1. Overview

1.1 Introduction As a 1990 sub-Saharan Africa survey by UK’s Originating in Central America and introduced into Natural Resources Institute (NRI) showed, MSV is Africa by the Portuguese in the 16th century, maize one of the two most important biotic constraints has become Africa’s most important staple food affecting maize production in Africa (the other is crop. Both large and small-scale farmers grow maize mottle virus). The most important abiotic maize in various agroecological zones. It is factor is drought, and when weather conditions increasingly replacing traditional food crops, combine with MSV epidemics, maize shortages including cereals such as sorghum and millet. can produce famines in Kenya’s semiarid regions. Maize is so important that in most of sub-Saharan The large fluctuations in maize production also Africa poor maize yields are usually linked to food generate large price swings. Prior to May 1999, for shortages and famine. example, prices for maize flour were low, averag- ing under 30 Kenyan shillings for 2-kg bags sold in Yet although maize is a crucial staple food crop, rural retail shops and major urban supermarkets. the average yield per hectare in Africa is the lowest But despite Kenya’s relatively well-developed in the world. Furthermore, annual maize yield agricultural infrastructure, May prices suddenly increases in Africa have not kept pace with popula- increased to 55 shillings per 2-kg packet—an tion growth (1.9 vs. 2.9%, respectively, for the years increase of more than 80%. This rise in price was 1961-65/1984-88). Unless this situation is reversed, due in part to an outbreak of MSV disease. food dependency will increase throughout much of the continent. According to the Food and Agricul- Given Kenya’s fertile agricultural regions, farming ture Organization (FAO), the worldwide average should be a profitable business for maize produc- maize yield per hectare is about 4 t, but in Africa it ers (two-thirds of whom are women according to is 1.7 t, less than half the global average. The yield experts at KARI—a significant measure of how is 8 t/ha in a highly industrialized nation such as the important maize is to families’ food security). But USA. African maize yields are clearly far below the MSV strikes at the heart of the region’s farming crop’s genetic capacity: on-farm yields range from system, destroying the main crop that small-scale only 10 to 35% of the potential. In Kenya, for farmers subsist on. And the problem is becoming example, on-farm yields were 2.3-4.7 t/ha below more acute. From the 1930s to the 1980s, MSV potential. In Malawi, on-farm hybrid maize yields epidemics were commonly reported, but in the last reached a mere 2.5 t/ha compared with 8 t/ha decade their frequency and destruction have during trials at agricultural stations. When open- increased. At the same time, subsistence farmers pollinated maize was used, the difference was 10- lack options to control the disease. Applying fold (0.6 t/ha compared with 6 t/ha). These numbers systemic insecticides helps reduce yield loss by show how important it is to raise maize yields to reducing vector populations, but this is not an improve food security for households in both rural option for most farmers in Africa, especially small- and urban Africa. scale farmers who cannot afford to buy chemicals.

One of the major problems facing maize farmers in The devastating impact of MSV is not confined to Kenya and other east African nations is maize Kenya alone. Serious MSV epidemics have been streak virus (MSV). Indigenous to Africa and its reported in at least 20 African nations, including offshore islands, MSV causes yield losses of up to Nigeria, Ethiopia, Sudan, Tanzania, Zimbabwe, 100% even in high potential agricultural zones. A Zambia, Angola, Mozambique, Malawi, Madagas- good example of this is the Githunguri region in car, Senegal, Ghana, Cameroon, Togo, Benin, Kenya’s Kiambu district. It is an agricultural zone Sao Tome, and Burkina Faso. MSV disease has with adequate rainfall and fertile red soil. MSV is been reported in all areas of the continent except wreaking havoc in this area. Since 71% of Kenya’s the north. land is arid or semiarid, such serious crop losses in high potential zones can devastate the national For farmers who suffer major yield losses due to economy and threaten food security, which is MSV in the high yield potential areas of Kenya and already precarious. Currently, Kenya’s annual other African countries, scientific data and surveys average maize production is 2.7 million t, slightly are no substitute for the practical solutions that they lower than the 3 million tons consumed each year— urgently need. Pauline Ruiru, a small-scale farmer and consumption is expected to increase in the from Githunguri in the fertile highlands north of future. Nairobi, put the question plainly to a visiting team

1 markers for breeding, and increasing the capacity of national programs to breed and adapt MSV resistant maize. Dr. Florence Wambugu, director of ISAAA’s AfriCenter, believes that a solution can be found within 5 years.

Much of the efforts to improve maize yields have so far focused on breeding for high-yielding varieties (HYV). Diseases and insects, which easily reduce yields from such varieties by over 50%—and much more if postharvest losses are included—tended to be sidelined. It was probably assumed that farmers would use chemical pesti- cides and take agronomic measures to control diseases and vectors. This era of “pure plant breeding” also tended to exclude plant patholo- gists, entomologists, and other scientific special- ists who possessed important information about maize and its cultivation. Furthermore, the promise of biotechnology was overlooked or underesti- mated. Yet it is increasingly clear that both small- and large-scale farmers stand to gain from various aspects of biotechnology, particularly its low-input requirements. This is because biotech packages the technology “in the seeds,” a delivery system that farmers have handled since the emergence of agriculture 10,000 years ago.

