid MA, Rao SLN. (1999). Receptor Toxicol. 5, 37-56. interactions of beta-N-oxalyl-L-alpha,beta- 9. Paleacu D, Cohn DF, Rabey JM (1999) Cognitive diaminopropionic acid, the Lathyrus sativus evaluation of patients with chronic neurolathyrism. putative excitotoxin, with synaptic membranes. Parkinsonism REL D. 5, 55-58. Neurochem Res 23, 1191-1196. 10. Tor-Agbidye J, Palmer VS, Lasarev MR, Craig 3. Kessler A. (1947). Lathyrismus. Psychiatrie und AM, Blythe LL, Sabri MI, Spencer PS (1999) Neurologie 113, 345-376. Bioactivation of cyanide to cyanate in sulfur amino 4. La Bella V, Piccoli F. (2000). Differential effect of acid deficiency: Relevance to neurological disease β-N-oxalylamino-L-alanine, the Lathyrus sativus in humans subsisting on cassava. Toxicol Sci 50, neurotoxin and (+/-)-α-amino-3-hydroxy-5- 228-235.

7 Lathyrus Lathyrism Newsletter 2 (2001)

Fermentation of teff (Eragrostis tef), could also decrease certain antinutritional factors like phytates, protease inhibitors and flatulence factors. grass-pea (Lathyrus sativus), and their mixtures: Aspects of nutrition and food safety. Summary of results In the present study fermentation of pure teff, grass- pea and their mixtures, 9:1 and 8:2 (teff:grass pea) ° 1 1 2 have been done at two temperatures (room and 35 C) Y. Yigzaw , L. Gorton , G. Akalu and T. in duplicate using strains of Lactobacillus plantarum 3 Solomon for bacterial fermentation and both Aspergillus oryzae and Rhizopus oligosporus in succession for 1. Department of Analytical Chemistry, Lund fungal fermentation. In addition two methods the University, Lund, Sweden. natural or spontaneous and back-slopping (previously 2. Ethiopian Nutrition and Health Research fermented culture used as a starter culture) methods Institute, Addis Ababa, Ethiopia. of fermentation have been tried on the above four 3. Department of Chemistry, Addis Ababa substrate groups. Although mixing teff with grass pea University, Addis Ababa, Ethiopia. has not been part of the traditional practice for food preparation in Ethiopia, exploring the potential of Email: (1) [email protected] fermentation of their blend may be beneficial. One obvious reason is developing an affordable nourishing crop for the poorer section of the population. However, in dealing with grass pea containing food products, the safety level has to be Introduction ascertained. This was the essence of the present work Grass pea (Lathyrus sativus) is one of the important and we devoted much attention towards the food legumes in countries like Bangladesh, India, and development of analytical potential for an accurate Ethiopia. It has desirable agronomic characteristics, evaluation of the toxin level after processing. In notably the ability to resist adverse climatic designing the experiment for fermentation we did not conditions. Nutritionally it is tasty and protein-rich. want to go higher than 8:2 (teff : grass pea) ratio as a However over consumption can cause an upper motor compromise between nutritional adequacy and neuron disease known as neurolathyrism, an sensory value. irreversible paraparesis of the lower limbs. The cereal grain teff (Eragrostis tef (Zucc.) Trotter) is one of the The pH and essential amino acid profiles of the major cereal crops of Ethiopia, where it is believed to different fermentation processes were compared. The have originated. Teff provides over two-thirds of the toxin levels of both raw and fermented grass pea human nutrition in Ethiopia (3), with a grain protein samples were also determined. The back-slopping content (10-12%) similar to other cereals. Besides with teff at a temperature of 35°C gave the sharpest providing protein and calories, teff is a good source pH drop. In all the cases, the fermentations done at of minerals, particularly iron. It has a very high elevated temperature showed a steeper slope in their calcium content and contains high levels of pH vs. time plot compared to their room temperature phosphorus, copper, aluminium, barium and counterparts. The spontaneous fermentation of pure thiamine. The principal use of teff grain for human teff and that using the pure culture of Lactobacillus food is the Ethiopian bread “injera”, a soft porous plantarum at room temperature showed similar pH thin pancake with a sour taste. profiles, the former being slightly lower in pH at any time during the fermentation period. Both of them The limited nutritional quality of cereal grains is due were slower fermentation processes as shown up in to their lower contents of proteins, fats, minerals, and their pH profile (pH vs. time plot). The fermentation vitamins compared with animal foods. It is an of pure grass pea with Lactobacillus plantarum at important fact that the amino acid composition of room temperature was the slowest among the legume and grain seeds, both of which exhibit substrate groups considered. One possible deficiencies of some of the essential amino acids for explanation for this could be the high protein content the human diet, are complementary. For example, of grass pea, which renders a higher buffering cereal grains tend to be deficient in lysine whereas capacity and hence slower change in pH. In a legume seeds are rich in this amino acid. The result is preliminary fungal fermentation experiment done, the that blending of the two types of seed makes a trend in pH change was a progressive increase to nutritionally better food than does either alone. more than 7. Such a trend is a common phenomenon Fermentation of cereals and their blend with legumes in tempe fermentation, whereby tender-cooked soy is a potentially important processing method that can bean or other legumes and legume/cereal mixtures be expected to improve the nutritive value such as are bound together in a white cake by mycelium of availability of proteins and amino acid profile. It

8 Lathyrus Lathyrism Newsletter 2 (2001) the mould Rhizopus spp. It is already documented in electrochemical detection is based on crosslinking literature that there is also a principle related to food horseradish peroxidase (HRP) and an Os-containing safety in such high alkaline fermentations. A mediating polymer as an inner layer and combination of high pH, free ammonia and rapid immobilizing L-glutamate oxidase (GlOx) as an outer growth of the essential proteolytic microorganisms at layer on top of a graphite electrode. GlOx has a high relatively higher temperature, make it very difficult activity primarily for L-glutamate but also for β- for other spoiling microorganisms to grow. In general ODAP and aspartate. In the catalytic reaction the a quick change in pH is a desirable phenomenon as it substrate is oxidized and the enzyme cofactor normally gives less chance for infecting reduced. The natural reoxidizing agent is molecular microorganisms to compete with the desirable ones in oxygen, which in the reoxidation reaction of the a given setting. enzyme is reduced to hydrogen peroxide.

The fungal fermentation improved the amino acid To entrap the enzymes in the hydrogel and to prevent profile for the essential amino acids in all the leakage of the mediator, a short chain polymer, poly substrate groups studied except for pure grass-pea ethylene glycol (400) diglycidyl ether (PEGDGE) substrate. Fermentation of teff:grass pea (8:2) , in was used as a cross-linking agent. Addition of particular has been found to be quite comparable in polyethylenimine (PEI) in the hydrogel is believed to essential amino acid profile to an ideal reference have sensitivity and stability enhancing effect on the protein recommended for children of 2 – 5 years old. biosensor. The double layer approach for the It is important to note that the complementary effect construction of the bienzyme sensor that avoids direct of mixing cereals and legumes, which improves the electrical wiring of GlOx has resulted in a high inherent nutritional deficiencies of cereals and sensitivity of 4.6 mA/M.cm2 and 14 mA/M.cm2 for legumes alone, may have a contribution in the amino β-ODAP and L-glutamate, respectively. The system acid profile of the final product. On the other hand was quite linear over the range (1- 250 µM) for both the bacterial fermentations (i.e. the spontaneous, glutamate and β-ODAP, the slope over this range was back-slopping and those using Lactobacillus used to calculate sensitivities. Concentrations higher plantarum as inoculum) did not bring a net change in than these were not tried. The limits of detection for the essential amino acid profile. the chromatographic-biosensor system were found to be 2 and 0.7 µM for β-ODAP and L-glutamate, The solid state fungal fermentation of pure grass pea respectively. The refractive index detection on-line was carried out with strains of Rhizopus oligosporous with the biosensor enabled full control of the and Aspergillus oryzae in succession in that order at chromatographic system for the determination of the ° 35 C with autoclaving at the start and in between the total amount of ODAP, selectively the amount of β- ° (1) inoculations at 100 C for 10 min . The substrate ODAP and L-glutamate in raw and fermented was allowed to ferment for about 48 h with each β 7 samples containing grass pea. The non- -ODAP inoculation (10 spores/5 g fermenting sample). Grass selective spectrophotometric method of analysis pea grits were subject to fermentation without any developed by Rao (2) has also been employed on raw further sample pre-treatment, except adjusting the pH -1 grass pea sample collections for the purpose of to 4 with lactic acid (50 g.L ). This fermentation comparing results by the two entirely different β process reduced -ODAP level in grass pea on the methods of analysis. Because the Rao method uses average by 82% for the high toxin variety, and by up alkaline hydrolysis of ODAP to DAP, it does not to 97% for the low toxin variety. β-ODAP levels distinguish between the generally accepted non toxic were measured by a further optimised α- and the toxic β-form. In literature it has been chromatography-biosensor coupled analysis method reported that the occurrence of α-form in raw grass β described below. The -ODAP levels for the high pea does not exceed more than 5%. The Rao and low toxin varieties of raw grass pea used for the spectrophotometric method and the present method present fermentation were originally 0.76 % and 0.52 show an extraordinary degree of agreement as % dry weight basis, respectively. The samples were revealed by parallel “t” test (90 % confidence limit). collected in areas of Ethiopia where lathyrism is at a higher prevalence rate. The present analytical system has operational stability of more than 50 h and storage stability of the biosensor about 4 days (dry, 4°C). Analysis time per sample is 10 min after extraction of ODAP which is Analytical method completed within 90 min. (TRIS/KCl(10/10mM) The chromatographic method of analysis employed buffer, pH 7). In general a fast, convenient and refractive index in combination with bio- reliable analytical system is developed in the present electrochemical detection for the simultaneous study that could dependably measure the safety level determination of the total amount of ODAP, of processed foods containing grass pea. selectively the amount of β-ODAP, and free glutamate. The biosensor construction for the

9 Lathyrus Lathyrism Newsletter 2 (2001)

References 1. Kuo YH, Bau HM, Quemener B, Khan JK, Lambein F. (1995) Solid-state fermentation of Lathyrus sativus seeds using Aspergillus oryzae and Rhizopus oligosporus sp T-3 to eliminate the neurotoxin beta-ODAP without loss of nutritional value. J Sci Food Agric. 69, 81-89. 2. Rao SLN. (1978). A sensitive and specific colorimetric method for the determination of alpha,beta-ODAP, the Lathyrus sativus neurotoxin. Anal. Biochem. 86, 386-395. 3. Uraga K, Narasimha HV. (1997). Effect of natural fermentation on the HCl-extractability of minerals from Teff (Eragrostis teff). Bull. Chem. Soc. Ethiop. 11, 3-10.

10 Lathyrus Lathyrism Newsletter 2 (2001)

Similarities between Tropical Neurolathyrism Neurolathyrism is a neurologic disorder caused by Spastic Paraparesis (TSP) and excessive ingestion of Lathyrus species. Lathyrism has neurolathyrism been known since ancient times; epidemics have occurred in some regions, including Russia, southern Europe, the Middle East and India, particularly during times of famine. At these times increased consumption of Lathyrus sativus, L. cicera and Vicia sativa has been implicated. Horses, cattle, swine and birds have been Vladimir Zaninovic’ affected.

Emeritus Professor of Neurology, Universidad del Clinically, lathyrism often presents relatively rapidly Valle, Cali, Colombia after a prolonged period (months) of ingesting large amounts of the grain, often in the context of general malnutrition. Disease often commences with Email: [email protected] OR complaints of pain or cramps in the legs or in the [email protected] region of the lumbar spine. Lower extremity weakness and sphincter disfunction then develop, soon evolving into permanent spastic paraparesis. The cramping pains and the sphincter dysfunction usually subside when the intoxication ceases and spasticity develops7.

There are a few pathologic studies of lathyrism, but a report by Hirano et al.8 confirms earlier descriptions of bilateral atrophy in the distal pyramidal tract in the lumbar cord. Additionally, there have been Tropical Spastic Paraparesis morphologic descriptions of degenerative changes in Tropical Spastic Paraparesis (TSP) is a slowly the spinocerebellar tracts and dorsal columns. In progressive spastic paraparesis with an insidious onset concert, these data suggest a central nervous system in adulthood. It has been found all around the world (CNS) disease expressed most pronouncedly in the (except in the poles), mainly in tropical and subtropical distribution of the longest CNS fibres. regions. It affects more women than men (almost 2:1) of low socioeconomic classes. Konzo or Buka-Buka is an acute or chronic form of spastic paraparesis more common in tropical countries In 1985 French investigators1 discovered the than in temperate climates. Deficiencies or toxicities association of TSP with HTLV-I, a retrovirus due to primitive diets as well as infectious agents have discovered in USA in 19802. A few months after the been implicated. Konzo is similar to lathyrism but French publication in 1985 TSP was also associated differs from TSP/HAM and from lathyrism in its with HTLV-I in Colombia and Jamaica3. In 1986 abrupt onset, nonprogressive course. Normal magnetic Japanese researchers4 published the association of resonance imaging scans of brain and spinal cord in HTLV-I with a similar syndrome and named the severely affected patients provide evidence of selective disease as HAM (HTLV-I Associated Myelopathy). damage of the upper motor neurons. All Konzo patients 9 Since 1988 the World Health Organization (WHO) were seronegative to retroviruses . recommended the conciliatory name TSP/HAM to both syndromes5. Until the end of 1996 only 45% (1,261 of Neurolathyrism affects more men and is usually 2,811 cases) of TSPs were associated to HTLV-I6. That epidemic whilst TSP/HAM affects more women and it means that more (55%) of TSPs remained idiopathic or is usually endemic. These are the some of the most at least HTLV-I seronegative. important differences between two similar clinical and neuropathological syndromes.

From the clinical point of view both TSP/HAM and neurolathyrism depict a pyramidal syndrome affecting

11 Lathyrus Lathyrism Newsletter 2 (2001)

mainly the corticospinal pathways and in a lesser grade Furthermore, vascular changes are restricted to a the sensory and spinocerebellar pathways of the spinal proliferation on the adventitia and no vasculitic cord. changes such as necrosis of the vascular wall, endothelial proliferation, or ischemic lesions in the surrounding tissue were found10.

Neuropathology of TSP/HAM The clinicopathological studies10 show that TSP/HAM has a clearly defined pathological pattern. The clinical Neuropathology of neurolathyrism differences in TSP/HAM patients are related to the “Comprehension of lathyrism has been made more extension and severity of this pattern. The cases difficult by the absence of a complete showed a close clinicopathological correlation. All of neuropathological study using contemporary them had lesions in the axons and myelin of the histopathological techniques. Studies of the brain are pyramidal tract of the spinal cord. This followed an lacking, save for one case of a subject who developed ascendant pattern similar to some degenerative the disorder 31 years prior to death in whom loss and diseases, as in familiar spastic paraparesis, with marked shrinkage of pyramidal neurons in the upper part of the abnormalities in the lumbar and thoracic segments of precentral gyrus was noted. The balance of the spinal cord that became less severe as the tract neuropathological studies has focused on the spinal reached the cervical segments. Most cases had lesions cord, which shows predominantly distal symmetrical in the Goll’s tract distributed in a descendant pattern, degeneration of lateral and ventral corticospinal tracts, with maximal involvement in the cervical region and sometimes with distal degeneration of spinocerebellar becoming negligible towards the caudal regions. and gracile tracts. One Rumanian subject, who died with mildly atrophic leg muscles 32 years after the The spinal cord lesions of patients with TSP/HAM, development of moderately severe lathyrism, showed ascendant in the pyramidal tract and descendant in the distal axonal degeneration of the fasciculus gracilus posterior columns, have been interpreted as and spinocerebellar pathways and changes in (but not demyelinating, either as a primary cytotoxic disorder or loss of) lumbar anterior horn cells, including swelling, secondary to inflammatory or immunological diminished Nissl substance, and Hirano bodies. disorders. However, primary demyelinating diseases, either viral or inflammatory in origin, damage myelin There are at least two interpretations of these changes. in several areas, usually confluently and in a transverse One possibility is that lathyrism is a central distal fashion. Disorders that damage the myelin affect axonopathy dominated by degeneration of the longest different systems simultaneously, such as multiple corticospinal tracts, with lesser involvement of shorter sclerosis or multifocal leukoencephatopathy. Central pyramidal pathways serving the upper extremities of nervous system demyelinating diseases involve groups the most severely compromised individuals. This of oligodendrocytes, and each oligodentrocyte interpretation is consistent with the neurophysiological myelinates several axons independent of their and neuropathological findings, except perhaps the loss functions. However, in TSP/HAM lesions, the myelin of upper motor neurons”11. follows the axons in a dying-back fashion (axial) that especially affects the longest axons. The lesions in the posterior columns also support the idea of an axomyelinic degeneration. The lesions of Goll’s tract Conclusions in the cervical spinal cord are selective, affecting the TSP/HAM and neurolathyrism have clinical and longest axons from the legs. neuropathological similarities which suggest a common or similar cause, possible of toxometabolic origin. It seems unlikely that these parenchymal changes are secondary to vascular changes. Abnormal vessels with Cassava (manioc, mandioca) consumption, the gross thickening of the adventidia were seen in all etiological trigger of Konzo in the setting of minimal patients, many of them with lymphocytic cuffs, nutrition (sulphur deficiency) is rapidly expanding in especially in the spinal cord, brain stem, midbrain, the tropics and subtropics and appears to be a strong thalamus, and meninges, but were unrelated to the candidate for some TSPs in HTLV-I seronegative location or severity of the parenchymal damage. populations. There is a possible relation of atypical