An elderly farmer experiences major yield loss due to Kenya’s past emphasis on breeding to improve MSV infection on local Kenyan hybrids. yields excluded other factors that reduce yields. This has resulted in high-yielding maize varieties that are equally highly susceptible to MSV, such as the popular 511 hybrid. Farmers can lose the whole of scientific experts: “Can you please tell me what I season’s crop if the MSV infection occurs at an should do?” early stage. Before 1980, losses due to MSV in Africa were around 10%. In the 1990s the trend Today, no one can provide farmers such as worsened: average MSV yield losses range today Pauline Ruiru with a direct, clear-cut answer to her from 30 to 50%—this in a continent that imports 25% MSV problem. But according to Dr. Anatole of its grain. Robert W. Herdt, director for Agricultural Krattiger, executive director of the International Sciences at the Rockefeller Foundation, has Service for the Acquisition of Agri-biotech Applica- pointed out that dependency on imported grains tions (ISAAA), the solution lies in biotechnology: has increased over the last three decades—and a “Millions of small-scale farmers can seize the quarter of the food imported is donation or food promise of better lives through the altruistic dis- aid. In short, there is an urgent need to control semination of such biotechnology.” Indeed, we major crop constraints such as MSV so that Africa cannot afford to ignore biotechnology applications can become self-sufficient in food production. that can help African maize farmers control MSV in a cost-effective, agriculturally sustainable, and Africa, with maize yields that are less than half the environmentally protective way. global average, never fully plunged into the Green Revolution that enabled other developing nations In 1996, ISAAA launched a project to help national in Latin America and Asia to become self-sufficient research groups in Africa combat MSV through in grain production through the increased use of improved resistance breeding. The collaborative high-yielding seed varieties, pesticides, fertilizers, project aims to provide a better understanding of and irrigation. Biotechnology now offers the the epidemiology of MSV, its insect vectors, and continent a major opportunity to curb hunger and genetic resistance to MSV in maize varieties. It unacceptable levels of food dependency. while at may involve developing appropriate molecular the same time reducing the use of toxic pesticides.

2 Identifying and breeding maize varieties that are distinguish between various levels or mechanisms resistant to MSV is a very practical contribution in of MSV resistance, especially when coupled with the war against food shortage and famine that stalk insect transmission tests. Developed by the John parts of Africa at the dawn of this new millennium. Innes Centre (JIC), UK, agroinoculation enables researchers to differentiate between resistance The procedures and capacity for breeding maize against the insect vector, resistance against the are well established in several African nations, transmission of the virus, resistance against the including Kenya, Zimbabwe, South Africa, Nigeria, virus spreading within the plant, and host infection Reunion Island, and others. But the technology and or immunity. expertise to understand the pathology of MSV and the modes and genetic basis of resistance are 1.3 Collaborating to Combat MSV missing. All these are needed before effective The MSV project is an example of a unique, well- resistance screening and breeding strategies can coordinated collaborative effort at global, regional, be developed to conquer MSV. and national levels to solve an urgent need of resource-poor farmers. Facilitated by ISAAA, the Such efforts will improve the living standards of MSV project involves international collaboration rural families and poor urban populations—those from both public institutions and the private sector. who are hit hardest by food price increases and shortages. Results already indicated that MSV can These institutions include the Kenya Agricultural be controlled through resistant hybrids adapted to research Institute (KARI), which hosted the project, Kenya’s ecological zones. International Center for Insect Physiology and Ecology (ICIPE) for insect vector studies, the John 1.2 Maize Streak Virus (MSV) Innes Centre for molecular marker studies and Indigenous to Africa and adjacent islands, MSV is agroinoculation, and the University of Cape Town believed to have evolved with native grasses. It is (UCT) for expertise in virology needed to identify transmitted by leafhoppers belonging to the genus MSV isolates. Other organizations also made Cicadulina, a species which varies in its ability to important contributions, including the International transmit the virus. Transmission, however, is Institute of Tropical Agriculture (IITA) in Nigeria, the inherited, dominant, and sex linked, with males Centro Internacional de Mejoramiento de Maiz y being heterozygous. Species known to transmit Trigo (CIMMYT) in Mexico, the Grain Crops MSV include C. mbila, C. similis, C. storeyi, C. Institute, PANNAR Seeds, Kenya Seed Company, triangula, C. arachidis, C. latens, C. bipunctata, C. and CIRAD. ghauri, and C. parazea. C. mbila is considered the most common vector. KARI, one of Africa’s leading national agricultural research service institutions, has direct and indirect A gemini virus, MSV is one of seven viruses that networks that can be used to share and dissemi- attack maize crops in Africa (worldwide 32 maize nate research results within the region. Dr. C. viruses have been identified). Eighteen grasses Ndiritu, KARI’s director, points out that “once the serve as alternate hosts—most of them are annual capacity for crop biotechnology, research, and species. In maize, the virus causes stunting, development has been well established, Kenya bareness, interveinal necrosis, and death. The will be able to share its technology and know-how younger the crop at the time of germination the with neighboring countries, including Uganda, higher the yield loss. In fact, yield losses easily Tanzania, Rwanda, and others with resource-poor reach 100% if MSV infects a maize crop in its first 3 farmers.” Working on MSV for two decades, KARI weeks. In Kenya, MSV outbreaks cause serious has taken a multidisciplinary approach that now losses in various agroecological zones, including includes pathologists, entomologists, and plant the coastal lowlands, central highlands, and the breeders. KARI’s Dr. Jane Ininda, for example, has lake basin. MSV exists in mixed populations with been receiving advanced training in mapping isolates that have different virulence. gene(s) with resistance to MSV at JIC. Dr. Benjamin Onudi has also been trained at JIC in The single-stranded DNA virus requires leafhop- agroinoculation techniques used for screening pers (Cicadulina spp.) for transmission; neither the maize genotypes with resistance to MSV. Other top virion nor the viral DNA can be mechanically experts include JIC’s Dr. Peter Markham and Dr. transmitted. The viral sequence or genome, Edward Rybicki of UCT. JIC is a centre of excel- however, can be delivered into the plant by lence that conducts research on plant genomes. Its agroinoculation techniques via Agrobacterium facilities can also maintain the vectors needed to tumefaciens. Agroinoculation makes it easy to carry out transmission studies. 3 1.4 Objectives of the MSV Workshop pants so that these results could be shared with In bringing together scientists, policymakers, others. government leaders, and others, the workshop • Further the main objective of MSV phase II, which sought to accomplish the following objectives: is to develop maize inbred lines with high levels • Discuss the results of the MSV project’s first of MSV resistance that can be used to develop phase in the context of an experienced, interna- hybrids. To effectively commercialize hybrids, tional community that could challenge and which is primarily a private sector activity, MSV interpret ideas and outcomes. phase II and related activities must be exposed • Update and familiarize the project with parallel or to private sector seed companies for their inputs. complementary work, especially in institutions The workshop also served as a good starting that supplied MSV-resistant germplasm. point for such connections. • Identify future issues. • Understand the progress made by other scien- Overall, the workshop and its proceedings re- tists involved in MSV research. vealed the extent of the progress made towards • Help to identify mutually beneficial collaborators developing maize varieties that are resistant to and linkages for future work. MSV disease. Partly by bringing together insights • Complement the KARI project’s efforts with from other MSV programs in the region, the work- progress being made elsewhere on MSV shop also helped to develop a strategy for future diseases. By widening their outlook, scientists research and development (R&D) of MSV-resis- can avoid working in isolation and reinventing tant maize varieties. KARI scientists gained from the wheel. the peer review and close analysis of the project • Challenge KARI researchers and other project by experts from the international commu0nity. collaborators to address the MSV problem from Institutions that had provided MSV-resistant maize a regional/African perspective. Both directly and germplasm, which included CIRAD’s C390, IITA’s indirectly, this helps facilitate reaching the Tzi3, CIMMYT’s CML202, PANNAR’s AO76, and intended goals more quickly. KARI’s Embu 11, took a keen interest in the project • Document and disseminate information gener- results. ated by KARI scientists and workshop partici-