12 Lathyrus Lathyrism Newsletter 2 (2001)

parkinsonism in the French West Indies with I, HTLV-I. WHO Weekly Epidemiology Record consumption of tropical plants like herbal tea and fruits 49, 382-83. from the Annonaceae family (Annona muricata and A. 6. Leon FE, Costa CM, Gaffga N. (1997). squamosa) containing alkaloids that have insecticidal Discrepancy, coincidence, or evidence in chronic activity12. This enhances the idea of chronic idiopathic spastic paraparesis throughout the neurotoxicity in some degenerative diseases of the world. A meta-analysis on 2811 patients. Arq central nervous system. Neuropsiquiatr 55, 530-35. 7. Zaninovic’ V. (1999). On the etiology of tropical spastic paraparesis and human T-cell lymphotropic virus I-associated myelopathy. Int J Infect Dis 3, References 168-76. 1. Gessain A, Barin F, Vernant JC, et al. (1985). 8. Hirano A, Llena JF, Streifler M, Cohn DF. (1976). Antibodies to human T-lymphotropic virus type-I Anterior horn changes in a case of neurolathyrism. in patients with tropical spastic paraparesis. Lancet Acta Neuropathol (Berl) 35, 277-83. ii, 407-10. 9. Tylleskar T, Legue FD, Peterson S, Kpizingui E, 2. Poiesz BJ, Ruscetti WF, Gadzar AF, Bunn PA, Stecker P. (1994). Konzo in the Central African Minna D, Gallo RC. (1980). Detection and Repubic. Neurology 44, 959-61. isolation of type C retrovirus particles from fresh 10. Cartier LM, Cea JG, Vergara C, Araya F, Born P. and cultured lymphocytes of a patient with (1997). Clinical and neuropathological study of six cutaneous T-cell lymphoma. Proc Natl Acad Sci patients with spastic paraparesis associated with USA 77, 7415-19. HTLV-I: an axomyelinic degeneration of the 3. Rodgers-Johnson P, Gajdusek DC, Morgan OS, central nervous system. J Neuropathol Exp Neurol Zaninovic’ V, Sarin PS, Graham DS. (1985). 56, 403-13. HTLV-I and HTLV-2 antibodies and tropical 11. Spencer PS. (1995). Lathyrism. Intoxications of spastic paraparesis. Lancet ii, 1247-48. the nervous system. Part II. Handbook Clin 4. Osame M, Usuku K, Izumo S, et al. (1986). Neurol 21, 1-20. HTLV-I associated myelopathy: a new clinical 12. Caparros-Lefebvre D, Elbaz A. (1999). Caribbean entity. Lancet ii,1031-32. Parkinsonism Study Group. Possible relation of 5. World Health Organization (1988). Report of the atypical parkinsonism in the French West Indies Scientific Group on HTLV-I and Associated with consumption of tropical plants: a case-control Diseases. Kagoshima, Japan, December 1988. study. Lancet 354, 281-86. Virus diseases. Human T-lymphotropic virus type

13 Lathyrus Lathyrism Newsletter 2 (2001)

may progress with continued consumption of cassava and grass pea, respectively, but in their absence the Following commentary by Peter S. Spencer, diseases are stable for decades. By contrast, clinical Professor of Neurology, Center for Research on advancement is the norm for TSP/HAM, and the Occupational and Environmental Toxicology; neurological picture may include features (sensory, Oregon Health & Science University, Portland, lower motor neuron and cognitive deficits) that are Oregon 97201, USA absent in Konzo and lathyrism. (Email: [email protected]) While these facts are solid, there are significant uncertainties in regard to the clinical disorder Professor Zaninovic highlights similarities between the associated with cassava dependency. Whereas spastic clinical and neuropathological features of lathyrism paraparesis (Konzo) in children and women of child- and HTLV-1-associated TSP/HAM and, on this basis, bearing age has been reported in central and southern suggests a common or similar cause, possibly of Africa, cassava-dependent populations in west Africa toxometabolic origin. This hypothesis is advanced to are more prone to an ataxic myeloneuropathy in middle account for the fact that 55% of TSP/HAM cases are age. Some of these latter cases may have been counted reportedly HTLV-1 seronegative. He further suggests as TSP/HAM. Whether the differences in age of onset that cassava dependency may be responsible for some and neurological picture arise from differences in cases of spastic paraparesis (Konzo) in HTLV-1 neurotoxin content of cassava-based food is a question seronegative populations. that merits prompt investigation. Further, Zaninovic's The hypothesis stands on a rocky foundation because proposal correctly implies that viral and non-viral of the differences between TSP/HAM and the two environmental factors may combine to cause the toxic-nutritional disorders. Whereas Konzo and expression of human disease. lathyrism may begin abruptly, improve and then stabilize, TSP/HAM has an insidious onset and is slowly progressive. It is true that Konzo and lathyrism

14 Lathyrus Lathyrism Newsletter 2 (2001)

Electrophoretic phenotypes of diversity and intra and interspecific-relationships among collections is quite poor (1,16). different enzymes in some entries of Lathyrus sativus L. The aims of this study were: a) to add information on the genetic structure of 1 2 1 Lathyrus sativus L. E. Alba , G. B. Polignano , D. De Carlo and b) to verify the possibility of classifying grasspea 3 A. Mincione entries based on isoenzyme traits.

1. Dipartimento di Biologia Difesa e Biotecnologie Materials and Methods Agro Forestali, Università della Basilicata, Potenza, Twelve grasspea populations, selected for morpho- Italy agronomic traits from the Southern Italy collection 2. Istituto del Germoplasma, C.N.R., Via Amendola maintained at the Centro Inter-universitario del 165/a, Bari, Italy Germoplasma Mediterraneo (CIGM), University of 3. Dipartimento di Agrochimica ed Agrobiologia, Basilicata, Potenza, Italy were used. The accessions are Università degli Studi di Reggio Calabria, Gallina described for 1000-seed weight, seed protein content (RC), Italy and seed coat colour (Table 1). Ten plants per population were grown in a controlled environment Email: (2) [email protected] chamber and analysed for 9 enzyme systems: acid phosphatase (APH), phospho-gluconate dehydrogenase (PGD), aspartate aminotransferase (AAT=GOT), Introduction shikimate dehydrogenase (SKD), endopeptidase (END), In Italy, grasspea (Lathyrus sativus L.) is one of the glucose-6-phosphate dehydrogenase (G-6-PDH), least common grain legumes. Today, it is not included glutamate dehydrogenase (GDH), diaphorase (DIAP) in the official statistics of Italy and its survival is due to and leucine aminopeptidase (LAP). the praiseworthy action of a small number of farmers in very limited areas of peninsular and insular central- Table 1. List of Lathyrus sativus L. entries tested. southern Italy. Since the Second World War the sharp reduction in cultivation, similarly for other grain Entry* 1000 Seed Seed protein Seed coat legumes, has certainly contributed to relegate grasspea weight (g) content (%) colour among neglected crops in the Italian and European ACL 2 372 24.8 White agricultural research programs. ACL 29 327 24.9 “ ACL 53 332 26.5 “ The development of sustainable and environmentally ACL 77 325 25.2 “ sound agricultural systems has stimulated a rising ACL 80 294 25.2 “ interest in the role of grain legumes and of this crop, in ACL 94 369 25.2 “ particular, in the world scientific community (2,4,8). As ACL 95 356 26.0 Grey for other neglected grain legumes, the lack of varieties ACL 104 316 25.1 White has favoured breeding programs using local material ACL 109 317 25.2 “ generally represented by old landraces (15). The renewed ACL 130 339 25.7 “ interest of both farmers and plant breeders for grasspea ACL 137 351 25.4 “ has induced a concomitant and necessary action on the ACL 146 303 25.4 “ primary gene pool of this endangered species. In *Entry number of Alba’s collection of Lathyrus spp. particular, research programs have been started for maintained at CIGM, University of Basilicata, Potenza, safeguarding, collecting and establishing several Italy. collections at different scientific institutions, complemented by diversity evaluation programs on them (11). Five hundred mg of younger leaves were picked from plants at the 4 to 5 true-leaf stage, and homogenised Among molecular methods, the electrophoretic analysis with 3 ml of Tris-HCl 75 mM pH 7.5 added with 5% of isoenzyme variation has proved to be particularly sucrose, 5% PVP w/v, 14 mM β-mercaptoethanol 0.1% useful in defining more precisely the size and structure v/v, 50 mM ascorbic acid, 10 mM of genetic diversity in the gene pools of different grain diethyldithiocarbamate, albumin bovine serum 0.1% (14) legumes (6,7,10). In grasspea, the literature on both genetic w/v . After centrifuging at 10 000 rpm for 15

15 Lathyrus Lathyrism Newsletter 2 (2001) minutes, the supernatant was removed and stored at – cathodal zone, PGD-1, G6PDH-1 and GDH-1, which 80°C until it was used. The extracts were displayed both homozygous and heterozygous genetic electrophoresed on 12% starch gel at 4°C and 180 mA. structures. The fastest band was always present, on the As regards the SKD enzyme system, a buffer of contrary, the slowest one was highly discriminating. histidine-citrate 0.2 M pH 6 was used, while a solution obtained from the running buffer with water (1:6 ratio) Glutamate oxalacetate transferase (GOT=AAT). This was used for the gel. For all other systems a buffer enzyme system exhibited two isozymes (GOT-1 and composed of Tris-citrate 0.2M pH 8.3 and lithium- GOT-2) corresponding to three phenotypes. GOT-1 was borate 0.5M pH 8.3 (9:1 ratio) was applied. At the end polymorphic, GOT-2 monomorphic (band a). of the run, starch gel slices about 1.5 mm thick were Heterozygous individuals exhibited a three-banded stained (14). pattern.

Band frequencies were calculated for each entry and for Diaphorase (DIAP). Three isozymes were detected in each isozyme system. The intra-population variation gel slices stained for DIAP. Of the three zones of was estimated by the Shannon-Weaver (H’) diversity activity resolved for this enzyme system, some index (12): electrophoretic variation was found only for the n intermediate one, characterized by five bands (c, d, e, f,

H’= -Σ pi log2 pi g). i=1 Endopeptidase (ENP). One zone of endopeptidase th where pi is the frequency of the i band and n is the activity was present. Heterozygous individuals exhibited number of bands observed for each enzyme. This index a three-banded pattern. applied by different authors (Hutchenson, 1970; Polignano et al., 1998) provides a quantitative estimate Leucine amino peptidase (LAP). Two clear regions of of the intra-population diversity. LAP activity were resolved, some electrophoretic variation was found only for the cathodal one, LAP-1, Based on the matrix of band frequencies and which exhibited both homozygous and heterozygous standardised data, the inter-population variation was phenotypes examined by cluster analysis, and Ward’s method (PROC CLUSTER) was utilized to calculate Euclidean Shikimate dehydrogenase (SKD). Assays showed two distances (1). regions of activity. The five band variants observed (a, b, c, d, e) suggest the presence of two loci SKD-1 and Results and Discussion SKD-2. The latter was made visible in one Polymorphism was present for 19 out of the 43 monomorphic band in all surveyed individuals. Three isozymes resolved in 9 enzyme systems (Table 2). the bands of isozyme activity were observed in the SKD-1 results obtained for each enzyme system are explained zone among the analysed individuals. below: A total of 43 bands with frequency values ranging from Acid phosphatase (APH). Eighteen patterns and three 0 to 1 were observed. Average frequencies below 5% zones of activity were detected: a cathodal (APH-1), an indicate a rare allelic presence; instead, if the frequency intermediate (APH-2) and an anodal zone (APH-3). The of the most frequent allele to a fixed locus is less than zone APH-3 with two bands (g, h) was monomorphic. 95% the population is considered to be polymorphic The zone APH-2 was polymorphic with three bands (d, (Brown and Weir, 1983). Values lower than 5% were e, f). Similarly, the zone APH-1 was polymorphic with a estimated for the following bands: APH-a (entries 137 slow (a), an intermediate (b) and a fast band (c). and 146), APH-b (entries 104 and 137), SKD-a (entries 146 and 53), END-a (entry 137) and DIAP-g (entries Phospho-gluconate dehydrogenase (PGD), glucose-6- 53, 80 and 137). The bands APH-g-h, PGD-b-c-d, AAT- phosphate dehydrogenase (G6PDH) and glutamate a-d, SKD-b-d-e, END-c, G6PDH-a-c-d, GDH-a-c-d, dehydrogenase (GDH). Of the two zones of activity DIAP-a b-c-e-h and LAP-b-c were found in all entries. detected for these enzyme systems, some The average values of the bands for each enzyme electrophoretic variation was found only for the system ranged from 4.3 to 4.7.

16 Lathyrus Lathyrism Newsletter 2 (2001)

Table 2. Band frequencies for 9 enzyme systems in 12 Lathyrus sativus L. entries (ACL number, see Table 1). Some bands are grouped signifying that frequencies were all of value 1.

Enzyme Band Entry Mean 104 109 130 137 146 2 29 53 77 80 94 95 APH a 0.10 0.30 0.60 0 0 0.10 0.20 0.30 0.40 0.60 0.50 0.70 0.32 b 0.07 1 0.90 0 0.70 0.60 0.50 1 0.70 0.70 0.80 1 0.41 c 1 0.70 0.70 1 1 1 1 0.60 0.80 0.90 0.70 0.50 0.41 d 0.50 0.80 1 0.90 0.60 0.60 0.90 0.70 0.90 0.50 0.40 0.80 0.63 e 0.80 1 1 0.70 0.90 1 0.90 1 1 1 0.90 1 0.35 f 1 0.60 0.70 1 0.90 1 0.70 0.80 0.80 0.80 1 0.50 0.48 g&h 111111111111 1 PGD a 0.50 0.50 0.40 0.50 0.50 0.50 0.50 0.50 0.60 0.30 0.50 0.60 0.49 b,c&d 111111111111 1 AAT a&d 111111111111 1 b 0.60 0.50 0.60 0.40 0.60 0.50 0.70 0.70 0.60 0.60 0.60 0.70 0.59 c 0.30 0.20 0.30 0.20 0.20 0.10 0.30 0.40 0.30 0.40 0.10 0.30 0.26 SKD a 0.80 0.90 0.30 0.60 0 0.90 0.20 0 0.80 0.30 0.60 0.70 0.51 b,d&e 111111111111 1 c 0.40 0.30 0.20 0.50 0.40 0.20 0.20 0.60 0.50 0.70 0.10 0.60 0.39 END a 0.10 0.30 0.20 0 0.10 0.10 0.30 0.20 0.30 0.10 0.40 0.10 0.18 b 0.70 0.80 0.50 0.50 0.90 0.90 0.70 0.80 0.70 0.40 0.70 0.50 0.68 c 111111111111 1 G6PDH a,c&d 111111111111 1 b 0.60 0.20 0.60 0.30 0.50 0.40 0.40 0.60 0.40 0.40 0.50 0.60 0.46 GDH a,c&d 111111111111 1 b 0.60 0.40 0.40 0.40 0.60 0.40 0.60 0.70 0.50 0.40 0.50 0.60 0.51 DIAP a,b,c,e&h 111111111111 1 d 0.40 0.50 0.50 0.50 0.40 0.40 0.40 0.40 0.30 0.30 0.60 0.40 0.43 f 0.50 0.50 0.40 0.40 0.40 0.40 0.60 0.50 0.60 0.30 0.50 0.40 0.46 g 0.20 0.10 0.20 0 0.10 0.20 0.20 0 0.20 0 0.20 0.20 0.13 LAP a 0.50 0.60 0.60 0.60 0.80 0.80 0.60 0.30 0.40 0.60 0.50 0.30 0.55 b&c 111111111111 1 Mean alleles per locus 4.7 4.7 4.7 4.3 4.5 4.7 4.7 4.7 4.7 4.7 4.6 4.7

The Shannon-Weaver diversity index H’ was calculated to compare the diversity expressed by the enzyme A high H’ value was estimated for the PGD system in systems and the entries (Table 3). For a single enzyme all entries except for entries 130 and 80, which were system a low H’ value indicates unbalanced frequency monomorphic. On the contrary, the AAT system was classes and lower diversity levels. On the contrary, polymorphic in entries 2, 29, 53, 94, 95 and 146. higher H’ values indicate a greater balance among frequency classes and greater diversity levels. The For both SKD and END systems lower H’ values were average value of H’, estimated for each entry, ranged estimated in all entries. In particular, the entries 146 and from 0.56 (entry 2) to 0.89 (entry 137). Similarly, the 53 resulted monomorphic for the SKD system, and the average H’ value for each enzyme system ranged from entry 137 for the END system. The highest H’ values 0.28 (END) to 0.94 (G6PDH). Lower H’ values were were estimated for both enzyme systems G6PDH and estimated for APH system except for entry 146 (0.61); GDH but not in all entries. The latter systems showed for the latter enzyme system, the entry 137 was the highest number of monomorphic entries with an monomorphic (1.00). index value of 1.

17 Lathyrus Lathyrism Newsletter 2 (2001)

Table 3. Shannon-Weaver (H’) diversity indices for 9 enzyme systems observed in 12 Lathyrus sativus L. entries (see Table 1).

Entry Enzyme Mean APH PGD AAT SKD END G6PDH GDH DIAP LAP

ACL 104 0.19 0.93 0.96 0.21 0.20 0.81 0.81 0.34 0.94 0.60 ACL 109 0.22 0.93 0.96 0.17 0.23 0.88 1.00 0.31 0.81 0.61 ACL 130 0.18 1.00 0.96 0.26 0.28 0.81 1.00 0.34 0.81 0.63 ACL 137 1.00 0.93 1.00 0.24 1.00 1.00 1.00 1.00 0.81 0.89 ACL 146 0.61 0.93 0.90 1.00 0.13 0.94 0.81 0.32 0.50 0.68 ACL 2 0.17 0.93 0.83 0.15 0.13 1.00 1.00 0.35 0.50 0.56 ACL 29 0.22 0.93 0.90 0.24 0.26 1.00 0.81 0.33 0.81 0.61 ACL 53 0.22 0.93 0.90 0.97 0.21 0.81 0.69 1.00 1.00 0.74 ACL 77 0.22 0.81 0.96 0.19 0.26 1.00 0.94 0.33 1.00 0.63 ACL 80 0.24 1.00 0.96 0.23 0.16 1.00 1.00 1.00 0.81 0.71 ACL 94 0.25 0.93 0.83 0.20 0.26 0.94 0.94 0.32 0.94 0.62 ACL 95 0.23 0.81 0.90 0.21 0.25 0.81 0.81 0.35 1.00 0.60 Mean 0.31 0.93 0.93 0.34 0.28 0.94 0.88 0.50 0.81 0.66

The DIAP system showed generally medium to low H’ been observed, including 19 with different frequency value and three entries (137, 53 and 80) resulted percentage, and 24 which were always present. No monomorphic. Similarly, the last system, LAP, showed variation was observed in the average number of medium to high H’ values except three entries (523, 77 bands/enzyme. and 95) which resulted monomorphic. A cluster analysis revealed four distinct groups in which Figure 1 is a phenogram constructed using Ward’s entries revealed a greater similarity. In other words, the minimum variance method. The most appropriate entries belonging to a cluster showed more similar number of clusters was found by looking for a patterns for the enzyme systems compared to those consensus among the four statistics R2 (RSQ), cubic located in different clusters. None of the nine enzyme clustering criterion (CCC), pseudo-F (PSF) and pseudo- systems observed was exclusive to define one or more t2 (PST2) (9). Consequently, four main clusters may be groups. In fact, all the enzyme systems showed a similar identified, which explain a moderate variance (49%). trend of variation in the groups obtained.

Five of the twelve entries are included in cluster III (29, The wide intra-population variation observed suggests 77, 94, 109 and 130). Cluster I is represented by one the opportunity to extend the survey to wider segments entry which is very distant from the other clusters. of the collection. More detailed studies including more Entries 53 and 95 are included in cluster II. The cluster differentiated materials could be useful to investigate IV includes four entries with entry 137 being more the associations among morpho-agronomic and isozyme distant from the other entries (2, 104 and 146). traits. Moreover, the inclusion of materials from different geographical regions could be useful to Conclusions ascertain the available variation and its distribution. Despite the limited number of tested entries and Finally, the lack of information on the genetic control of isozyme loci examined in this study, some useful single enzymes in Lathyrus has not allowed the conclusions may be drawn. Lathyrus sativus entries examination of allelic frequencies. Future research showed a wide polymorphism for all the enzyme should be directed to more accurate investigations to systems being tested. Other similar results have been elucidate the genetics of different enzyme systems. reported for L. sativus (1,16). A total of 43 bands have

18 Lathyrus Lathyrism Newsletter 2 (2001)

Semi partial R squared.

Fig. 1. Phenogram from cluster analysis of twelve Lathyrus sativus L. entries (ACL number, see Table 1) based on a distance matrix of allele frequencies at 19 polymorphic loci.

Acknowledgements Reggio Calabria - Gallina, 16-17 /6/ 97. pp. 107- We wish to thank Mr. G. Scippa for his technical 121. support. Critical reading of the manuscript by Dr. P. 3. Brown AHD, Weir RS. (1983). Measuring genetic Perrino is gratefully appreciated. variability in plant populations. In: Tanksley SD, Orton TJ (Eds.). Isozyme in plant genetics and References breeding. Elsevier, Amsterdam. pp. 219-239 1. Alfaro M, Paredes OM, Becerra V L. (2000). 4. Campbell CG, Mehra RB, Agrawal SK, Chen YZ, Genetic variation in Lathyrus sativus and their Abd el Moneim AN, Khawaja HIT, Yadov CR, Tay relationships with other species. In: Proceedings of JU, Araya WA. (1994). Current status and future International Food Legume Research Conference strategy in breeding grass pea (Lathyrus sativus L.). III. Adelaide, Australia, 22-26 September 1997. p. Euphytica 73, 167-175. 134. 5. Hutchenson K. (1970). A test for comparing 2. Bozzini A. (1997). Utilizzazione di germoplasma in diversities based on the Shannon formulae. J. alcune leguminose da granella presso il C. R. Theor. Biol. 29, 251-254. Casaccia dell’ENEA. In: Atti del 3° Convegno 6. Labdi M, Robertson LD, Singh KB, Charrier A. Nazionale “Biodiversità: Tecnologie-Qualità , (1996). Genetic diversity and phylogenetic relationships among the annual Cicer species as

19 Lathyrus Lathyrism Newsletter 2 (2001)

revealed by isozyme polymorphism. Euphytica 88, 11. Sarker A, Robertson LD, Campbell CG. (2000). 181-188. Lathyrus spp.: Conserved Resources, Priorities for 7. Polignano GB, Quintano G, Bisignano V, Uggenti Collection and Future Prospects. In: Proceedings of P, Della Gatta C, Alba E. (1998). Enzyme International Food Legume Research Conference polymorphism in faba bean (Vicia faba L. minor) III. Adelaide, Australia, 22-26 September 1997. p. accessions. Genetic interpretation and value for 73. classification. Euphytica 102, 169-176. 12. SAS Institute, 1989. SAS user’s guide: Statistics. 8. Polignano GB, Uggenti P. (1996). Risorse Version 6 ed. SAS Inst., Cary, NC, USA genetiche e selezione varietale per produzioni 13. Shannon CE, Weaver W. (1949). The mathematical biologiche in alcune leguminose da granella theory of communication. Univ. Illinois Press, neglette. In: Atti 2° Convegno Nazionale Urbana, USA. Biodiversità e Produzioni Biologiche. 11 Giugno 14. Soltis DE, Soltis PS. (1989). Isozymes in Plant 1996, Matera. pp. 163-172. Biology. Dioscorides Press, Portland Oregon. 9. Polignano GB, Uggenti P, Scippa G (1993). The 15. Tavoletti S, Porfiri O, Petrini A, Papa R, Taffetani pattern of genetic diversity in faba bean collections F. (1998). La riscoperta della cicerchia (Lathyrus from Ethiopia and Afghanistan. Genetic Resources sativus L.) come coltura tradizionale. In: Abstracts and Crop Evolution 40, 71-75. IV Congresso Nazionale BIODIVERSITA’: 10. Rodriguez MM, Paredes OM, Becerra VL. (2000). Germoplasma locale e sua valorizzazione. Alghero, Isozyme diversity of Chilean lentil germplasm 8-11 Settembre 1998. (Lens culinaris Medik). In: Proceedings of 16. Yunus AG, Jackson MT, Catty JP. (1991). International Food Legume Research Conference Phenotypic polymorphism of six enzymes in the III. Adelaide, Australia, 22-26 September 1997. p. grass pea (Lathyrus sativus L.). Euphytica 55, 33- 135. 42.

20 Lathyrus Lathyrism Newsletter 2 (2001)

Autogamy and allogamy in genus pots, 20 cm in diameter, the plants were grown under homogenous conditions (25°C and 12h /day length) in Lathyrus an experimental garden, at the National Agronomic Research Institute of Tunisia (Upper Semi-arid). Plant flowering occurred from April (L. cicera, L. sativus and L. aphaca) to June (L. latifolius and L. sylvestris). N. Ben Brahim1, D. Combes2 and M. 3 Three types of treatments were performed: selfing by Marrakchi bagging flowers in order to prevent insect visitation of the flower (Treatment A), natural pollination without 1. Laboratoire de botanique, Institut National de la castration of anthers (Treatment B) and natural Recherche Agronomique de Tunisie, pollination following castration of anthers (Treatment Rue Hédi Karray, 2049, Ariana, Tunisie. C). Each type of operation was performed on 15 2. Laboratoire d’Ecologie Moléculaire, IBEAS, flowers per plant, 5 plants were studied per species. Université de Pau et des Pays de Ladours, France. Two fertility parameters were measured for each plant 3. Laboratoire de Génétique et de Biologie on 15 flowers per plant: the frequency of flowers Moléculaire, Faculté des Sciences de Tunis. giving pods (p/f) and the mean number of seeds per pod (g/p). Email : (1) [email protected] The values of the ratio p/f ranged from 0 to 1, the transformation

Introduction f(x) = arc sin√x, where x = p/f, was used. In Tunisia the genus Lathyrus is represented by 15 (11) native species , distributed essentially in marginal Statistical analysis. The data of the fertility parameters environments in the north of the country. Many of were analysed by analysis of variance to test them are of considerable agronomic interest as forage differences between species and between species and (L. ochrus and L. articulatus) and as human food (L. breeding system effects. The Duncan test (7) was used cicera and L. sativus). These species are under utilised to compare means. The SAS procedure ANOVA was and in the past little effort has been made to expand the used. selection and breeding of Lathyrus as a food crop. However, many Lathyrus species have potential as an alternative pulse in many cropping systems as they are Results tolerant to low rainfall conditions. Research is now being conducted to address aspects of improvement of Selfing by bagging flowers. The three perennial some species such as L. sativus, including reducing species L. latifolius, L. sylvestris and L. tuberosus did concentrations of the neurotoxin β-N-oxalyl-L-α, β- not develop pods by selfing, the ratios p/f and g/p were diaminopropanoic acid (ODAP) in the seed, both zero (Table 2) and indicate that these species are introducing insect and disease resistance, and strictly outcrossing. For that reason they were excluded increasing yield (4). from statistical analysis. The variance analysis showed highly significant differences between annual species In order to identify useful material for agriculture and for the two studied fertility parameters: number of for breeding programs, we studied breeding systems flowers giving pods p/f (F ratio 6.04) and number of and analysed some aspects of floral biology of fourteen seeds per pod g/p (F ratio 37.41). annual and perennial Lathyrus species. The comparison of means using Duncan test at 5 % Materials and methods (Table 3) reveals a regrouping of most species for high Fourteen annual and perennial Lathyrus species means for the number of flowers giving pods (p/f). belonging to four sections according to Kupicha’s However, L. odoratus is distinguished from other classification (10) were analysed, each species was species by a low mean for p/f (0.25) (Table 2). represented by one population. All these species are diploid (2n=2x=14). Studied species, their biological The comparison of means at 5% showed that L. type and their sites of origin are listed in Table 1, only nissolia had the highest number of seeds per pod (9.30) L. sativus, L. cicera and L. odoratus (the ornamental (Tables 2 and 4). This can be explained partly by the sweetpea) are cultivated. high number of ovules per ovary in this species (Table 5). L. aphaca, L. cicera, L. sativus and L. setifolius had The seeds used in this study were germinated in petri the lowest means for g/p (range 2.44 to 4.12). dishes on wet filter paper. They were manually scarified in order to increase their germination percentage. Seedlings were transplanted in October to

21 Lathyrus Lathyrism Newsletter 2 (2001)

Table 1. Lathyrus species studied, nomenclature of Kupicha(10).

Species Abbreviation Section Habit Origin L. annuus AN Lathyrus Annual Tunisia L. aphaca A Aphaca Annual Italy L. articulatus AR Clymenum Annual Tunisia L. cicera C Lathyrus Annual Tunisia L. hirsutus H Lathyrus Annual Italy L. latifolius L Lathyrus Perennial Tunisia L. nissolia N Nissolia Annual France L. odoratus O Lathyrus Annual Tunisia L. ochrus OC Clymenum Annual Tunisia L. sativus SA Lathyrus Annual Tunisia L. setifolius SE Lathyrus Annual Italy L. sylvestris S Lathyrus Perennial France L. tingitanus TI Lathyrus Annual Portugal L. tuberosus T Lathyrus Perennial Hungary

Table 2. Frequency of flowers giving pods and mean number of seeds per pods in selfing and in natural pollination.

Species Frequency of Mean number of flowers giving seeds per pod pods (p/f) (g/p) ABCABC L. annuus AN 0.82 0.84 0 5.33 5.43 0 L. aphaca A 0.72 0.79 0 4.12 4.32 0 L. articulatus AR 0.64 0.88 0 5.31 5.38 0 L. cicera C 0.91 0.86 0.13 3.58 3.44 3 L. hirsutus H 0.88 0.90 0 5.80 6.25 0 L. latifolius L 0 0.85 0.11 0 6.34 6.37 L. nissolia N 0.75 0.82 0 9.30 9.75 0 L. ochrus OC 0.81 0.80 0 5.11 4.63 0 L. odoratus O 0.25 0.81 0.20 5.84 6.67 5.46 L. sativus SA 0.53 0.80 0.09 2.82 2.50 2.57 L. setifolius SE 0.66 0.72 0 2.44 2.70 0 L. sylvestris S 0 0.73 0.13 0.20 5.07 6.40 L. tingitanus TI 0.68 0.89 0 6.33 6.70 0 L. tuberosus T 00.880040

A selfing by bagging B natural pollination without castration C natural pollination following castration

However, L. nissolia is characterized by high values of Natural pollination without castration of anthers. g/p. L. setifolius and L. sativus have the lowest ratio of Variance analysis indicated no significant differences g/p and the lowest number of ovules per ovary (Table for the ratio p/f (F ratio1.56) and highly significant 5). differences for fertility parameter g/p (F ratio 48.35). In natural pollination without castration of anthers, the The comparison of means by Duncan test at 5 % (Table values of the two fertility parameters are higher than 4) demonstrated a regrouping between all species for those observed in selfing with bagging anthers (Table the number of pods per flower (p/f) (Table 3). 2). This is particularly prominent for the three

22 Lathyrus Lathyrism Newsletter 2 (2001) perennial species L. latifolius, L. sylvestris and L. with castration than in natural pollination without tuberosus , and also L. odoratus, all of which did not castration. Some of this reduction could be attributed to develop pods in selfing. The values of p/f increase lesion caused by castration. respectively from 0 to 0.88 for L. tuberosus, from 0 to Comparison of species by breeding system effects. 0.85 for L. latifolius, from 0 to 0.73 for L. sylvestris Variance analysis with two classification criteria and from 0.25 to 0.81 for L. odoratus. indicated that there is a species by breeding system effect, which is highly significant for the two studied Natural pollination following castration of anthers. parameters (F ratio 9.75, for p/f and F ratio 47.23 for The results of natural pollination following castration g/p). revealed that the castrated flowers from annual species (L. aphaca, L. nissolia, L. articulatus, L. ochrus, L. The comparison of averages reflects a regrouping for tingitanus, L. annuus, L. setifolius and L. hirsutus) did species for the parameter p/f. However, a heterogeneity not develop pods and seeds (Table 2). However, for is observed for the ratio g/p: L. nissolia is characterised five species (L. odoratus, L. sativus, L. latifolius, L. by the highest means, while L. setifolius and L. sativus sylvestris and L. cicera) the ablation of anthers caused have the lowest means (Table 4). a clear reduction in the number, but did not completely inhibit the development of pods and seeds. This Comparison between selfing with bagging, natural indicates clearly that the three cultivated annual species pollination without castration of anthers and (L. odoratus, L. sativus and L. cicera) can be pollinated natural pollination following castration of anthers. either by their own pollen or by foreign pollen. The comparison of the three breeding systems using Duncan test at 5 % reveals a separation of breeding Variance analysis revealed no significant differences systems for the two parameters p/f and g/p. For the between the five species for the parameter p/f (F ratio ratio g/p, the two breeding systems: natural pollination 2.58) (Table 3) and highly significant differences for and selfing by bagging are grouped with the highest g/p (F ratio 12.83) (Table 4). The comparison of the means. Also, we observed a classification of the three means shows that L. latifolius, L. sylvestris and L. breeding systems in the following decreasing order of odoratus have the highest means for the number of efficacy: natural pollination without castration of seeds per pod (g/p). anthers B, selfing by bagging flowers A, natural pollination following castration C. This occurred for It is important to mention that the values obtained by both p/f and g/p. the analysed parameters are lower in natural pollination

Table 3. Comparison of means using Duncan test at 5% for the fertility parameter p/f. Treatments are selfing by bagging flowers (Treatment A), natural pollination without castration of anthers (Treatment B) and natural pollination following castration of anthers (Treatment C). Species are ranged in order from least to greatest, left to right. Species under a continuous bar are not significantly different to each other. Species abbreviations are found in Table 1.