Participants to the MSV International Workshop held 15-17 September 1999 at KARI and ISAAA AfriCenter, Nairobi, Kenya.

4 For Africa, the workshop was a big step towards • In addition 69% in OSU23I was grouped as controlling or eradicating MSV, which would greatly resistant followed by 62% in C390; 42% in CML02 strengthen food security and alleviate poverty. and 40% in Tzi3 were also considered as MSV Significantly, sources of resistance to MSV were resistant. identified in maize varieties Reunion Yellow and • Trials at Alupe in the Lake Victoria region Arkells Hickory as early as 1931. But 60 years after showed that 38% in OSU23I was highly resistant, recognizing MSV disease, Kenyan farmers still followed by 28% in C390. This dropped to 14% in have no maize hybrids with MSV resistance. The CML202 and 4% in Tzi3. workshop’s theme, “Advances in MSV Disease • Data from field trials at Muguga and Alupe Research in Eastern and Southern Africa,” reflects showed that MSV resistance in a specific maize the new and increasingly practical efforts used to variety may vary from one ecological zone to control MSV disease in a sustainable way. As Dr. another. Krattiger pointed out, “farmers need practical • Seasonal abundance and aerial densities of advice about how to protect their crops from MSV— trapped Cicadulina spp. showed two annual they do not need a parade of high scientific peaks—July-August and November-December. learning.” Vector transmission efficiency varies with Cicadulina chinai, leading with 70% MSV trans- The workshop was divided into six sessions: mission efficiency. The efficiency of MSV disease epidemiology, vector studies, MSV status transmission by Cicadulina varies from species reports in Africa and individual countries, breeding to species and between individuals within the and biotechnology, seed production, and MSV in same species with Cicadulina chinai being the the private sector. Research results from the most efficient. project show that maize varieties like CIMMYT’s • There is still a very high incidence of MSV in OSU23I, CIRAD’s C390, and IITA’s Tzi3 are Kenya. According to the latest survey (1998), 80- resistant to MSV. It should therefore be possible to 100% of maize crops in Southwest Kenya identify MSV resistance genes using advanced suffered from the disease. biotechnology skills. Their location will help clarify the relationship among various resistance sources These data suggest that the potential to incorpo- and facilitate the incorporation of MSV resistance rate MSV resistance into various hybrids is high. genes into hybrids or commercial genotypes. According to CIMMYT’s Dr. Alpha Diallo, there is Some of the most exciting results of the project no data to support the widespread notion that were from a study evaluating the “Expression of resistance to MSV breaks down under various disease symptoms among F3 lines of different environmental conditions—although there may be parental sources of MSV resistance.” KARI maize variability in resistance levels due to other factors. breeders already have their eyes on short-, Analysis of his lead paper on “MSV status in Africa” medium-, and long-term commercial products that shows that by 1998 CIMMYT Harare had devel- use MSV-resistant populations as a source of oped 54 MSV-resistant/tolerant inbred lines that future inbred lines. are available to maize researchers upon request (see section 2.1.2 for more information about Workshop participants were also provided with CIMMYT’s efforts). But participants agreed that the data and reports of achievements of phase I of the final focus should be full immunity against MSV—not MSV Project, including epidemiology studies that tolerance—because the project can avail of ad- are continuing into phase II. These include: vanced biotechnology tools and skills such as marker-assisted selection (MAS), agroinoculation, • Trials in Muguga, near Nairobi, resulted in some and genome mapping.

highly resistant F3 lines derived from the maize variety OSU23I provided by CIMMYT. At least According to Prof. Peter Markham, however, if this 46% in OSU23I had high resistance to MSV. is the goal many unanswered questions still • 26% in C390 provided by CIRAD had high prevent scientists from reaching it. Where does the resistance to MSV. virus go after harvest? Where does it come from? • At least 2% in Tzi3 from IITA was found to be What are the factors that control vector migration or highly resistant; resistance dropped to 0.01% in movements? Is the epidemic linked to mutation or CML202 from CIMMYT. new strains that are more virulent? His remarks made clear that much remains to be studied about the relationship between the virus and its hosts.

5 1.5 Workshop Summary: • Shift in emphasis from breeding to improve yield Achievements of Phase I to increasingly focusing on resistance to dis- of the MSV Project eases and pests to stabilize yields. The workshop also provided an opportunity to • Obtaining a wide range of much-needed maize share information about the project’s structure, germplasm from both private and public sectors programs, and achievements: for resistance breeding programs. • An increase in the national capacity to breed and • Advanced training on sophisticated biotechnol- adapt MSV-resistant maize. ogy skills such as DNA diagnostics, • An increased capacity and opportunity for KARI agroinoculation, and molecular markers. to collaborate at regional and global levels with • Closer collaboration among plant breeders, partners in both public and private sectors. pathologists, and entomologists who are used to • A visit by experts from JIC and UCT to MSV- working in isolation even within a national infected maize fields in Kenya. research institute like KARI. • Detection of MSV isolates with different biologi- Overall, this project has made considerable cal and molecular properties, which proves that progress in understanding MSV, its vectors, its mixed MSV populations do exist. This discovery epidemiology, and basis for resistance. Work in is very important for resistance breeding. phase II will focus more on efforts to determine the • Identification of zones or regions prone to MSV, locations of MSV-resistant genes. As this report of besides peak periods of MSV disease. the Workshop indicates, there is a lot of excitement • Gathering of preliminary evidence for the vari- about the preliminary data supplied by phase II. ability of virulence in MSV populations within agroecological zones and individual maize crops.