A C H AN OC N A TI SE AR SA O

B H TITARS C LANNOOCSAASE

C SA L S C O

Comparing species across all of Treatments A, B and C:

THANOCCANARTISESASLO

Comparing Treatments A, B and C:

ABC

23 Lathyrus Lathyrism Newsletter 2 (2001)

Table 4. Comparison of means using Duncan test at 5% for the fertility parameter g/p. Notation as for Table 3.

A NTIOHANAROC A CSASE

B NTIOL HANARSOCAT CSESA

C SLO CSA

Comparing species across all of Treatments A, B and C:

D N TI O H L OC AN AR S A T C SE SA

Comparing Treatments A, B and C:

E BAC

Table 5. Means comparison using Duncan test at 5% for the parameter mean number of ovules per ovary (No/OV). Notation as for Table 3.

Species LNSTOTIANOCAHARCSASE No/OV 18.6 12.2 11.8 11.5 11.3 8.8 8.0 7.5 6.6 6.4 5.8 5.3 3.5 2.9

Table 6. Means comparison using Duncan test at 5% for the floral parameters: mean flower length in mm (LF) and mean number of flowers per inflorescence (NF/IN). Notation as for Table 3.

Species O L TI OC T S AR SA N C SE H A AN LF 33.5 26.1 23.4 17.5 15.6 15.6 15.6 14.1 13.6 12.0 11.8 11.2 11.1 11.0

Species L S T O TI AR H SA C A N OC SE AN NF/IN 107332221111111

24 Lathyrus Lathyrism Newsletter 2 (2001)

Discussion In natural pollination following a castration, ablation of In selfing by bagging flowers, with the exception of L. anthers leaded to the absence of developing pods in odoratus (p/f = 0.25), the values of p/f are high in annual species (L. aphaca, L. nissolia, L. articulatus, L. annual species and zero in the three perennial ones. The ochrus, L. tingitanus, L. annuus, L. setifolius and L. values of p/f in the ten annual species ranged from 0.53 hirsutus) and the perennial L. tuberosus. For annual (L. sativus) to 0.91 (L. cicera) and indicate that these species, this result shows that pollination is assured by species are preferentially autogamous (3,9,12). their own pollen. For the three cultivated species (L. odoratus, L. sativus and L. cicera) ablation of anthers The autogamous breeding system, which is a caused a great reduction of the fructification but did not characteristic for annual species of genus Lathyrus, inhibit it. This indicates that these species are seems to be favored by certain characters related to the pollinated by their own pollen and can receive foreign floral biology of plants. The comparison of percentage pollen. The absence or the reduction of fructification of selfing in ten annual species with means of the floral indicates that the pollen takes part in pollinator parameters (Table 6) measured on 60 flowers and with attraction, particularly bees (2). the flower colour show a relationship between the selfing ratio and these parameters. In fact, L. cicera, L. hirsutus, L. annuus, L. ochrus, L. aphaca, L. nissolia, References L. sativus, L. tingitanus, L. setifolius and L. articulatus 1. Asmussen CN. (1993). Pollination biology of the are characterised by a reduced number of flowers (1 to sea pea, Lathyrus japonicus: floral characters and 2) per inflorescence (Table 6), flowers of small size activity and activity and flight patterns of (range mean 11 mm in L. annuus to 23.4 mm in L. bumblebees. Flora 188, 227-237. tingitanus), a pale or dark flower colour (white or blue 2. Ben Brahim N. (1990). Etude du mode de in L. sativus; yellow in L. annuus, L. aphaca and L. reproduction et de la variabilité génétique chez ochrus; blue in L. hirsutus; red in L. cicera, L. trois espèces de Lathyrus: L. cicera L. , L. setifolius and L. nissolia). These parameters make these articulatus L. et L. ochrus D.C. Thèse de doctorat, flowers less attractive to pollinators (6,13,15,17). All these Tunisie. 200 p. species have percentages of selfing higher (ranging 3. Ben Brahim N, Boussaid M, Marrakchi M. (1990). from 0.53 in L. sativus to 0.91 in L. cicera) than those Analyse de la biologie florale chez trois espèces observed in L. odoratus (0.25) and in the three annuelles de Lathyrus. AN. INRAT. 65 (15), 1- 21. perennial species L. latifolius, L. sylvestris and L. 4. Campbell CG, Mehra RB, Agrawal SK, Chen YZ, tuberosus (0). Abd El Moneim AM, Khawaja HIT, Yadov CR, Tay JU, Araya WA. (1994). Current status and The lowering of fertility observed in selfing in all future strategy in breeding grasspea (Lathyrus species could be explained by the stress on flowers due sativus). Euphytica 73, 167-175. to bagging. The bagging of flowers increases the 5. Chérifi K. (1993). Analyse de la diversité temperature and causes the fall of pollinated flowers génétique de populations naturelles de deux before the maturity of pods. Similar observations have espèces de Medicago annuelles : Medicago ciliaris been made in the genus Medicago (5). (L.) Krock et Medicago intertexta (L.) Mill. Thèse de 3ème cycle, Fac. Sci. Tunis, 115 p. The high values of the ratio p/f obtained in natural 6. Chriki A. (1986). L’hérédité des anthocyanes pollination without castration express that the three florales chez Hedysarum coronarium, H. carnosum perennial species are strictly outcrossing and L. et H. capitatum. Thèse Doctorat d’Etat. Univ. de odoratus is preferentially outcrossing (14). The Pau (France). 229 p. development of pods and seeds in perennial species is 7. Dagnelie P. (1975). Théorie et Méthodes due to pollinators, especially bees and bumblebees. The statistiques Vol. I and II. Gembloux Presse Agron. flowers of L. latifolius are visited by many insects the 378p and 371p. same day; such as Coleoptera (Bruchus beetles), 8. Hossaert M, Bashar A, Mackey D. (1986). Floral Hymenoptera (Apis, Bombus and Xylocopidae) and nectar production in relation to insect activity in Lepidoptera (16). In fact, the stigma of these species Lathyrus latifolius. In Kaul AK & Combes D becomes receptive only after the rupture of the cuticle (eds.). Lathyrus and lathyrism. Third Medical of the stigma papillae (1). Reproductive characters Research Foundation: New York. 213-221. maintained by natural selection allow these allogamous 9. Kiyoshi Y, Toshiyuki F, Blumenreich ID. (1985). species to be adapted to entomophilic pollination. Isozymic variation and interspecific crossability in Flowers of large size, bright flower colour (red in L. annual species of genus Lathyrus L. In Kaul, A.K. latifolius and L. sylvestris, pink in L. tuberosus) flower & Combes, D. (eds.): Lathyrus and lathyrism. Pau density (an average of 10 flowers per raceme in L. (France), 118-129. latifolius, 7 in L. sylvestris and 3 in L. tuberosus) and 10. Kupicha FK. (1983). The infrageneric structure of nectar production (8). Lathyrus. Notes from the Royal Botanic Garden Edinburgh 41, 209-244.

25 Lathyrus Lathyrism Newsletter 2 (2001)

11. Pottier-Alapetite G. 1979. Flore de la Tunisie. differences between strains of Lathyrus odoratus Angiospermes Dicotylédones. Apétale- L. Cytologia 24, 359-402. Dialypétales. Imp. of. de Tunisie. 554 p. 15. Solbrig O T. (1976). On the relative advantages of 12. Plitmann N, Heyn CC, Weinberger H. (1985). cross-and self fertilisation. Ann. Missouri. Bot. Comparative taxonomy of some wild species allied Gard. 63, 262-276. to Lathyrus sativus. In Kaul, A.K. & Combes, D. 16. Valero M, Youssef A,Vernet P, Hossaert M. (eds.): Lathyrus and lathyrism. Pau (France), 8- (1985). Is there a polymorphism in the breeding 21. system of Lathyrus latifolius. In Kaul AK & 13. Salhi-Hannachi A. (1996). Analyse de la diversité Combes D. (eds.): Lathyrus and lathyrism. Pau génétique des populations spontanées de Medicago (France), 105-117. L: Variabilité morphologique et polymorphisme 17. Wyatt R. (1983). Pollinator plant interrelations and isoenzymatique. Thèse de doctorat, Tunisie. 178p. the evolution of breeding systems. In Real, L. (ed). 14. Sharma AK, Datta PC. (1959). Application of an Pollination biology. Academic Press. London. improved techniques in tracing karyotype

26 Lathyrus Lathyrism Newsletter 2 (2001)

Evaluation of selected traits in • Select those with interesting characteristics to be used in breeding. grasspea (Lathyrus sativus L.) • Find correlations of individual traits, and to genetic resources. characterise the set from the point of all traits.

Materials and methods Michaela Benková1 and Mária Žáková During 1998 and 1999 years a set of 35 grass pea genetic resources, comprising 34 of Slovak (SVK) and 1 of Syrian (SYR) origin, was evaluated in a maize Research Institute of Plant Production (RIPP), growing area at RIPP Pieštany. The trial was sown in Bratislavska cesta 122, 921 68 Pieštany, Slovak plots with a harvest area of 10 m2 (6.25 x 1.6 m). Republic. Distance between rows was 0.4 m. During the vegetative period the genetic resources were evaluated Email: (1) [email protected] with morphological, phenological, biological and economic parameters using relevant descriptors for Lathyrus spp. Introduction Grass pea (Lathyrus sativus L.) is one of the most In mechanical analyses values of the following traits important species of the Lathyrus genus. It has two were measured in 20 plant samples: (8) subspecies : spp. asiaticus Zalk., which has small, • Time to maturity (sowing date to harvest) (d). coloured seeds, occurring mainly in the continental • Plant height (mm). part of Asia and Africa and spp. europaeus Zalk. with • Height of setting the first pod (mm). large seeds, occurring predominantly in Central • Plant dry weight (g). (8) Europe and European Russia. According to Vavilov • Number of pods per plant. its genetic centre is in the Mediterranean region. It is • Number of seeds per plant. utilised for nutritional, fodder and technical purposes. • Weight of seeds per plant (g). It is undemanding to grow and so has been cultivated • 1000 seeds weight (or TGW) (g). predominantly in dry and warm regions of Southern • Yield (t.ha-1). (5) Europe, North Africa and Asia Minor . It is less • Resistance to lodging, on 1 (low) to 9 (high widespread in Central Europe. resistance) point scale. • State of the stand, on 1 (low leaf area index) to 9 Under the leadership of ECP/GR (European (high) point scale. Cooperative Programme for Crop Genetic Resources • Resistance to the fungus Fusarium oxysporum, on Network) the European database of grass pea was 1 (low resistance) to 9 (high) point scale established in 1985. Its holder has been IBAES in • Protein content was measured by CSN-2000 Pau, France. The database consists of approximately analyser. 4000 grass pea accessions (2). The largest collections of grass pea genetic resources are in India (Raipur, The data were evaluated by variance analysis 2659 genotypes), in Syria (ICARDA, 1560 genotypes) (ANOVA) and correlation coefficients between tested (1) and in France (University of Pau, 1807 genotypes) . traits and characteristics were calculated. For the purpose of classification of the 13-dimensional In Slovakia grass pea has not been grown on a large structure two multivariate techniques were used: area, and nowadays is only grown in the south. In the a principal component analysis (7) and a cluster Research Institute of Plant Production (RIPP) in analysis (6) with Ward’s method. Because different Pieštany, Slovakia the grass pea genetic resources scales were used for the descriptors, data were first have been maintained since 1994, when grass pea standardised as recommended (3). All analyses were landraces were obtained by collecting expeditions. carried out using the Statgraphic v.7.0. and the SPSS Within the National Programme for Cultural Plant v.8.1. package. Genetic Resources Preservation in Slovakia the collection has been expanding, mainly with further Results domestic landraces, comprising a valuable national The results of variance analysis showed that the year resource. At present the grass pea collection is contributed highly significantly to variability in most composed of 49 genotypes, mainly of domestic origin. traits. The effect of genotype was highly significant only for TGW and significant for the height of the first The objectives of this study were to: pod. The effect of the year was not recorded in the • Evaluate the selected biological and economic other traits (Table 1). All genotypes mentioned are of traits and characters of 35 grass pea genetic Slovak (SVK) origin unless specifically designated resources. Syrian (SYR).

27 Lathyrus Lathyrism Newsletter 2 (2001)

Table 1. Mean values and ranges of the set of measured traits, and F ratios of the effect of genotype and years from ANOVA.

Traits F-ratio Mean Range Genotype Years Time to maturity (d) 127 125 - 129 0.82 4366.93 ** Plant height (mm) 854 742 - 997 1.27 400.15 ** Height of lowest pod (mm) 242 192 - 303 1.88 * 0.02 Plant dry weight (g) 30.9 20.0 - 52.2 1.04 0.02 Number of pods per plant 27.2 17.6 - 45.1 1.14 2.03 Number of seeds per plant 53.1 33.6 - 80.9 1.10 9.91** Weight of seeds per plant (g) 14.4 8.6 - 25.1 1.10 13.31** 1000 grain weight (g) 286 232 - 354 4.51** 15.53 ** Yield (t.ha-1) 4.02 3.24 - 4.79 1.23 28.33 ** Resistance to lodging 4.7 3.5 - 6.0 1.78 2.25 State of stand 7.5 6.5 - 8.3 1.73 4.7* Resistance to Fusarium disease 8.1 7.8 - 8.5 1.25 21.35 ** Protein concentration (%) 25.9 23.2 - 27.3 1.64 236.55 ** * significant at P<0.05, ** significant at P<0.01

The whole set had good resistance to Fusarium • time to maturity and height of setting the first pod. oxysporum, the average of this trait was high, 8 points • plant weight and the traits: number of pods per according to the scale. All genotypes were resistant to plant, number of seeds per plant and seed weight the disease during the study. per plant. • number of pods per plant and the traits: number of PCA and cluster analysis. More detailed seeds per plant, seed weight per plant and relationships were revealed by the principal resistance to lodging. component analysis (PCA). A principal component • number of seeds per plant and the traits: seed analysis was carried out to transform the weight per plant and yield. interdependent traits into a set of independent traits as well as to reduce the dimensionality of the structure. Negative strong correlation appeared between the traits: Results of correlation analysis showed the correlation • height of setting the first pod and number of seeds of individual traits. Table 2 shows the matrix of per plant. correlation coefficients. • height of setting the first pod and the traits: plant weight, number of pods per plant and seed In the tested set during 1998-1999 positive strong weight. correlation appeared between the traits, as follows:

Table 2. Selected correlation coefficients between the evaluated traits of grass pea (1998 - 1999)

PH LP PW PP SP Plant height PH Height of lower pod LP 0.53** Plant weight PW - -0.36 * No. pods/plant PP - -0.39 * 0.95** No. seeds/plant SP - -0.56** 0.88 ** 0.89** Seed weight/plant - -0.42 * 0.95** 0.88** 0.88** TGW - 0.40* - - -0.26 Yield - - - 0.13 0.43 ** Lodging resistance - - 0.37 * 0.46 ** 0.34 * * significant at P<0.05, ** significant at P<0.01

28 Lathyrus Lathyrism Newsletter 2 (2001)

Table 3. Correlation coefficients between principal components (PC) and the set of traits.

Trait PC1 PC2 PC3 PC4 PC5 Plant weight 0.97* 0.08 0.08 0.08 0.01 Weight of seed/plant 0.97* -0.04 0.08 -0.04 0.11 Number of pods/plant 0.96* 0.13 -0.03 0.17 0.05 Number of seed/plant 0.93* -0.02 -0.29 0.05 -0.43 Plant height -0.07 0.85* 0.15 -0.09 -0.07 Resistance to Fusarium 0.01 0.67* -0.06 -0.05 0.00 Resistance to lodging 0.36 0.64* -0.15 0.33 0.19 1000 grain weight 0.04 -0.07 0.94* -0.09 0.03 Height of the lower pod -0.50 0.43 0.56* 0.11 -0.30 Protein concentration 0.10 0.02 0.05 0.90* 0.03 Stand -0.07 0.02 0.47 -0.61* 0.18 Time to maturity 0.12 0.30 -0.04 -0.18 0.75* Yield 0.20 0.30 -0.06 -0.13 -0.67* * significant at P<0.05

Five principal components were obtained. Correlation Table 4. Variance of principal components (PC) coefficients between the five principal components and the cumulative contribution to the total (PC) and the set of characters are given in Table 3. variance. Table 4 shows variance of individual components and their cumulative contribution (%) to the total variance. Component Variance Cumulative (%) variance (%) As can be seen in Table 4, the five principal PC1 31.5 31.5 components explained approximately 80% of the total PC2 15.1 46.6 variance. In spite the fact that the evaluated samples PC3 12.0 58.6 did not show strong variability and clusters created by PC4 10.8 69.5 cluster analysis showed no significant differences, four PC5 9.1 78.6 small different groups can be separated (Fig. 1).

The formed groups varied mainly in plant height, height of setting the first pod and seed size. The average values of the evaluated traits in individual groups of grass pea sets are given in Table 5.