2. Status of MVS in Africa

2.1 Activities and Roles of International Prior to 1980, the disease was a minor lowland Organizations problem in Kenya, but it has since spread to all parts of the country, including the highly productive 2.1.1 Introduction (Dr. A. Diallo, CIMMYT) highlands and Central province. It is found in both forests and savanna, in coastal lowlands, and in Although MSV losses can be reduced through highlands up to about 2,300 m. Late planting is also agronomic practices such as early planting and linked with increased MSV infection, in areas with seed treatment with systemic insecticides, a more bimodal rainfall MSV infection levels are higher in reliable and environmentally sound strategy would the second season. be to develop MSV-resistant/tolerant cultivars. The genetics of maize streak disease has been 2.1.2 CIMMYT and MSV (Dr. A. Diallo, CIMMYT) studied and reported (Kim et al 1989, Pixley et al 1997, Nhlane 1997, Dintinger et al 1997, Welz et al Breeding for MSV resistance remains an important 1998), while various breeding methodologies have objective of CIMMYT’s maize program, and been applied and many MSV-resistant germplasm several MSV-resistant inbred lines have been have been developed (Bjarnason 1985, Efron et al developed at CIMMYT’s center at Harare, Zimba- 1989, Pixley and Zambezi 1996, Rodier et al 1995, bwe. In 1991 there were 22 inbred CIMMYT maize Diallo 1996, Dintinger et al 1997, du Plesis 1997, lines developed for MSV resistance. Designated Pixley et al 1997). as CML217 to CML238, another 22 lines were added to the CML series in 1993, extending the MSV disease is erratic and still unpredictable: any range from CML195 to CML216 for a total of 44 given year may bring insignificant yield losses or MSV-resistant lines. In 1998, CIMMYT’s MSV- losses of 100%. The 1983-84 MSV epidemic in resistant inbred lines increased by another 10, West Africa, for example, coincided with drought which pushed the total to 54. The last groups were and irregular rains that resulted in catastrophic 386ZM60, 387ZM609, 388ZM609, 389ZM609, losses. MSV is spreading rapidly in the region but 390EV7992, 391EV7992, 392Broad, 393Broad, there are no data to support claims of resistance 394Broad, and 395IITA1. Upon request, CIMMYT breakdown. There may be variations in resistance provides interested maize researchers with up to levels due to other factors. 30 kernels.

6 CIMMYT’s data and experience in breeding resistant materials, CIMMYT’s breeding work no germplasm for MSV resistance indicates that this longer uses backcross methods. Instead, can be a straightforward task. Significantly, how- recurrent selection within or between MSV- ever, the MSV resistance of specific crosses resistant populations and pedigree-breeding cannot always be reliably predicted from pedigree techniques are used to develop streak-resistant information or by evaluating parental performance. inbred lines. Some MSV-resistant lines are actually better than • CIMMYT also uses shuttle breeding. In Tanza- the donors’ average resistance (e.g., CML202), nia, this approach upgraded the streak resis- and some susceptible lines are particularly difficult tance levels of three elite populations. to convert to MSV resistance. Some of the new • State scientists can use many breeding strate- MSV-resistant inbred lines found among gies to develop MSV-resistant maize CIMMYT’s other maize lines (CML and downy germplasm. These include: mildew-resistant lines from CIMMYT Thailand), - Evaluating existing streak-resistant germplasm originated from germplasm not previously known - Converting locally adapted germplasm to to contain MSV resistance genes (Diallo 1996). MSV-resistant varieties through line recycling, Furthermore, Pixley (1997) reported that Tuxpeno backcrossing introgression, and mass selec- Sequia, a CIMMYT maize population from Mexico, tion was unexpectedly found to contain significant - Developing new MSV-resistant open-polli- resistance to MSV in CIMMYT’s Harare nurseries nated varieties (OPVs) and/or populations by in 1995 and 1997. The dosage effect of streak recombining elite streak-resistant inbred lines resistance genes in different types of hybrids - Developing MSV-tolerant hybrids using involving resistant and susceptible parents was resistant inbreds as one or two parents. studied and reported (Diallo 1996). References There have been reports of variation in MSV Bjarnason, M. 1985. Progress in breeding for disease development and of MSV-resistant resistance to the MSV disease. p 197-207. germplasm becoming susceptible when exposed Diallo, A.O. 1996. West and Central Africa, Lowland to various strains (Rodier 1995, Njuguna 1996). Tropical Sub-Program Annual Research Such changes were generally small, however, and Report, 1996. CIMMYT-IITA, Bouake, Cote Dr. Diallo emphasized the need for dialogue and d’ Ivoire. caution: “We are not aware of instances where the Dintinger, J., A. Pernet, A. Rodier, B. Reynaud, and resistance in MSV-resistant lines developed by J.L. Marchand. 1997. A complete resis- CIMMYT at Harare has broken down or become tance to MAV originating in Mascarene more susceptible. Anyone who has facts and data germplasm, inheritence and gene map- is very welcome to come to us.” Indeed, resistance ping. Abstracts of Maize Streak Sympo- through the cultivar revolution has never been sium, 9-11 September 1997. ARC Grain disproved elsewhere (Rodier 1995). Dosage Crops Institute Private Bag x1251, effects on MSV resistance genes in different Potchestroom 2520, South Africa. hybrids involving resistant and susceptible parents Du Pleis, J.G. 1997. The Vaalharts Composite as have been studied and reported (Diallo 1996). source of streak resistance. Abstracts of Maize streak symposium, 9-11 September Highlights of CIMMYT’s breeding and resistance 1997. ARC Grain Crops Institute Private efforts: Bag x1251, Potchestroom 2520, South Africa. • In addition to its facilities at Harare, Zimbabwe, Efron, Y., S.K. Kim, J.M. Fajemisin, J.H. Mareck, CIMMYT helped to establish MSV screening C.Y. Tang, Z.T. Dabrowski, H.W. Rossel, facilities at Embu, Kenya. and G. Thottappilly. 1989. TITLE??? Plant • A collaborative MSV screening project has also Breed. 103:1-36. been initiated in Uganda under the institution’s Kim, S.K., J.M. Fajemisin, and I.W. Buddenhagen. African Maize Stress (AMS) Project, which is also 1989. Mode of gene action for resistance designed to address other factors affecting to MSV. Crop Sci. 29:4. maize production, such as drought, low nitrogen Nhlane, W.G. 1997. Genetic analysis of MSV levels, striga disease, and stem borers. disease in maize, Abstracts of Maize • With 20 years of experience in breeding for MSV Streak Symposium, 9-11 September 1997. resistance and the availability of numerous MSV- ARC Grain Crops Institute Private Bag x1251, Potchestroom 2520, South Africa.