Table 5. Means ± standard errors for plant traits of the 4 clusters (see Fig. 1).

Traits Cluster I. II. III. IV. Time to maturity (d) 126.88±0.24 126.88±0.35 126.57±0.43 127.0±0.58 Plant height (mm) 865±14 781±15 852±29 858±9 Height of lower pod (mm) 257±6 232±10 231±8 221±12 Plant weight (mm) 28.5±0.8 28.6±1.0 36.7±1.2 42.8±4.7 Number of pods/ plant 24.8±0.6 24.8±0.8 33.7 ±1.5 36.8±4.3 Number of seeds/ plant 47.9±1.7 50.4±1.9 66.7±3.1 70.8±7.6 Weight of seed /plant (g) 13.2±0.4 13.5±0.8 16.9±0.6 21.2±1.9 1000 seed weight (g) 294±5 282±5 263±10 313±11 Yield (t.ha-1)4.09±0.06 3.98±0.15 3.98±0.21 3.87±0.47 Resistance to lodging 4.8±0.2 4.0±0.2 5.4±0.2 4.7±0.4 Stand 7.6±0.1 7.4±0.1 7.0±0.1 8.0±0.2 Resistance to Fusarium 8.2±0.03 8.0±0.1 8.2±0.1 8.0±0.2 Protein concentration (%) 25.8±0.2 25.6±0.2 26.4±0.2 26.0±0.6

29 Lathyrus Lathyrism Newsletter 2 (2001)

Fig. 1. Phenograph of the tested traits, created by the cluster analysis. Num denotes the individual genotypes, 35 is of Syrian origin. Clusters I to IV in order moving down the page.

Rescaled Distance Cluster Combine

Discussion underutilized and neglected crops. 18. Institute of At present there is little data about this nearly Plant Genetics and Crop Plant Research, “forgotten” crop. The evaluated set contained mainly Gatersleben/International Plant Genetic Resources landraces and one genetic resource of Syrian origin. Institute, Rome Thus the set was quite homogenous. During a 2 year 2. FAO Yearbook, Production Vol.52. 1998. FAO, period we found that the Syrian genetic resource had Rome. the lowest value of 1000 seed weight (TGW) and the 3. Fox P.N., Rosielle A.A. (1982). Reducing the lowest yield. This is in agreement with Gáborcík (4) influence of environmental main–effects on who found that TGW of Slovak genotypes was up to pattern analysis of plant breeding environments. twice as high, with yield 1.4 times greater, than foreign Euphytica 31, 645-656. genotypes. We found positive correlations among the 4. Gáborcík N., Pastucha L. (1995). Príspevok cícera number of pods per plant and the traits: number of baranieho a hrachora siateho k ekologizácii seeds per plant and plant weight, which corresponds polnohospodárstva. Plant genetic resources, SPU with the results of Cambell (1), according to his Nitra, Annual report 1995, 23 – 27. statement that seeds of high weight have set more pods 5. Hruška J., Luskoviny S.Z.N. (1956). Legume. per plant. Number of seeds per plant correlated with SZN, Praha. the seed yield, which again agrees with results of 6. Tryon R. C. (1959). Domain sampling Cambell (1). The genotype H-17 was markedly formulation of cluster and factor analysis. different to the others. Although it was not tall, it Psychometrika 24, 113-135 showed the highest values of yield, seed weight per 7. Uberla K. (1976) Faktorová analýza. ALFA, plant, number of seeds and number of pods per plant. Bratislava. 8. Vavilov N.I. (1957). Mirovye resursy sortov References chlebnych zlakov, zernovych bobovych, lna i ich 1. Campbell C.G. (1997). Grass Pea. Lathyrus ispolzovanie v selekcii. Izd. AN ASSR, Moskva - sativus L. Promoting the conservation and use of Leningrad.

30 Lathyrus Lathyrism Newsletter 2 (2001)

Morphological characterisation of The characterisation site was located in Alcalá de Spanish genetic resources of Henares (Madrid) in 40º31’N, 3º17’W and 610 masl. Lathyrus sativus L. Each accession was sown in an experimental plot of 5 1 rows 3 m length, 35-40 cm between rows, with 150 L. De la Rosa and I. Martín. seeds/accession. Observations were recorded about the whole plot for the phenological characters and flower Centro de Recursos Fitogenéticos, INIA. Apdo. and seed colour. The mean of ten plants, pods or seeds 1045, 28800 Alcalá de Henares, Madrid, Spain. was calculated for the metric traits.

Email: (1) [email protected] Pearson correlation was performed to determine the interrelationships between traits. Principal Component Analysis (PCA) was utilised to show the patterns of Introduction covariation of quantitative variables among Grass pea (Lathyrus sativus L.) is known in Spain as accessions. Statistical calculations were made using “almorta”, “muela”, “tito”, “guijo” or “chícharo”, and the SPSS package. was probably one of the first cultivated plants. At present, in Spain, this species remains a marginal crop Results and Discussion in some areas around all of the country. In 1996 the Table 1 summarises the results of field area dedicated to grass pea was only 273 ha, with a characterisation for quantitative characters. For each production of 174 t of grain and 52 t of straw. In Spain trait, mean, standard deviation, maximum and there are no commercial varieties of L. sativus and for minimum value about the total accessions is recorded. this reason the seed sown belong to traditional local These results show the existence of high variability in types. plant, pod and seed traits, in a similar range of variation to that in a collection that included the The Spanish Plant Genetic Resources Centre (CRF), known geographical distribution of L. sativus (7). The belonging to the National Institute of Agriculture and populations evaluated in this study were later and their Food Research Technology (INIA) and preserves a range of variability for phenological data was lower total of 445 accessions of genus Lathyrus, mainly L. that those evaluated in a collection of Mediterranean cicera and L. sativus, and also some accessions of L. sativus lines (6). The lesser variation for these traits other wild species including L. amphicarpos and L. can be as a consequence of the more restricted (1) annus . geographical origin of the CRF collection. Furthermore, Spanish grass pea exhibited bigger seeds Material and Methods that Mediterranean lines mentioned above; the larger The CRF L. sativus collections consist of 150 seed size is probably derived from their preferential accessions, 132 corresponding to Spanish landraces. use as grain. Sixty of them were characterised for the following agro/morphological traits from November 1994 to Table 2 shows the percentage of each class for flower July 1995: and seed colour. The coloured flower category includes completely blue flowers and white ones that DFF: Days to first flowering. DF: Days to 50% flowering. have some blue or pink pigmentation on the standard DEF: Days to end of flowering. petal, lateral petal or both. Although grass pea is DM: Days to maturity. reported as an autogamous species, the great PLH: Plant height (cm). FPOH: First pod height (cm). percentage of accessions with a mixture for seed and STPL: Number of primary stems per plant. flower colour could be due to outcrossing. Data of PEDL: Peduncle length (cm). hereditability of flower colour (3) and isozymes POPL: Number of pod per plant. (4) POL: Pod length (mm). heterozygosity indicated a variable level of POW: Pod width (mm). outcrossing depending of environmental conditions. SEPO: Number of seeds per pod. Field observation in our experimental plot confirm this SEL: Seed length (mm). SEW: Seed width (mm). aspect, for this reason we use spatial isolation for seed SET: Seed thickness (mm). increase of landraces of grass pea in order to maintain 100SEW: 100-seed weight (g ). the genetic integrity of individual accessions. FC: Flower colour. SEC: Seed colour.

31 Lathyrus Lathyrism Newsletter 2 (2001)

Table 1. Mean, standard deviation, maximum and minimum for the quantitative evaluated traits (see text for the explanation of descriptor codes).

Descriptor Mean Std Dev Max Min DFF 139.95 2.63 147.00 136.00 DF 147.18 2.57 154.00 143.00 DEF 191.81 2.44 197.00 187.00 DM 209.57 5.82 222.00 200.00 PLH 53.35 8.15 77.10 39.30 POH 15.18 3.69 27.70 9.30 STPL 2.46 0.42 4.50 1.90 PEDL 3.69 0.52 4.93 2.67 POPL 29.37 6.52 46.80 17.40 POW 15.03 0.94 17.00 11.90 POL 42.08 3.07 52.45 34.50 SEPO 2.94 0.33 4.00 2.20 SEL 8.95 0.97 11.04 6.47 SEW 9.00 0.70 10.57 7.15 SET 4.79 0.29 5.71 4.21 100SEW 25.77 3.92 36.80 13.80

Table 2. Class frequency of grass pea non metric characters.

Descriptor Class Frequency (%) Flower colour (FC) White 16.6 Coloured 43.3 Mixture 40.3

Seed colour (SC) Light 23.3 Light with blue or black line 36.6 Light with wide black ornamentation 1.6 Mixture 38.3

Correlations between metric characters were first flowering with days of end flowering and days to calculated (Table 3), many of them were as expected. maturity, days of end flowering with first pod height, The strongest positive correlations were between days number of pod per plant with seed length and width to first flower with days to 50% flowering, days to end and 100 seed weight with number of seed per pod. flowering with days to maturity, plant height with first Similar correlations were found in a collection of pod height, peduncle length with plant height and first Spanish faba bean germplasm (8). Seed and flower pod height, number of primary stems per plant with colour data were not included in correlation analysis number of pod per plant, pod width with pod length, but, as reported elsewhere (7), in this species there is an seed length with seed width and 100 seed weight with association between less coloured seed and flowers. seed length and width. Significant negative The Spanish landraces, probably more domesticated correlations (P< 0.05) were found between days to forms, contain a low ODAP content (5).

32 Lathyrus Lathyrism Newsletter 2 (2001)

Table 3. Correlation coefficient between quantitative traits (see text for the explanation of codes). Coefficients not shown were not significant at P<0.05.

100 DFF DF DEF DM PLH FPOH STPL PEDL POPL POL POW SEPO SEL SEW SET SEW DFF 1.00 DF 0.73 1.00 DEF -0.32 - 1.00 DM -0.30 - 0.51 1.00 PLH ----1.00 FPOH ---0.39-0.62 1.00 STPL -0.26---0.370.421.00 PEDL ---0.32-0.56 0.55 -1.00 POPL ----0.440.360.59 0.34 1.00 POL ------1.00 POW -----0.27---0.61 1.00 SEPO ------0.26--0.28--1.00 SEL ------0.25--0.38---1.00 SEW ---0.30-----0.36---0.82 1.00 SET ------0.27-----0.331.00 100 ---0.25-----0.35---0.320.86 0.82 - 1.00 SEW

Table 4. Correlation of the analysed traits with the three first principal axes F1, F2 and F3 (see text for the explanation of codes).

F1 F2 F3 DFF 0.106 -0.494 0.676 DF 0.111 -0.453 0.629 DEF 0.393 0.135 -0.627 DM 0.177 0.562 -0.362 PLH -0.608 0.480 0.025 FPOH -0.645 0.394 0.274 STPL -0.560 0.391 -0.022 PEDL -0.551 0.422 0.247 POPL -0.703 0.053 0.160 POL -0.103 0.501 0.172 POW -0.011 0.616 0.130 SEPO 0.009 -0.140 -0.410 SEL 0.736 0.427 0.344 SEW 0.696 0.533 0.264 SET 0.375 0.090 -0.269 100SEW 0.687 0.474 0.414

Eigen value 3.75 2.85 2.18 % Variation explained 23.47 17.82 13.65 % Variation accumulated 23.47 41.30 54.96

33 Lathyrus Lathyrism Newsletter 2 (2001)

Results of PCA were summarised in Table 4, which legumes are a very interesting option and grass pea shows the correlation of each character with the three could play an important role in some regions of principal components, the percentage of variation Southern Europe. explained by this components, the eigen values and the variability explained and accumulated by the 3 PC. F1 explained 23.47% of variability. In this axis, References the traits with the most important contribution were 1. De la Rosa, L., Martín, I., Varela, F. La colección related to plant architecture (plant height, first pod de Lathyrus del CRF-INIA. Monografías del height and pod per plant) and to seed characters (100 INIA, serie Agrícola, INIA, Madrid (in press) seeds weight and seed length and width), these two 2. De la Rosa, L., Martín, I., Varela, F. (1999). La groups of variables showing an inverse relation. F2 colección de algarrobas (Vicia articulata (17.82% of variation) was mainly loaded by traits Hormen) del Centro de Recursos Fitogenéticos related to pod size. F3 (13.65% of variability) was del INIA. Invest. Agr.: Prod. Prot. Veg. 14(3), positively correlated with the phenological characters 367-381. days to first flowering and days to 50% of flowering 3. Chowdhury, M.A., Slinkard, A.E. (1997). Natural and correlated negatively with days to end of outcrossing in grasspea. The Journal of Heredity flowering. Similar results of variability distribution 82(2), 154-156. have been found for a Spanish collection of Vicia 4. Chowdhury, M.A., Slinkard, A.E. (2000). Genetic articulata (2). diversity in grass pea (Lathyrus sativus L.). Genetic Resources and Crop Evolution 47, 163- The total amount of variability accounted for the three 169. principal components was 54.96%; this low 5. Franco, F. (1996). El lathyrus . In “El cultivo de percentage indicates that traits did not show a strong las leguminosas de grano en Castilla y León.” association, due probably, to the scarce breeding work F.Franco and A. Ramos (Eds) Junta de Castilla y made in Spanish populations of grass pea. In a similar León. Valladolid (Spain) 209-213. study (9) of grass pea accessions, collected in the 6. Hanbury, C.D., Siddique, K.H.M., Galwey, N.W., Marche region of Italy, the accumulated variation in Cocks, P.S. (1999). Genotype-environment the first 3 principal components was 91.74%. This interaction for seed yield and ODAP high difference in percentages of variation could be concentration of Lathyrus sativus L. and L. cicera explained by the different provenance of accessions; L. in Mediterranean-type environments. Italian material was collected in a small area whereas Euphytica 110, 45-60. our landraces were gathered in a wider range of 7. Jackson,M.T., Yunus, A.G. (1984). Variation in places. the grass pea (Lathyrus sativus L.) and wild species. Euphytica 33, 549-559. 8. Suso, M.J., Moreno, M.T., Cubero, J.I. (1993). In spite of the problem of lathyrism, which has Variation among Spanish faba bean cultivars: restricted the exploitation of L. sativus, the variability taxonomy and evolutionary implications. Genetic found in our collection for plant architecture and seed Resources and Crop Evolution 40, 105-111. traits should be taken into account for future breeding 9. Tavoletti, S., Capitani, E. (2000). Field evaluation programs in this species. In a moment when Europe is of grass pea populations collected in the Marche looking for diversification of cropping systems and for region (Italy). Lathyrus Lathyrism Newsletter 1, more environmental friendly agriculture, the grain 17-20.

34 Lathyrus Lathyrism Newsletter 2 (2001)

Protoplast, cell and tissue cultures Materials and Methods Embryo axes from seeds of L. sativus genotypes LB, for the biotechnological breeding of LIII and L12 were germinated on hormone-free grass pea (Lathyrus sativus L.) medium and hypocotyl segments without any pre- existing meristem were used as initial explants as S. Ochatt1, P. Durieu, L. Jacas and C. reported (6). Pontécaille Bud regeneration media were supplemented with Laboratoire de Physiologie et Culture in Vitro, BAP + NAA, with thidiazuron alone, or with zeatin + URGAP, INRA, CR de Dijon, B.P. 86510, 21065 ABA, as used with pea (6). Regenerated shoots were Dijon cedex, France grown on the germination medium described above prior to rooting on half-strength hormone-free MS Email: (1) [email protected] medium (5). Rooted plants were weaned in the greenhouse (6), and kept until they set seed. To be presented at the forthcoming 4th European Experiments were repeated thrice (≥10 replicates per Conference on Grain Legumes to be held in medium per genotype). Results, expressed as the Cracow, Poland, July 2001. mean regeneration percentage and number of shoots/explant, were statistically analysed with Newman-Keuls test Introduction Regenerants were compared with seedlings in terms Grass pea consumption leads to neurolathyrism and a of phenotype, and they were analysed by flow safer use of this environmentally adaptable crop cytometry (7), with two independent DNA content needs cultivars with decreased neurotoxin levels. analyses/sample, compared to leaf controls from Were they cross-compatible, grass peas would be seedlings of each genotype. useful genetic resources for disease resistance breeding of peas (Pisum sativum L.). In vitro Protoplasts were isolated from leaves of all genotypes selection (9) and gene transfer may provide grass peas and cultured in the media tested for pea (7). They were better adapted for consumption, while interesting labelled with fluorescein diacetate (green), while Lathyrus traits could be introduced into pea by rhodamine B isothiocyanate (red) was preferred for somatic hybridisation (2, 4). pea protoplasts (2), and electrofusion (at 750, 1000, 1250, 1500 or 2000 V/cm) was compared with The main bottleneck in tissue cultures of Lathyrus chemical fusion. The standard macro-method (3) was sativus has been plant regeneration (8, 9). Cell compared with a micro-method we developed (2), and suspensions were established from which non- both were tested with glycine, PEG 6000 and PEG dividing protoplasts were isolated (4), and the 1540. Heterokaryon viability was assessed and they sustained division of leaf protoplasts and their fusion were cultured in various media. with pea protoplasts was reported (2). Hyperhydricity (vitrification) has been observed in cultured tissues of Results many species. Their morphology was described and True-to-type plants were regenerated and the whole the influence of various physiological parameters procedure, from hypocotyl explants to grains reviewed (1), but information on the fundamental harvested from regenerants, took 4 (LB) to 5 months genetic mechanism(s) underlying the occurrence of (LIII and L12). The optimum hormonal combination hyperhydricity is practically non-existent. This article for regeneration was genotype-specific. LB (white- describes strategies for in vitro regeneration of fertile seeded) responded best on auxin-free medium with L. sativus plants, it examines the genetic background 5.0 mg/l BAP, while coloured-seeded genotypes concomitant with the appearance of hyperhydricity responded best with 0.01 mg/l NAA, plus 5.0 mg/l and, also, reports the isolation and culture of grass BAP for LIII, but 3.0 mg/l BAP for L12 (Table 1). pea leaf protoplasts, and their fusion with pea Rooting was optimum on half-strength hormone-free protoplasts to give somatic hybrid calluses. MS medium, but differences appeared within genotypes between shoots regenerated on different hormonal balances.

35 Lathyrus Lathyrism Newsletter 2 (2001)

Table 1. Mean caulogenesis and hyperhydricity related to the regeneration medium

Genotype Cytokinin % regenerationc buds/explantc hyperhydric (mg/l) (mean number) shootsc (%) LB BAP (1.0)a 0 f 0 f 0 f BAP (3.0)b 20.8 b 1.9 c 44.8 c BAP (5.0) b 45.83 a 3.25 a 74.8 b TDZ (2.2) 10.4 c 2.5 b 84 b Zeatin (10.0) 0 f 0 f 0 f

L III BAP (1.0) a 0 f 0 f 0 f BAP (3.0)b 6.125 d 2.0 c 0 f BAP (5.0) b 12.25 c 1.25 d 0 f TDZ (2.2) 1.02 e 1.0 e 0 f Zeatin (10.0) 0 f 0 f 0 f

L 12 BAP (1.0) a 16 bc 1.625 cd 7.74 e BAP (3.0) b 52 a 1.345 d 20 d BAP (5.0) b 14 bc 3.0 a 14.3 d TDZ (2.2) 2 e 1.0 e 100 a Zeatin (10.0) 2 e 1.0 e 100 a a, containing 0.0, 0.01 or 0.05 mg/l NAA b, containing 0.0, 0.01, 0.05, 0.1 or 0.5 mg/l NAA c, figures within a column followed by different letters differed significantly (P < 0.05)

Shoots with severe hyperhydricity symptoms were clear effect of cytokinins was when BAP was produced on media with low NAA (0.01 or 0.05 compared to TDZ, the latter provoking more mg/l) plus high (5.0 mg/l) BAP, or when TDZ or hyperhydricity. zeatin and no auxin was added. Such shoots were non-rootable, while non-hyperhydric shoots rooted Large numbers of viable protoplasts were obtained readily, giving fertile plants. Hyperhydricity never from the leaves of all genotypes, and they occurred with LIII, whatever the hormones used for proliferated to give calluses, but plants have not been regeneration, while regenerants of LB and L12 were regenerated to date. In terms of fusion with pea very prone to hyperhydricity (Table 1). When protoplasts, glycine was the least effective agent (~10 regenerated shoots were analysed by flow cytometry, % heterokaryons), while PEG was best (>20 % hyperhydricity was associated with abnormal heterokaryons). With electrofusion, heterokaryon profiles, with three peaks corresponding to 2C, 4C formation was increased from 750 to 1500 V/cm, but and 8 C DNA levels (Fig. 1). Conversely, all fell drastically beyond this threshold. In all, phenotypically normal regenerants had a diploid electrofusion permitted the largest heterokaryon DNA content (profiles with 2C and 4C DNA peaks), formation but with a large variability, and PEG 6000 comparable to seedling tissues. The time in culture appeared as the most efficient and reproducible did not modify DNA content, nor did it affect the fusion agent (Table 2). Heterokaryons fluoresced occurrence of hyperhydricity. Auxins had a stronger both green and red under UV light, and divisions effect than cytokinins, an increase in BAP having no could be monitored during several days (up to 4-6 effect on hyperhydricity or DNA content. The sole celled colonies). Several hundreds of calluses have addition of NAA to a medium with hitherto high been obtained from the fused protoplasts and BAP levels sufficed to induce hyperhydricity and regeneration experiments are under way. increased the DNA content of regenerants. The only

36 Lathyrus Lathyrism Newsletter 2 (2001)

2C 4C Hyperhydric shoot regenerated 8C on 5 BAP / 0.01 NAA

2C 4C 8C Hyperhydric shoot regenerated on 5 BAP / 0.05 NAA

2C

4C Normal shoot regenerated on 5 BAP

2C 4C Hyperhydric shoot regenerated on 8C TDZ

Figure 1. Flow cytometric profiles of LB regenerants on different regeneration media

Table 2. Plating efficiency of Pisum (+) Lathyrus heterokaryons (mean data from ≥200 heterokaryons/treatment and 3 independent experiments)

Fusing agent Fusion method Heterokaryon IPE FPE formation(%) (%) (%) Non-fused LB protoplasts NANA 63.28 2.15 Non-fused LIII protoplasts NANA 37.55 1.44 Non-fused L12 protoplasts NA NA 29.22 0.58

Glycine Micro 11.4 16.12 0.0 Macro 9.2 19.25 0.0

PEG 1540 Micro 20.6 23.5 2.15 Macro 19.7 4.65 0.22

PEG 6000 Micro 21.3 28.0 2.75 Macro 22.5 7.4 0.42

Electrofusion 750 10.1 21.81 1.86 (V/cm) 1000 13.9 19.46 1.42 1250 18.8 22.3 2.85 1500 22.1 25.0 3.11

NA: not applicable; IPE: % heterokaryons dividing once; FPE: % heterokaryons undergoing sustained division

37 Lathyrus Lathyrism Newsletter 2 (2001)

Discussion Plant regeneration competence was strongly Acknowledgements. This work is part of Research genotype-dependent, with LIII more recalcitrant than Agreement 10420/RO, within the FAO/IAEA Co- LB, and L.12 intermediate in responses. Production ordinated Research Programme “Genetic of rootable shoots before was rare (8,9), with a very improvement of under-utilised and neglected crop sin specific and narrow growth regulator requirement (8). LIFDCs through irradiation and related techniques” However, when our regenerants had a normal (1998-2003). phenotype and a normal DNA content, rooted and fertile plants were consistently and reproducibly obtained on hormone-free medium. References A clear relationship had been established in the past 1. Debergh P., Aitken-Christie J., Cohen D., Grout between the hormones in regeneration media and B., von Arnold S., Zimmerman R., Ziv M. (1992). hyperhydricity of shoots recovered (1), in line with our Reconsideration of the term "vitrification" as used in results. Additionally, here, flow cytometry permitted micropropagation. Plant Cell Tiss Organ Culture 30, correlation of an abnormal DNA content with the 135-140. hyperhydricity of regenerants and with a reduced 2. Durieu P., Ochatt S. (2000). Efficient intergeneric competence of such shoots for subsequent rooting fusion of pea (Pisum sativum L.) and grass pea and fertile plant production. A similar correlation was (Lathyrus sativus L.) protoplasts. J Exp Bot 51, 1237- observed between the media used and the 1242. regeneration of fertile plants from pea hypocotyls (6), 3. Kao K.N., Constabel F., Michayluk M.R., while the regeneration of fertile protoplast-derived Gamborg O.L. (1974). Plant protoplast fusion and plants has been correlated with the absence of growth of intergeneric hybrid cells. Plant Sci 120, endoreduplication in calluses derived from the 215-227. cultured protoplasts (7). 4. McCutchan J.S., Larkin P.J., Stoutjesdik P.A., Morgan E.R., Taylor P.W.J. (1999). Establishment of Hyperhydricity was more strongly linked to the shoot and suspension cultures for protoplast isolation presence of auxin and to the auxin/cytokinin balance, in Lathyrus sativus L. SABRAO J Breed Gen 31, 43- than to the cytokinin level or the number of 50. subcultures, contrasting claims that a prolonged 5. Murashige T., Skoog F. (1962). A revised medium period of culture may result in hyperhydricity (1). Roy for rapid growth and bioassays with tobacco tissue et al (8) also underlined the paramount influence of cultures. Physiol Plant 15, 473-497. hormone interactions on the regeneration of rootable 6. Ochatt S., Pontécaille C., Rancillac M. (2000a). shoots of Lathyrus, but provided no explanation for The growth regulators used for bud regeneration and the fundamental mechanism underlying this shoot rooting affect the competence for flowering and phenomenon. In addition, the difference observed in seed set in regenerated plants of protein peas. In susceptibility to hyperhydricity depended on the type Vitro Cell Dev Biol – Plants 36, 188-193. of cytokinin used and on the genotype studied, such 7. Ochatt S., Mousset-Déclas C., Rancillac M. genotype-specific responses being commonplace in (2000b). Fertile pea plants regenerate from legumes (6, 7). protoplasts when calluses have not undergone endoreduplication. Plant Sci 156, 177-183. The availability of a reliable regeneration technique 8. Roy P.K., Barat G.K., Mehta S.L. (1992). In vitro for the production of fertile plants may help in the plant regeneration from callus derived from root breeding of Lathyrus sativus itself, while explants of Lathyrus sativus. Plant Cell Tiss Organ regenerating somatic hybrids might yield genotypes Culture 29, 135-138. with the disease resistance from grass pea coupled 9. Van Dorrestein B., Baum M., Abd El Moneim A. with the grain quality from protein pea. These results, (1998). Use of somaclonal variation in Lathyrus added to those obtained with pea (7), stress the interest sativus (grasspea) to select variants with low β- of flow cytometry as an early screening strategy to ODAP concentration. Proc 3rd Eur Conf Grain avoid hyperhydricity in cultured tissues, and for the Legumes, 14-19 November, Valladolid, Spain. AEP optimisation of plant regeneration in grain legumes. (eds). p. 364.

38 Lathyrus Lathyrism Newsletter 2 (2001)

Progress in isolation and purification Lathyrus sativus is primarily a self-pollinated crop (see Ben Brahim et al. this issue) and breeding programmes of Lathyrus sativus breeding lines have been followed accordingly. However, the natural outcrossing in this species is at a higher frequency than acceptable in a truly self-pollinated crop (4). The extent of 1 M. Anisur Rahman , M. Matiur Rahman outcrossing varies according to the flower colour and has and M. Akhtaruzzaman Sarkar been recorded as high as 27.8% (3). This high frequency of natural outcrossing has meant that the improvement programme will not be very reliable, as it will be difficult Pulses Research Centre, Bangladesh Agricultural to maintain varietal purity in farmers’ fields because of Research Institute (BARI), Ishurdi-6620, Pabna, outcrossing with unimproved plants nearby. Therefore, Bangladesh. breeding steps need to be taken to overcome the outcrossing problem by developing a visible morphological marker such as white, pink or red flowers (1) [email protected] in low toxin varieties. It has been observed in some studies that blue-flowered lines had relatively lower BOAA contents (2). It may be possible to associate high Introduction level of BOAA with white, blue or red flowered lines. Lathyrus sativus L. is the most important pulse crop in Therefore, effort should be made to initiate this research Bangladesh, where it is popularly known as “khesari”. by isolating and purifying L. sativus breeding lines. Among the pulses L. sativus occupies first position both in area and production and contributes about 35% of the Research underway and results total pulse production in Bangladesh. The national The segregating generation developed so far in F6, F7 and average yield of L. sativus is 750 kg/ha (1). In Bangladesh other breeding lines have shown mixtures of flower the crop is generally grown in relay with Aman rice. colour. Therefore, the PRC have planned to purify the However, it is also grown as a forage crop in some parts segregating lines along with other advanced lines with of the country where cattle are allowed to graze the crop. controlled pollination to develop homozygous pure lines based on flower colour. Attempts have been made to ban its cultivation in Bangladesh due to its toxic effect in human beings. In 1997-98, a total of 78 lines (7 lines from F6, 28 lines Lathyrus sativus seed contains a free amino acid known from F7, 20 lines from observation trial, 16 lines from as β-N oxylaminoalanine (BOAA or ODAP) which can preliminary yield trial and 7 lines from multilocation cause the neurological disorder called “lathyrism” (5). yield trial) were grown. Seeds from 10 plants from each Despite this problem, there is no alternative to L. sativus of the 78 lines were harvested separately. In 1998-99, all at present since it is a very hardy crop that tolerates these seeds were grown as plant to row at Pulses adverse environmental conditions such as drought and Research Centre, Ishurdi, Bangladesh with 50 x 50cm excessive soil moisture. The crop produces yields with plant spacing. Initially 2 seeds were sown per hill and at little or no inputs such as fertilizer and chemicals. seedling stage 1 plant per hill was maintained. At 1-2 Therefore, farmers choose to grow this crop extensively. flowering stage, each and every plant was observed for The Pulse Research Centre (PRC) has taken initiatives to flower colour and a mixture of flower colour were solve the neurotoxin problem, developing two varieties noticed within each breeding line. The flower colour of a in 1995 and 1996 with high yield potential and lower breeding lines was decided by seeing the majority of the BOAA content. The varieties and neurotoxin contents plants having similar flower colour. Variable numbers of are found in Table 1. healthy plants within each breeding lines of similar flower colour were bagged with nylon mosquito nets of 5 2 Table 1. Two L. sativus varieties released by the cm x 30 cm size (mesh 6 holes per cm ). A stick of 30 Pulses Research Centre, Bangladesh and the mean cm length was used to support the net and the plant. The BOAA (or ODAP) contents. flowers those already bloomed were removed from each plant during bagging. At maturity seeds from all the Varieties BOAA content bagged plants were harvested and packed separately. (%) Altogether 1779 plants of blue, pink, red and white Barikhesari-1 0.29 flower colour were harvested in 1998-99 (Table 2). Barikhesari-2 0.27 In 1999-2000, all of those 1779 plants were grown again as plant to row in the similar way to the previous year. Plants were bagged with the same nylon mosquito net. It

39 Lathyrus Lathyrism Newsletter 2 (2001) was observed that most of the breeding lines had almost will be used to study the inheritance pattern of flower attained homozygosity. At the end of the season 1872 colour and be used simultaneously to develop varieties. plants in total were harvested individually. References 1. Bangladesh Bureau of Statistics. (1987). Year Book Table 2. The number of plants selected from each of Agricultural Statistics for 1985/86. Dhaka, population in 1998-99. Bangladesh : BBS. 2. Quader, M., Ahad, M., Wahiduzzaman, M, Rahman, Populations Flower colour No. of plants M. (1988). Present status of Lathyrus sativus L. harvested research in Bangladesh. Presented at the conference F6 Blue/pink/red 159 on Lathyrus sativus L. and the eradication of F7 Blue/pink 627 lathyrism. 2-6 May 1988. London, UK. : Ciba Observation trial Blue/white/red 485 foundation. Preliminary yield trial Blue 334 3. Rahman, M.M., Kumar, J., Rahman, M.A., Ali Multilocation yield trial Blue/pink 174 Afzal, M. (1995). Natural outcrossing in Lathyrus Total 1779 sativus L. Indian J. Genet. 13, 204-207. 4. Smartt, J. (1984). Evaluation of grain legumes. 1. Mediterranean Pulses. Experimental Agriculture 20, These 1872 will be tested for their BOAA content. The 275-296. low BOAA (or zero if found) content line(s) will be used 5. Spencer, P.S., Roy, D.N., Palmer, V.S., Dwivedi, as parents for crossing with the adapted cultivars or M.D. (1986). The need for a strain lacking human varieties to develop varieties for wider cultivation. and animal neurotoxic properties. Pages 297-305 in L. sativus and Lathyrism (Kaul, A.K. and Combes, Meanwhile, this year (2000-2001) these 4 different D., eds.) New York, USA: Third world Medical flower coloured lines (white, blue, red and pink) have Research Foundation. already been planted in the field for hybridization. They

40 Lathyrus Lathyrism Newsletter 2 (2001)

Mutants of grasspea (Lathyrus sodium azide (NaN3). Biological injuries of M1 plants observed in the field were a measure of reaction to the sativus L.) obtained after use of mutagens. Selection of morphological mutants was chemomutagens. conducted on the M2 progeny.

The measured morphological traits and yield showed Wojciech Rybinski that the sensitivity of M1 plants depended on the mutagen, dose concentration and genotype. Sodium Institute of Plant Genetics, Strzeszynska 34, 60-479 azide induced a lower level of injuries compared to Poznan, Poland. MNU. The cultivar Derek showed a higher susceptibility to MNU than the cultivar Krab, the Email: [email protected] opposite was observed for NaN3.