7 Pixley, K.V. and B.T. Zambezi. 1996. Maize Combating MSV at IITA germplasm available from CIMMYT Screening techniques. At present, IITA can Zimbabwe. CIMMYT Harare, P.O. Box MP produce 200,000 viruliferous leafhoppers to infect 163, Mont Pleasant, Harare, Zimbabwe. 50,000 maize plants per week, with a susceptible Rodier, A., J. Assie, J.L. Marchand, and Y. Herve. check planted at regular intervals to monitor 1995. Breeding maize lines for complete uniformity of screening. IITA has developed an and partial resistance to MSV. Euphytica effective screening technique by mass rearing 81:57-70. both C. mbila and C. triangulata in glasshouses, Welz, H.G., A. Schecert, A. Pernet, K.V. Pixley, and using maize plants for both oviposition and rearing. H.H. Geiger. 1998. A gene for resistance to The adult leafhoppers are transferred to cages the MSV in the African CIMMYT maize with MSV-infected plants for 24 h before being inbred line CML202. Molec. Breed. 4:147- released en masse into large screenhouses. This 154. procedure was useful for identifying sources of resistance but not ideal for evaluating many segregating populations. 2.1.3 West Africa: Breeding for MSV Resistance—IITA Experience Sources of resistance. IITA primarily uses maize (Dr. S.O. Ajala) germplasm IB32 as a source of MSV resistance. IB32 originates from a maize population, TZ-Y, IITA’s experience and work in Nigeria shows that derived from a cross between Tuxpeno Planta successfully developing MSV-resistant maize Baja and an unknown yellow germplasm source varieties requires an efficient screening technique from East Africa. In 1976, MSV resistance was also to achieve uniform infestation, identification of identified in La Revolution from Reunion Island. sources of resistance, and coherent breeding Much work, however, has been done on TZ-Y, and strategies. one of the S1 lines identified from it, designated TZ-Y-32, had all its plants rated as highly resistant. IITA considers breeding for resistance the most This line was advanced to S6 and designated as appropriate means of controlling the disease, and IB32. Its MSV resistance is characterized by very we have found that recurrent selection and back- few, short, white streaks on leaves, no yield loss, crossing are efficient ways of developing MSV- and no stunting—unlike highly susceptible lines resistant germplasm. Currently, all germplasm characterized by 75% of the leaf area being originating from IITA are MSV resistant. covered by numerous long streaks, severe stunting, and death. Environmental conditions including rainfall and temperature are causally related to disease At IITA, plants used as MSV resistance sources outbreaks. In West Africa, epidemics are associ- had to at least show a trace of the streak, which ated with drought or irregular rains (IITA 1983). indicated that they had truly been infected but Studies conducted in Nigeria and Togo could resist the reproduction and spread of the (Dabrowski and Okoth 1985) have also showed virus. To represent degrees of MSV resistance, a that the vector C. mbila is the most predominant six-point rating system has been developed. In species in cool and mid-altitude environments general, plants rated 0 are immune, while those (1,000 m-1,800 m). This, however, was not the case rated 1 are highly resistant. Ratings of 2-3 are when temperatures exceeded 28 oC, since the tolerant to moderately tolerant, while 4 and 5 are critical maximum temperature for the development sensitive (susceptible/intolerant) to highly sensi- of C. mbila is 25 oC (Rose 1973, van Rensburg tive. 1982). Natural populations of C. mbila and C. triangulata tend to have more active transmitters IITA’s MSV Resistance Rating System than other species since the proportion of females 0 = No symptoms is 2-3 times higher. Females are better transmitters. 1 = Very few streak on leaves. 2 = Light streaking on older leaves, gradually decreasing on younger ones. 3 = Moderate streaking on old and young leaves, slight stunting. 4 = Severe streaking on 60% of leaf area, plant stunted. 5 = Severe streaking on 75% of leaf area, severely stunted or death.

8 Inheritance of MSV resistance. The mode of which resulted in a narrow genetic base. As a resistance to MSV in IB32 has been studied using result, other germplasm were introgressed into generation mean analysis involving four suscep- these populations to form TZSR-W-1 and TZSR-Y- tible inbreeds: B14, B68, B73, and Mo17. These 1, respectively. To form these new populations, were crossed to IB32 to produce F1, F2, and TZSR-W was crossed with TZB and TZPB from backcross generations. All plants were artificially Nigeria, and Pop 21 and Pop 22 from CIMMYT. infected with 4-5 viruliferous leafhoppers placed at The other line, TZSR-Y, was crossed with Poza whorls of individual plants at 10 to 12 days after Rica 7428 from CIMMYT, 096EP6 from Nigeria, and emergence. Symptom expression on the segre- the cross between the two resistant sources IB32 x gating population was continuous. La Revolution. Half-sib family selection under MSV pressure selection was used to improve the new Study results suggest that MSV resistance in IB32 populations. The best 50 families were selected in is qualitative (oligogenic). There are 2-3 gene each group to form TZSR-W-1 and TZSR-Y-1 (Kim pairs involved, and modifier genes also play a role 1981). Early maturing populations TZESR-W and in disease expression. Bjarnason (1984) noted that TZESR-Y were also developed during this period MSV resistance in La Revolution is monogenic, (Kim 1981). These two early populations were while Efron (1989) reported that segregation of formed by crossing TZSR-W and TZSR-Y with several S1 lines from different sources of resis- early maturing germplasm mainly from Burkina tance identified in Tanzanian maize populations Faso. Crosses were inbred to the S3 level under was discrete. The S1s segregated in approxi- artificial MSV pressure and selected lines were mately the ratio of three immune plants to one recombined (Kim 1982b). Populations were susceptible plant. This suggests monogenic subsequently subjected to recurrent selection, inheritance. A quantitative train analysis (QTA) of a mainly reciprocal full-sib family selection schemes. CIMMYT maize line (CML202) also strengthens the Intrapopulation schemes were also employed to theory that resistance is monogenic (Weiz 1998). build up the frequency of MSV resistance genes in several populations. In addition to MSV resistance, IITA’s experience shows that by using a simple there has been progress in improving agronomic selection procedure, including backcross breed- parameters and yield potential (Table 1). ing, resistance levels can be raised from a mere 2% to 100% in a few generations of selection under 2.2 Status of MSV in Various African MSV pressure. Because susceptibility can almost Countries be eliminated in just two generations, it is highly likely that only a few genes condition resistance to 2.2.1 MSVD Occurrence, Development, and MSV. Distribution in Malawi (Dr. P. Ngwira)