In M2 a wide spectrum of chlorophyll mutations To be presented at: Eucarpia - Genetic Resources (albina, xantha, chlorina and other types) were Section Symposium to be held at the Institute of observed. Morphological mutants showed changes in Plant Genetics, Poznan , Poland, 16-20 May, 2001. stems, leaves, flowers and seeds. Particularly interesting were semi-dwarf and dwarf mutants, although many were completely sterile. Others Abstract flowered and ripened early compared with control The growing of grasspea (Lathyrus sativus L.) has a plants. All mutants could be distinguished by yield long tradition in Europe (2). The recent production of characteristics, 1000 seed weight was particularly new genotypes with a low content of neurotoxin has affected. The mutants will be retained in our plant opened new possibilities of broader use of grasspea in collection and those with desirable agricultural traits many breeding programmes as well as for feeding will be used for breeding purposes. purposes (1). In Poland, the grasspea is an agricultural plant of marginal importance. However, in recent years the popularity and importance of this crop has increased. One of the factors limiting the use of References grasspea in breeding is the narrow genetic variability 1. Campbell C.G., Briggs C.J. (1987). Registration of traits (3). Mutagenesis can be used to create of low neurotoxin content Lathyrus germplasm additional genetic variability that may be utilised by LS 8246. Crop Sci 27, 821. plant breeders in the development of cultivars for 2. Przybylska J. (1999). The grasspea (Lathyrus specific purposes or with specific adaptation. sativus L.) as a potential valuable pulse crop. Postepy Nauk Rolniczych 1/99, 33-43. The initial material used for mutagenic treatment 3. Singh M., Chaturvedi S.N. (1987). Effectiveness constituted two Polish cultivars- Krab and Derek. The and efficiency of mutagens or in combination seeds were treated with different doses of two with dimethyl sulphoxide in Lathyrus sativus chemomutagens: N-methyl-N-nitrosourea (MNU) and Linn. Ind J Agric Sci 57, 503-507.

41 Lathyrus Lathyrism Newsletter 2 (2001)

Development of low ODAP variation in phenotype, as well as ODAP content and yield characteristics were observed among the somaclones of Lathyrus sativus. regenerated plants. Phenotypic variation was observed in flower colour, seed coat colour, leaf size and morphology, pod morphology and seed weight. Analysis 1 1,2 of seeds of about 100 R1 generated plants for seed ODAP I.M. Santha and S.L. Mehta showed it to vary from 0.02 to 0.80%. From these plants those having low ODAP and high yielding characteristics (as compared to the parent cultivar P-24) were selected 1. Division of Biochemistry, Indian Agricultural and advanced to further generations. ODAP content and Research Institute, New Delhi 110 012, India. yield characteristics of these low toxin lines were stable 2. Present address: Deputy Director (Education), over generations. ICAR, ICAR Bhavan, New Delhi 110 012, India. A few of the very promising somaclones were tested in various field trials at different locations in India. Based Email: (2) [email protected] on these trials one of the somaclones Bio L212 (Ratan) was released by the Government of India for cultivation in the North Eastern Plain and Central zones. Bio L212 or Bio L12 (Ratan) was developed from the leaf explant of P-24 and differs from its parent. It has extremely low Abstract ODAP content (0.05%) coupled with high yield potential (1800 kg/ha), large seeds (100 seed weight of 9.29g Lathyrus sativus is a grain legume high in protein compared to P-24 of 6.94g) and high biomass (almost (25-30%) that can be grown under limited moisture 150% of P-24). It also has remarkable vegetative growth, conditions. However, the Government of India has characteristic pod marking and long, broad leaves which imposed a ban on its sale because of the association distinguishes it from its parent in the field. The average with human neurolathyrism in the population seed yield of Bio L212 in our experimental farm is 2340 subsisting on it. Over the years, by mutation and ± 170 kg/ha. conventional breeding, attempts were made to reduce the amount of oxalyl diamino propionic acid In total 18 somaclones were selected for very low ODAP (ODAP), the chemical implicated in neurolathyrism. content (<0.1%). These low ODAP containing Although a decrease in ODAP was obtained, it was somaclones have now been fully characterised at the not stable. Therefore, biotechnological approaches biochemical and molecular level using isozyme analysis, offered the scope to develop strains that were either RAPD, RFLP, mitochondrial specific genes and devoid of ODAP or contained low levels which were photosynthetic efficiency. All these somaclones have safe even with prolonged consumption. Intially phenotypic distinguishing features. Two of the reproducible in vitro regeneration protocols were not somaclones are white flowered with a white seed coat. A available. However, successful regeneration protocols few of the somaclones are now being further tested in were developed using leaf, root and internode different locations. The phenotypic variability as explants of L. sativus cultivar P-24. About 300 in observed will be a boon for plant breeders in further vitro regenerated plants were taken to the field, of improving quality and yield. which 102 survived and set seed. A wide range of

42 Lathyrus Lathyrism Newsletter 2 (2001)

Isozymes variability of grasspea and to assess diversity, which of course is important in germplasm collection and conservation. This study was (Lathyrus sativus L.) in Ethiopia. carried out to determine the isozyme variability of in Ethiopian grasspea populations. Wuletaw Tadesse 1 and Endashaw Bekele2 Materials and Methods 1. Adet Research Center, P.O. Box 08, Bahir Dar, Isozyme analysis was done in ten accessions of grasspea Ethiopia. selected from five clusters based on morphological 2. Addis Ababa University, Department of Biology, diversity, i.e. two populations from each cluster were P.O. Box 1176, Addis Ababa, Ethiopia. used. The selection of the two populations from each cluster was based on diversity index, origin and ODAP content. Twenty-five seedlings per accession were Introduction studied. Three and seven days old leaf samples were Grasspea (Lathyrus sativus L.) is the third important compared for extraction and better resolution. The three pulse crop after faba bean and chickpea in Ethiopia day old leaf gave better resolution and hence these were (CSA, 1998). It is grown in the off season (planted in used for extraction. Crude extracts were prepared by September/October and harvested in January/February) macerating leaves in two drops of extraction buffer on residual moisture in vertisols at altitudes ranging (0.05M sodium phosphate, pH 7.0, plus 0.2M 2- from 1600-2700 m.a.s.l., across the different mercaptoethanol). The perspex extraction trays were administrative regions of the country. The ability of kept on crushed ice during maceration to prevent grasspea to tolerate both drought and flooding denaturation of the enzymes. Extracts were absorbed on conditions, the low cost required for its production, its to wicks made from Whatman 3MM chromatography high protein content and capacity to ameliorate soil paper. Horizontal electrophoresis was carried out in fertility make it an important subsistence and insurance 12% starch gels. crop of the poor family (2). Two buffer systems were used: Excessive consumption of grasspea seeds by humans 1. Lithium borate buffer pH 8.3: the gel buffer for this can result in an irreversible disorder of the lower limbs, system contains 5.4 g tris base and 1.28 g known as “neurolathyrism”, caused by the toxin oxalyl- anhydrous citric acid. The electrode buffer contains diaminopropanoic acid (ODAP) present in the seeds (6,9). 1.2 g lithium hydroxide and 11.9 g boric acid (pH Identification of varieties with low ODAP content is the 8.3). primary objective in grasspea improvement for which a 2. Histidine tris citrate buffer pH 7.5: the gel buffer wider genetic base of germplasm is essential. for this system contains 8.3 g histidine-HCl and Germplasm collection and characterisation is one of the 0.03 g EDTA. The electrode buffer for this system strategies to increase the genetic base to meet the above contains 15.1g tris-base and 7.3 g citric acid. objective. Diversity studies are highly important to Twenty five samples were run on each gel plus two guide such germplasm collection and selecting sites for wicks dyed with bromophenol blue to act as a in situ conservation (1). marker control. Electrophoresis was carried out at 4oC with a constant current of 70 milli ampere (250 Isozymes are practical and useful genetic and volts) for lithium borate gels and 50 milli ampere biochemical markers as well as good estimators of (200 volts) for tris-citrate gels. Gels were run genetic variability in plant populations (5). The most approximately 8 cm within 4-5 hours. commonly used measures of intra-population variation are the percent of polymorphic loci, the effective Three enzyme systems were selected for detailed number of alleles per locus and the mean proportion of analysis after a preliminary survey of five enzymes loci heterozygous per individual. The last parameter is (ACP:EC 3.1.3.2, AAT:EC 2.6.1.1, EST:EC 3.1.1, the expected mean heterozygosity assuming Hardy- PRX:EC 1.11.1.7, and LAP:EC 3.4.11.1 ) since they Weinberg equilibrium. Studies to evaluate the total gave consistent results with this species. The three variation of isozymes are thus of paramount importance enzymes analysed were esterase (EST), aspartate in order to determine genetic variability of populations aminotransferase (AAT) and acid phosphatase (ACP). Buffer system 1 was used for EST and AAT, while

43 Lathyrus Lathyrism Newsletter 2 (2001) buffer system 2 was used for ACP. In the first gel visible. An assessment of isozyme phenotypic system, the gel was cut in to three slices. The top slice polymorphism was made using the overall banding was discarded since most enzymes did not stain well on patterns. Phenotypic polymorphism, genetic distance, it. For buffer system 1, the second slice was used for degree of differentiation (FST) and hetrozygosity were EST and the third slice for AAT. For buffer system 2, determined using Biosys software (7). A tentative genetic the second slice was used for ACP. interpretation of the banding patterns was made based on the reported structure of each enzyme in different The following staining recipes were used following plant species (11) and particularly in related genera such protocols developed in the literature (2, 11). as Pisum, Lens and Vicia, where the information was available. For ACP: 50 ml 0.4 M sodium acetate buffer pH 5.0 which was used to pre-soak the gel for 20 minutes at 4oC, 50 milligram beta napthyl acid phosphate, 50 Results and Discussion milligram fast blac kk salt, and 0.5 ml 10%MgCl2. For AAT: 50 ml 0.1 M tris-HCl (pH 8.5), 18 mg alpha Population variability. The genetic variability at seven ketoglutaric acid, 65 mg DL-aspartic acid, 250 mg PVP, loci in all populations is presented in Table 1. The mean 50 mg disodium EDTA, 710 mg Na2HPO4 and 200 mg number of alleles per locus ranged from 1.6 to 2.1, the fast blue BB salt. lowest in population 219950 and the highest in For EST: 20 ml distilled water, 20 ml 0.2 M NaH2PO4, population 236705. The polymorphic loci ranged from 10 ml 0.2 M Na2HPO4, 2 ml 1% α-naphtyl acetate and 57.1 to 85.7%, the lowest in population 219950 and the 125 mg fast blue BB salt and 1 ml acetate were used. highest in populations 236705 and 46024. A locus is polymorphic if more than one allele is detected. As per Variation in banding patterns was determined by the this criteria, polymorphism was detected in all migration from the origin towards the anode. Isozyme populations. The highest polymorphism was detected in zones were designated to define the general area on the populations collected from the Gondar region. This is in zymogram within which the bands migrated. The zones line with morphological data (not reported here). Mean were numbered from the fastest to the slowest migration heterozygosity for the populations ranged from 0.081 in from the point of insertion of the wicks in the gel. population 46035 to 0.313 in population 226001. Scoring was made for those bands which were clearly

Table 1. Genetic variability at 7 loci in all populations.

Population Origin Altitude (m) Mean no. of % loci per locus He* alleles polymorphic 208449 Gojam 2300 1.7 71.4 0.189 207499 Gondar 2600 2.0 71.4 0.214 46035 Wollo 2375 1.9 71.4 0.081 219945 Tigray 1870 1.9 85.7 0.302 46024 Shoa 2460 1.9 85.7 0.297 219950 Tigray 2230 1.6 57.1 0.196 226001 Wollo 2400 1.9 71.4 0.170 46027 Shoa 2420 1.7 71.4 0.196 236711 Gojam 1840 1.7 71.4 0.170 236705 Gondar 1800 2.1 85.7 0.241 *He is expected heterozygosity

44 Lathyrus Lathyrism Newsletter 2 (2001)

Marked differences in the extent of differentiation Correlation between morphological and isozyme (FST) were shown between many loci (Table 2). The diversity indices. The association between populations were differentiated markedly for AAT-1, morphological diversity estimates (Shannon Weaver AAT-2, and EST-2. The level of differentiation was low diversity index) and genetic diversity estimates from for ACP-3 (0.031) and for ACP-1 (0.118). The degree isozyme data (expected hetrozygosity estimate, H) at of differentiation (FST) of the individual loci ranged population level was negative and not significant (r = - from 0.031 for ACP-3 to 0.784 for AAT-2. The mean 0.25). Yunus et. al. 1991 also observed the absence of FST value (0.346) is medium as compared to the correlation of isozymes with morphological data in average FST value for inbreeding species (0.510) (4). grasspea. The absence of correlation between markers Grasspea is both drought and flooding tolerant. In indicates that there is no one best marker that can be Gojam and Gondar it is grown in flooded vertisols, used for diversity study. Hence, it is important to study while in Wollo and Tigray it is grown in moisture stress diversity by using both morphological and molecular conditions. Accordingly, the high FST value reflects markers. Similar results have been reported by many adaptation to strong environmental dissimilarities or authors in other crops (1,8,10). high level of genetic drift maintained by restricted gene flow between populations. Similar observations have Table 2. Summary of FST at all loci been noted in barley (1). Locus FST No. of alleles The distance between populations ranged from 0.001 to AAT-1 0.6302 2 0.341 (Table 3). The highest distance (0.341) was AAT-2 0.784 2 between population 219950 from Tigray and population EST-1 0.188 2 46035 from Wollo. These two populations were from EST-2 0.353 4 different regions separated by some distance, indicating ACP-1 0.118 2 that geographic isolation is one of the important factors ACP-2 0.157 3 responsible for the observed genetic distance disparity. ACP-3 0.031 3 Mean 0.346

Table 3. Matrix of genetic distance coefficients. Below diagonal: unbiased minimum distance (7).

Population Population12345678910 1- 2 0.067 - 3 0.117 0.174 - 4 0.068 0.031 0.164 - 5 0.109 0 .039 0.183 0.001 - 6 0.212 0.152 0.341 0.151 0.155 - 7 0.262 0.231 0.213 0.24 0.245 0.097 - 8 0.058 0.044 0.140 0.008 0.021 0.132 0.228 - 9 0.078 0.040 0.202 0.030 0.037 0.204 0.278 0.067 - 10 0.179 0 .180 0.120 0.180 0.214 0.285 0.166 0.168 0.255 -

45 Lathyrus Lathyrism Newsletter 2 (2001)

Acknowledgements 7. Nei M. (1978). Estimation of average hetrozygosity This study was financially supported by the Bio- and genetic distance from a small number of Diversity Institute of Ethiopia. The first author wishes individuals. Genetics 89, 583-590. to express his gratitude to the Amhara National 8. Price SC, Shumaker KM, Kahler AL, Allard RW, Regional State for covering personal costs during the Hill JE. (1984). Estimate of population study. differentiation obtained from enzyme polymorphisms and quantitative characters. Heredity 75, 141-142. 9. Roy DN, Kisby DE, Robertson RC, Spencer PS. References (1989). Toxicology of Lathyrus sativus and the 1. Abebe D, Bjornstad A. (1997). Geographical, neurotoxin BOAA. In: Spencer PS (ed) Grass pea: altitude and agroecological differentiation of the threat and Promise. Proceedings of the isozyme and hordine genotypes of landrace barley International Network for the Improvement of from Ethiopia: Implications to germplasm Lathyrus sativus and Eradication of Lathyrism conservation. Gen. Res. Crop Evol. 44, 43-55. Workshop, London, May 1988. Third World 2. Chamberlain JR. (1998). Systematics and Medical Research Foundation, New York. pp. 76- population genetics laboratory protocols for 85. isozyme analysis. London. 10. Seifu T. (1997). Genetic diversity and structure of 3. CSA. (1998). Agricultural sample survey, area and tetraploid wheat landraces of the central highlands production of major crops. Statistical Bulletin 189. of Ethiopia. Doctoral Thesis. Swedish Unversity of 4. Hamrick JL, Godt MJW. (1990). Allozyme Agricultural Sciences. Svalov, Sweden. diversity in plant species. In: Brown ADH, Clegg 11. Wendel JF, Weeden NF. (1990). Visualization and MT, Kahler AL, Weir BS (eds). Plant population interpretation of plant isozymes. In: Soltis, Soltis genetics, breeding and germplasm resources. (eds.) Isozymes in plant biology. London, Chapman Massachusetts, Sunderland. pp. 43-63. and Hall. pp. 5-45. 5. Hamrick JL, Godt MJW. (1997). Allozyme 12. Wuletaw T, Wollelie M, Adugna K. (1997). diversity in cultivated crops. Crop Sci. 37, 26-30. Genetic improvement of grass pea (Lathyrus 6. Haque A, Mannan MA. (1989). The problem of sativus) for low toxin content and other agronomic lathyrism in Bangladesh. In: Spencer PS (ed) Grass characters. In: Crop Science Society of Ethiopia pea: the threat and promise. Proceedings of the (CSSE) Sebil. Vol 7. Proceedings of the Seventh International Network for the Improvement of Annual Conference, 27-28. Lathyrus sativus and Eradication of Lathyrism 13. Yunus AG, Jackson MT, Catty PJ. (1991). Workshop, London, May 1988. Third World Phenotypic polymorphism of six enzymes in the Medical Research Foundation, New York. pp. 27- grasspea. Euphytica 55, 33-42 35.