Development of adapted and resistant MSV Overview genotypes. TZSR-W and TZSR-Y were the first Maize streak virus disease (MSVD) was first two populations with high MSV resistance. They reported in 1953 (Wiehe) and has since increased were developed by crossing resistant plants from in occurrence and severity. It is associated with TZ-Y with TZPB, an MSV-susceptible white variety leafhopper vector population growth and their with high yield potential. The cross was separated infectivity rate. The major maize disease in areas into white and yellow grain types to form the two below 800 m, it has spread to all maize-growing populations TZSR-W and TZSR-Y. There was parts of the country. Malawi is setting up a screen- some drawback because each population was ing facility at Chitedze in collaboration with formed by recombining only 10 full sib families,

Table 1. Grain yield, streak resistance level, and other agronomic characteristics of different cycles of La Posta under artificial streak infestation at Ibadan, 1986

Cycle Grain yield (t/ha) Streak levela Streak plant (%) Plant height (cm) Days to 50% silk C4 3.0 3.8 69.1 164.0 54.5 C5 2.7 3.3 60.0 167.8 54.3 C6 4.8 1.7 10.2 187.5 53.3 C7 4.6 1.2 0.0 190.5 53.5 Source: Efront et al (1989). aStreak level: 1= resistant, 5=susceptible.

9 CIMMYT and is starting to focus on breeding and Twenty planting stations with three plants per developing MSV-resistant varieties. station were chosen along the path. Plants were assessed for the presence and severity of MSV In 1998, the MSV-resistant variety Pan 6193 was disease or infection. Infected plants were counted introduced to farmers. In the past, breeding work and divided by 60 multiplied by 100 to obtain a concentrated on high-yielding varieties and not on percent incidence per field. On average, eight disease resistance, so there is still much work to fields were surveyed in each Agricultural Develop- be done. ment Division. The minimum size of the fields was 0.5 ha. Samples were also collected for lab In terms of vector research, studies conducted at confirmation and the identification of other viral the Chitedze station show that Malawi has three diseases. All samples were sent to the Ohio main Cicadulina species: C. storeyi, C. parazeae, Agricultural Development Center, Ohio State and C. mbila. Their densities peak between March University, with duplicates for comparison sent to and April when rains decline. This study examined the Crop Science Department of the University of MSV development in maize planted at different Zimbabwe. Grain yield and rainfall data were also times during the growing season and included an collected. Statistical and regression analysis was analysis of the relationship between Cicadulina conducted on the data using the MSTATC (1988) leafhopper population growth, rainfall distribution, statistical package of Michigan University. and MSVD development. Results and Discussion Materials and Methods Species composition. Among the specimens Plantings were made every 14 days. The first sent to IIE (UK), C. mbila was the most predominant began on 7 December 1988; in the second year it of the three species identified, followed by C. began on 5 November 1989 using a local maize storeyi and C. parazeae. Vector populations were variety. All the plots of 4 ridges (5 m x 0.9 m) low at the beginning of the rainy season; the received a basal dressing of 60 kg/ha phosphorus greatest growth occurred in March and April when supplied as 20-20-0, 120 kg/ha supplied as 20-20- the rainy season tapered off. 0, and calcium ammonium nitrate as a topdressing. Population density and species data was obtained 2.2.2 Country Report: Uganda (Dr. V.A. Okoth) by daily catches of the vector (Cicadulina). Yellow cylindrical traps were placed around the trial plots, Overview and collected insects were sent to the International In Uganda, 60% of the maize planted by farmers is Institute of Entomology (IIE) in UK for identification. local and unimproved, 35% is improved open- These identified specimens were used to deter- pollinated, and 5% is hybrid. And while maize mine different types of species and their population hectarage in East Africa—including Uganda—may be densities. increasing, yields are still very low, averaging 0.8- 1.2 t/ha (Annon 1997). Reasons for these low yields MSVD incidence and development were moni- are lack of high-yielding cultivars, drought, pests, tored and evaluated by counting the number of diseases, and poor soil management, including virus-infected plants at 7-day intervals in each declining soil fertility. The most important pests are planting, beginning 2 weeks after emergence. The lepidopteran stem and cob borers belonging to the percentage of infected plants gave the disease families Noctuidae and Pyralidae (Gounou 1994). incidence: Chilo partellus swinhoe, Busseola fusca Fuller (Noctuidae), termites, grain weevils, Sitophilus Disease No. of MSV-infected plants x 100 spp., and Mussidia nigrivella also cause serious incidence = losses to maize in storage (Gounou 1994). The (DI) Total plant stand count parasitic weeds Striga hermonthica and S. asiatica are also a major threat to maize crops. The important maize diseases in Uganda and the MSVD distribution. A countrywide survey to region include maize streak virus (MSV), northern determine the distribution and prevalence of MSV corn leaf blight, rusts (Puccinia sorghi Sch and P. disease was conducted from 1988 to 1990 by polysora Uner), common smut, Ustilago maydis sampling fields 25 km apart along the main roads. (DC) Cda, and ear rot caused by Fusarium MSV infected plants were sampled along a zigzag monilliforme Shed., and Diplodia maydis (Berk) path 25-50 m long, depending on the size of the Sacc. (Annon 1992). Furthermore, two other dis- field. eases have recently gained prominence: sorghum