46 Lathyrus Lathyrism Newsletter 2 (2001)

Effect of foliar spray of potassium grasspea by broadcasting grassspea seeds just before nitrate and calcium nitrate on harvest of wet season photosensitive rice (cv. grasspea (Lathyrus sativus L.) grown Rupsail). The experiment was a randomised block in rice fallows. design with three replications. The treatments consisted of foliar spray of potassium nitrate, calcium R.K. Sarkar and G.C. Malik nitrate, water spray and control (Table 1). The foliar sprays of potassium nitrate (KNO3) and calcium nitrate (Ca(NO3)2) were given at 3 concentrations, - Department of Agronomy, University College of supplied equal amounts of N (as NO3 ) in the Agriculture, Calcutta University, Calcutta, West respective treatments (Table 1). The seeds of grasspea Bengal 700 019, India. (cv. Nirmal I) at 60 kg/ha were broadcast uniformly, 10 days before harvest of the physiologically mature rice crop, over the field under muddy conditions. In both years this was carried out in the second fortnight Introduction of November without any tillage operations. Of 21.04 million ha rainfed lowland rice areas in South Asia, India occupies 16.7 million ha, where Starter nitrogen, basal phosphorus and potassium were photosensitive traditional rice is grown during June- applied, broadcast at 20 kg/ha each of N, P2O5 and July to November-December every year. Following K2O a day after seed sowing. Dilute solutions of the harvest of rice, the majority of land remains fallow nutrient salt as per treatment were applied at 800 litres till the next rice in the following year for a large of water/ha as a foliar spray during 50% flowering number of agro-economic factors. stage of the crop. Data were recorded on 10 random plants for yield attributes, seed yield from plot, However, there is a possibility of raising another crop treatment wise. The rainfall received during the crop as a paira crop in a relay cropping system before the duration in 1998 and 1999 was 54 and 46 mm, harvest of rainfed lowland rice. Grasspea is suitable respectively. for relay cropping with paddy rice (1) and it has potential among grain legumes for its tolerance to dry conditions and its adaptibility to unfavourable environments (2). Since grasspea in relay cropping Results with rice suffers from nutrient stresses during the Foliar spray of KNO3 and Ca(NO3)2 salts exerted reproductive phase, late application of N is effective in conspicuous effects on yield attributing characters of reducing flower and pod drop in legume and in grasspea (Table 1). Foliar spray of KNO3 at 0.50% increasing seed yield (3). Legumes, in general, require during 50% flowering stage showed maximum values K(4) and Ca(5). This study, therefore, was initiated to of pods/plant, length of pod, seeds/pod and 1000 seed investigate the effect of foliar spray of potassium weight; it was significantly superior to water spray nitrate and calcium nitrate on a paira crop of grasspea and unsprayed control, but was on par with Ca(NO3)2 grown in rainfed lowland rice fallows. at 0.406%.

KNO3 at 0.50% when sprayed during 50% flowering stage while equivalent to Ca(NO3)2 at 0.406%, Materials and Methods recorded a significantly and appreciably higher seed The field experiment was conducted during the late yield than spray of KNO3 at 0.25 and 1.00%, winter season of 1998 and 1999 at the Experimental Ca(NO3)2 at 0.203 and 0.812%, water spray and Farm of Calcutta University, Baruipur, West Bengal, unsprayed controls. This was the case in both years India (88.28° E, 22.22° N and approximately 1 m and also in the pooled data (Table 1). The increase in above sea level). Soil was Gangetic alluvial (Entisol), seed yield due to spraying of KNO3 at 0.50% at 50% having 0.72% organic carbon, 22 kg available P2O5/ha flowering stage was 78.4 and 85.7% over the water and 240 kg available K2O/ha with pH 6.2. The sprayed and controls, respectively, on a pooled basis. experiment was set in rice fallows, with paira crop of

47 Lathyrus Lathyrism Newsletter 2 (2001)

Table 1. Effect of foliar spray of KNO3 and Ca(NO3)2 on yield attributes and yield of grasspea grown as a paira crop (pooled data for 2 years, 1998 and 1999).

Seed yield (t/ha) Pods Pod length Seeds 1000 seed Seed yield 1998 1999 Pooled Treatment /plant (cm) /pod weight (g) /plant Control 15.34 2.1 1.90 126.0 5.84 0.49 0.48 0.49 Water 16.70 2.4 2.25 127.3 6.10 0.50 0.53 0.51 KNO3 (0.25%) 18.68 2.9 3.34 135.4 7.28 0.82 0.76 0.78 KNO3 (0.50%) 21.75 3.6 4.14 149.5 8.89 0.89 0.93 0.91 KNO3 (1.00%) 19.80 3.2 3.75 141.6 8.21 0.84 0.80 0.82 Ca(NO3)2 (0.203%) 18.29 2.7 3.25 132.3 7.19 0.74 0.76 0.74 Ca(NO3)2 (0.406%) 19.88 3.4 3.90 143.2 8.28 0.86 0.82 0.84 Ca(NO3)2 (0.812%) 18.90 3.1 3.69 139.6 7.96 0.77 0.81 0.79

LSD (P<0.05) 0.99 0.45 0.23 1.12 0.09 0.36 0.01 0.25

Discussion References Grasspea in a relay cropping system with rainfed 1. Sharma RN, Chitale MW, Ganvir GB, Geda AK, lowland rice responded favourably to foliar spray of Pandey RL. (2000). Observations on the KNO3 and Ca(NO3)2. Improvement in yield attributes development of selection criterion for high yield of grasspea due to foliar spray of KNO3 at 0.50% and low neurotoxin in grass pea based on genetic might be attributed to hastened availability of N in the resources. Lathyrus Lathyrism Newsletter 1, 15- plant system, more chlorophyll synthesis, greater 16. accumulation of protein in plants and efficient 2. Sarno R, Stringi L. (1979). Proc. Cong. translocation of assimilates to reproductive parts. “Prospective delle proteaginose in Italia”, Perugia, pp. 365-370. Foliar spray of KNO3 at 0.50% during 50% flowering 3. Brevadan RE, Egli DB, Leggett JE. (1978). resulted in higher seed yield than other treatments Influence of N nutrition on pod abortion and yield (Table 1). Foliar spray of KNO3 at 0.50% increased of soybeans. Agron. J. 70, 81-84. the seed yield by 85.7% over the unsprayed control in 4. Barta, AL (1982). Response of symbiotic N2- pooled data owing to the favourable effect on yield fixation and assimilate partitioning of K supply in attributes. Spray of KNO3 at 0.50% during flowering alfalfa. Crop Sci. 22, 89-92. supplied N and K which are effectively absorbed as 5. Albrecht WA, Davis FL. (1929). Physiological anion and cation by plants, and might have delayed importance of calcium in legume inoculation. Bot. the synthesis of abscisic acid and promoted cytokinin Gaz. 88, 310-321. activity(6), causing higher chlorophyll retention. This 6. Brevadan ER, Hodges MA (1973). Effect of may secure higher photosynthetic activity in effective moisture deficit on 14C translocation in corn (Zea leaves and supplied developing pods with current mays L.) Plant Physiol. 52, 436-439. photosynthates for proper filling, resulting in higher 7. Mengal K. (1976). Potassium in plant physiology yield. Besides the beneficial functions of nitrate and yield formation. Indian Soc. Soil Sci. Bull. nitrogen, the prevalence of K+ in KNO3, might have 10, 23-40. improved grain filling and phytomass production, due to increasing photosynthetic activity and effective translocation of assimilates to reproductive parts (7) resulting in higher yield. Foliar spray of KNO3 at moderate rate of 0.50% proved more effective than lower (0.25%) and higher rate (1.00%) and all the rates of Ca(NO3)2 supplying equivalent amount of NO3- as in KNO3 for grasspea in rice fallow land.

48 Lathyrus Lathyrism Newsletter 2 (2001)

The nutritional value of Lathyrus assess the nutritional value and animal health aspects of L. cicera grain as a feed for sheep. Comparisons cicera and Lupinus angustifolius were made against lupin grain (Lupinus grain for sheep. angustifolius), which is commonly used in Australia for supplemental feeding of sheep.

Colin White1, Colin Hanbury2 and One hundred individually penned Merino wethers (35 Kadambot Siddique3 ± 0.3 kg) were allocated to 5 treatment groups of 20 sheep. Following a 4 week adjustment period, sheep were fed ad libitum the following treatment diets for 1. CSIRO Livestock Industries, Private Bag, PO up to 13 weeks: basal (oaten hay plus 2.5% urea), Wembley, Australia, 6014 35% lathyrus, 35% lupin, 70% lathyrus and 70% 2. CLIMA, University of Western Australia, lupin. The remainder of the diet was oaten hay and Crawley, Australia, 6009 minerals. After 10 weeks on treatments, 14 sheep per 3. Agriculture Western Australia, Locked Bag 4, group were killed for carcass assessment. The Bentley Delivery Centre, Australia, 6983. remaining 6 sheep were fed to maintenance metabolisable energy (ME) requirements for one Email: (1) [email protected] week followed by a 10 day faecal and urinary collection period.

Chemical and in sacco analysis of the grains showed Paper submitted to Animal Feed Science and that lathyrus contained less protein (28% vs 36%), fat Technology, in press for 2001 (0.7 vs 4.1%) and fibre (25% vs 36% NDF) but more starch (42% vs 1%) and antinutritional compounds (ODAP, tannins, trypsin inhibitor) than lupin. Essential amino acid composition was similar for the Summary two grains, as were in sacco degradabilities of dry Lathyrus sativus and L. cicera have recently been matter (84 vs 81%) and protein (92 vs 94%). found suitable as multipurpose legume crops in the southern Australian dryland regions, with a potential The feeding experiment showed that lathyrus had a growing area of 100,000 to 300,000 ha. However, higher nutritional value than lupin in terms of their use is limited by the presence, mainly in seeds, voluntary feed intake, liveweight gain (P < 0.01; of 3-(-N-oxalyl)-L-2,3-diamino propionic acid Fig.1), carcass weights (P < 0.05) and feed efficiency (ODAP), an agent which causes lathyrism. Research (P = 0.05). Wool growth reflected ME intake, and at CLIMA has produced a high yielding variety of L. there was no independent effect of grain type (P > cicera (cv Chalus) which contains consistently low 0.05). levels of ODAP. The aim of the present study was to

50 48 46 44 42 40

Liveweight (kg) 38 36 34 0 20406080 Days on treatment diets

Fig 1. Growth of sheep fed diets containing Lathyrus cicera grain (open circles) or Lupinus angustifolius grain (closed circles). Values are mean ± SE, combined for the two inclusion levels of grain (35% and 70%).

49 Lathyrus Lathyrism Newsletter 2 (2001)

Results from the balance study showed that ME Meat from sheep fed lupin tended to be yellower than concentration was the same for both grains (14 MJ/kg that from those fed lathyrus (P = 0.05). Apart from dry matter). this, there were no differences in meat quality due to grain type (e.g. redness, pH, taste or tenderness). There were no visible or biochemical signs of ill health (inappetence, lethargy, shaking or instability) in In conclusion, compared with lupin grain, low ODAP any sheep fed lathyrus. Several sheep fed the 70% Lathyrus cicera grain appears to be of high nutritional lupin diet had mild diarrhoea, and two were eventually value for sheep, with no evidence of adverse effects removed from the experiment due to anorexia. on sheep health.

50 ! RECENT PUBLICATIONS

This section is intended to provide details of recent proceedings and other larger publications, and details of how to obtain copies of the publications.

Title Beta-N-Oxalylamino-L-alanine. In: Spencer PS, Schaumburg HH (eds) Experimental and Clinical Neurotoxicology, 2nd edition, pp 925-938. Publication date 2000 Author(s) Hugon J 1, Ludolph AC, Spencer PS

Contact details 1. Professor Jacques Hugon Fac. Medicine Chair and Head, Dept. Anatomy The University of Hong Kong 5 Sassoon Road, Hong Kong Email: [email protected]

Publisher Oxford University Press, New York.

51 RECENT PUBLICATIONS- SOME PAPERS

This is a list of some very recently published papers relevant to Lathyrus and lathyrism. If authors would like to add recently published work (no older than 1999) to future versions of this list they are requested to post a copy (or Email an electronic version) of any publication to the editor. Any copies sent will be retained by the editor. Many publications can of course be found through electronic abstracting means, however not all papers turn up in all searches and this will provide an additional method of obtaining copies or contacting the authors.

Title and Publication Details Authors Contact details Analysis of β-N-oxalyl-L-α,β-diaminopropionic acid and Liang Zhao, Zingguo homoarginine in Lathyrus sativus by capillary zone Chen, Zhide Hu, electrophoresis. (1999). Qianfeng Li, Zhixiao J. Chromatography A 857, 295-302. Li

Kinetics studies on thermal isomerization of β-N-oxalyl-L- Liang Zhao, Zhixiao α,β-diaminopropionic acid by capillary zone electrophoresis. Li, Guanbin Li, Prof. Li Zhixiao (1999). Xingguo Chen, Zhide National Laboratory of Applied Organic Phys. Chem. Chem. Phys. 1, 3771-3773. Hu Chemistry Lanzhou University Determination of neurotoxin 3-N-oxalyl-2,3- Fei Wang, Xiong Lanzhou 730000 diaminopropionic acid and non-protein amino acids in Chen, Qian Chen, P.R.China Lathyrus sativus by precolumn derivatization with 1-fluoro- Xinchen Qin, Zhixiao 2,4-dinitrobenzene. (2000). Li Email: [email protected] J. Chromatography A 883, 113-118.

Accumulation of ABA and ODAP in Lathyrus sativus under Xing Gengsheng, water stress. (2000). Zhou Gongke, Li Chinese J. App. Ecol. 11, 693-698. Zhixiao, Cui Kairong

Water stress and accumulation of beta-N-oxalyl-L-alpha,beta- Gengsheng X, Cui diaminopropionic acid in grass pea (Lathyrus sativus). (2001) KR, Ji L, Wang YF, J. Agric. Food Chem. 49, 216-220. Li ZX.

Reduction efficiency of the neurotoxin beta-ODAP in low- Kuo YH, Bau HM, Prof. Fernand Lambein toxin varieties of Lathyrus sativus seeds by solid state Rozan P, Chowdhury Fysiologische Scheikunde, University of fermentation with Aspergillus oryzae and Rhizopus B, Lambein F. Gent. Jozef Kluyskensstraat, 27 microsporus var chinensis. (2000). B-9000 Gent, Belgium J. Sci. Food Agric. 80, 2209-2215. Email: [email protected]

Beta-N-Oxalylamino-L-alanine. (2000) Hugon J, Ludolph AC, Prof. Jacques Hugon In Spencer, P.S., Schaumburg, H.H., eds., Experimental and Spencer PS. Fac. Medicine Chair and Head, Dept. Clinical Neurotoxicology, 2nd edition, Oxford University Anatomy, University of Hong Kong Press, New York, pp 925-938. 5 Sassoon Road, Hong Kong Email: [email protected] Effect of β-ODAP, the Lathyrus sativus neurotoxin, and Kusama T, Kusama- Tadashi Kusama related natural compounds on cloned glutamate receptors and Eguchi K, Ikegami F, Laboratory of Physiology and Anatomy, transporters expressed in Xenopus oocytes. (2000). Yamamoto A, Kuo Y- Nihon University College of Pharmacy, Res. Comm. Pharm. Toxic. 5, 37-55. H, Lambein F, Funabshi, Chiba 274-8555, Japan Watanabe K. Email: [email protected]

52 ITEMS OF INTEREST

Includes any items of interest to Lathyrus and lathyrism researchers. Please send any suggested inclusions for future editions to the editor. Conferences, meetings or other items will be included.

Lathyrus discussion on-line

A Lathyrus Email discussion group has been initiated. To join send an Email to: [email protected] or notify Colin Hanbury at [email protected] and you will be added to the list. Alternatively if you visit: http://groups.yahoo.com/group/Lathyrus and follow the prompts under “Join This Group!”, you will be able to join.

Lathyrus bibliography on-line

“An Annotated Bibliography for the Genus Lathyrus” compiled by Dirk Enneking is available on-line at http://mansfeld.ipk-gatersleben.de/RIS/RISWEB.ISA This bibliography is available in written form and CD-ROM from CLIMA. See Lathyrus Lathyrism Newsletter Vol. 1 for details.

Lathyrus publications on-line from IPGRI.

1. Lathyrus Genetic Resources in Asia: Proceedings of a Regional Workshop, 27-29 December 1995, Indira Gandhi Agricultural University, Raipur, India. 1996. Available at: http://www.ipgri.org/publications/pubfile.asp?ID_PUB=656

2. Lathyrus Genetic Resources Network: Proceedings of a IPGRI-ICARDA-ICAR Regional Working Group Meeting, 8-10 December 1997, National Bureau of Plant Genetic Resources, New Delhi. 1998. Available at: http://www.ipgri.org/publications/pubfile.asp?ID_PUB=94

3. Grass Pea. Lathyrus sativus L. Promoting the conservation and use of underutilized and neglected crops. 18. 1997. Available at: http://www.ipgri.org/publications/pubfile.asp?ID_PUB=430

4. Descriptors for Lathyrus spp. 2000. Available at: http://www.ipgri.org/publications/pubfile.asp?ID_PUB=547

For any further information on these publications contact [email protected]

53