10 downy mildew, caused by Peronoscelerospora Maize Research sorhi C.G. Shaw (Weston and Uppal), and gray leaf The Institute of Agricultural Research coordinates spot, caused by Cercospora zea-maydis (Tehon the maize research activities of a team of special- and Daniels). ists in various fields like breeding, weed science, agronomy, physiology, entomology, pathology, Maize streak virus surveys. Uganda has also soil and plant nutrition, farming systems research, undertaken surveys on MSV and created a mass and extension services. Other institutions involved rearing system for the vector. Surveys conducted include the Alamaya University of Agriculture, the in four districts (Mubende, Masindi, Tororo, and Awasa College of Agriculture, nongovernment Kumi) showed that MSV incidence ranges from 8% organizations (NGOs), and government organiza- to 29.5% even in nonepidemic periods. It was, tions. however, generally lower in the crop planted early in both the first and second rainy seasons. MSV Field observations of the 1992-93 crop season at was also more severe in the Mubende and Kumi Gambella, which is 500 m above sea level, districts. showed that the MSV infection rate was 80% at research stations and up to 20% in farmers’ fields. The 1990 surveys carried out in these four districts Another field observation in 1996-97 was con- also showed that C. mbila, C. triangula, C. ducted in the north at Gojam, Tigray, which is 2,530 arachidis, and C. chinai were common. Bioassays m above sea level. Researchers recorded MSV obtained by caging individual insects collected infection rates of 60%. from the field on maize varieties susceptible to MSV disease showed that between 1.3% to ---% MSV vectors in Ethiopia include C. mbila, C. were already infected. Although the percentage bipunetella, C. ghauri, C. storeyi, and C. niger. seems low, on a large geographic scale it can result in many infective leafhoppers capable of causing widespread epidemics.

Data on insects collected in the June and October 1990 surveys indicated that the Masindi district, with a mean of 2.7%, had the highest percentage of naturally infective Cicadulina. Other studies carried out during 1990-93 include analyses of the com- parative biology of Cicadulina species from various districts, host plant suitability, and adapta- tion to artificial rearing. Pearl millet was found to be the most responsive to the rearing system and was also the most prolific and efficient in virus transmis- sion. Trials were also carried out in the screenhouse to simulate dry conditions to study what happens when host plants dry up and be- come unsuitable. CIMMYT was also involved in the collaborative MSV screening program initiated in Uganda.

2.2.3 MSV in Ethiopia (Dr. A. Yusuf)

Introduction Begun in 1950, Ethiopian maize research is nowadays coordinated under a national network. Maize is grown in nearly 1 million ha. MSV was identified and recorded in 1967, and in 1980 the gemini virus was diagnosed using electron microscopy. MSV infestation reaches 80% at research stations and up to 20% in farmers’ fields MSV mid-season infected field causing poor comb at Gambella in the southwest. set without grains.

11 The objectives of maize research include: • Study vector dynamics, population, and immigra- tion • Identify maize viruses in major regions of the country Finally, we believe that the most promising way to • Establish a screening center control MSV is through developing MSV-resistant • Determine the most prevalent maize viruses of varieties. economic importance with particular emphasis on MSV in four regions

3. The Ecology of MSV

3.1 Vectors major sex-linked chromosome (Storey 1932). Based on the latency period and the persistence 3.1.1 Cicadulina spp. and MSV (Dr. V.A. Okoth) of transmission efficiency in spite of the very short acquisition access period, Storey postulated that MSV multiplied in the insect. This, however, was Overview and Introduction doubted by Bawden (1964). Maize streak virus was first described as mealie variegation by Fuller in 1901. In 1925, Storey Bouton and Markham showed that the concentra- suggested that the infectious agent was a virus. tion of the virus in C. mbila decreases after transfer MSV belongs to the gemini virus group and was from infected maize to an immune plant (1986). first purified by Bock, Guthrie, and Woods in 1974. Combined with the inability to detect replicate DNA The virus was first recorded in South Africa where it forms of the virus in the insect, this led them to occurred in maize (Zea mays L.), oats, and conclude that MSV did not multiply in C. mbila. grasses (Fuller 1901). It also infects other grain Reynauld and Peterschmitt (1992) also noted that crops like wheat, barley, finger millet, pearl millet, MSV does not multiply in C. mbila. They also oats, and sorghum. Spread by leafhoppers, MSV observed that the transmission efficiency of C. does not replicate in the insect vector. In Kenya, 17 mbila was subjected to 50-h acquisition access grass plants have been reported to serve as MSV periods (AAP) and that there were more ELISA- hosts. Farmers in many countries have reported positive insects after 50-h AAP than 3-h AAP. A heavy yield losses during epidemics. MSV mean value of 0.36 ng of MSV per C. mbila was reduces plant height by 50% and yield losses found 3 days after the 50-h acquisition, whereas 14 reach 100%, especially if plants are infected days later there was 0.20 ng of virus per insect. before 7 weeks. High yield losses have been When leafhoppers were kept on infected maize, recorded in various parts of Africa, including Kenya however, they substantially accumulated MSV up (100%), Nigeria (100%), and Mauritius (91 %) to an average of 3.83 ng per insect after 35 days of (Fajesmisin 1984). continuous acquisition. The amount of virus detected in females was usually greater than the MSV Transmission (Vector-MSV studies) amount detected in males. The virus is transmitted by several species of leafhoppers in the genus Cicadulina (Cicadellidae, Studies carried out in Uganda from 1990 to 1993 Homoptera). Cicadulina mbila (Nude) is the main showed that pearl millet was the most suitable host vector species (Rose 1983). Storey defined the plant for rearing Cicadulina spp., while C. virus cycle in the insect and showed that acquisi- triangulata was the most prolific and efficient in tion is very rapid (less than a minute) if the insect transmitting the virus. MSV is exclusively transmit- feeds on mesophyll or phloem in chlorotic areas of ted by Cicadulina spp. Accordingly, researchers infected leaves (1933, 1938). The minimum latent recommended cultural and chemical methods to period varies from 6-12 h depending on the reduce vector population density and migration temperature (Storey 1928). According to Storey, into maize fields. Unfortunately, the main cultural only active races of C. mbila can transmit the virus method—rouging infected plants—actually increased (1933). In other races, however, the virus cannot MSV infection unless infected plants were sprayed. cross the gut wall. This permeability depends on a

12 3.1.2 Cicadulina Population Dynamics and Materials and Methods Species Distribution (Drs. M. Gethi, Field surveys to determine leafhopper population K.A. Nyarko, and B.O. Odhiambo) and species diversity began in June 1997. Areas selected were spread over about 10 maize- Overview and Introduction growing sites in Kenya and are in the different MSV evolved with wild grasses and has been ecozones described below: recorded in 18 wild grasses, most of them annual species (Onudi 1996). The size of the vector 1. Lake Region Ecozone (Mbita Point, Oyugis) population mostly depends on the number of host This area borders Lake Victoria, with altitudes of grasses suitable for oviposition and nymphal 1,000-1,200 m above sea level. Rainfall is bimodal development. The population, therefore, is lowest (750-1,200 mm), which allows for two crops per at the onset of the rainy season and highest at the year. Temperatures and relative humidity range beginning of the dry season. Seasonal abundance from 30 oC to 34 oC, creating moisture deficits in and aerial densities of trapped Cicadulina peak some areas. twice annually—in July-August and November- December. Relay cropping allows leafhopper 2. Western Region (Kiminini, Amagoro, and Kitale) populations to persist to the next season. The This area covers Kenya’s western province and a vector can also fly long distances, especially at the part of the Rift Valley province. Altitude is about end of the rainy season. 1,500-2,500 m above sea level, and the rainfall is the country’s highest (1,000-1,780 mm). Depending Studies on the Cicadulina biology and field on the elevation, the area has either a bimodal or population of various species have indicated unimodal rainfall pattern. similarities in life cycle (Rose 1978). Recently, Asanzi et al (1995) indicated that survival and 3. Central Ecozone (Kagumo and Mitunguu) development of most Cicadulina species de- This area covers some parts of the central and creased as temperature increased, while longevity eastern provinces and is referred to as the high- increased as temperature decreased. land east of the Rift Valley. The climate is influ- enced by both Mt. Kenya and the Aberdare The size of the hopper population mainly depends Ranges. Rainfall is bimodal with an average 750 on the number of host grasses suitable for oviposi- mm per annum. MSV is endemic to this area. tion and nymphal development. These numbers are determined by rainfall patterns and result in 4. Coastal Ecozone (Mtwapa) decreased population at the onset of rains. They These are known as the coast lowlands, lying 20- are highest at the beginning of the dry season 125 m above sea level. Temperatures and relative (Okoth and Dabrowski 1987). humidity are relatively high: 26-35 oC and 60-90%, respectively. The mean annual rainfall is about 700 Fajemisin et al (1986), working in Nigeria, also mm spread over two seasons in a bimodal pattern. demonstrated that there exists a relationship Most rainfall is received during the long rains; the between weather conditions, leafhopper develop- short rains are unreliable. ment, growth of grasses, and the rate of virus transmission to maize. Observations by Rose Leafhopper sampling. Sites that were used for (1973) also indicated that Cicadulina can fly long virus epidemiology studies were located in areas distances particularly at the end of the rainy where maize was the main crop. These areas also season. Dabrowski (1983) showed that Cicadulina had various grass species throughout the year. A leafhoppers existed in two morphological forms, modified cage as described by Okoth and long and short, depending on their mean flight Dabrowski (1987) was used to sample the leafhop- duration. pers. It measured 1.0 x 1.0 x 1.25 m, and was made of metal frame covered with a transparent nylon In this study, field surveys were carried out at 10 mosquito netting. sites in different ecological zones. It was noted that even resistant materials brought in after the 1988 To trap the leafhoppers, the cage was gently lifted MSV outbreak were more susceptible to infection and swiftly placed to cover fresh grasses border- and that there were indications of resistance ing the maize. A person then entered the cage breakdown. through a side slit and shook the grasses to disturb the insects. All insects that respond to light would then attach themselves to the net side. The

13 Cicadulina leafhoppers were then aspirated At the Mituguu site, which is an irrigated area, the individually. Twenty-five samples were taken at population increase followed the expected pattern— each site and the number of Cicadulina caught per low population at the beginning of the long rains square meter was recorded. and a peak at the end of the rainy season (1.3 hoppers/m2) in August. The population then Male leafhoppers were separated from the declined until December 1998 and started rising females. Females were then caged on living plants again. grown on trays and then taken to the laboratory, where they were reared singly using PVC cages (5 At Mtwapa in the coastal region, the trend was x 25 cm). They were allowed to oviposit and the somehow oscillatory, with the highest density in the first generation progeny was then identified using middle of the long rains in April, September, and the pygofer process and the aedeagus, which are December 1998. Population trends for Mwea morphologically different in Cicadulina species. started rising in January and peaked in February, which marks the real dry period. The population At Embu and Mbita Point, a Malaise trap obtained declined and had a small peak again in May. This from the UK was set up to trap hoppers flying at meant that, as expected, the population increased least 1 m above the ground. At Mbita Point another during the rainy period. trap known as MPFS pylon, a type of stationary yellow trap, was set. The sticky traps were At all sites, however, population trends were mounted on poles using a pulley mechanism at 1- affected by increased rainfall due to El Niño. In m intervals up to 12 m high. some cases, flooding of sampling areas reduced insect numbers. The trap would monitor and determine the direction of insect flight to establish the importance of Species composition. Data on species composi- migration in MSV epidemiology. Leafhopper tion and mean leafhopper density across different sampling on grasses was also carried throughout agroecozones are shown in Table 2. Six species the year to establish other possible alternative of Cicadulina were recorded. These were C. hosts. bipunclata, C. chinai, C. ghauli, C. mbila, C. latens, and C. storeyi. C. mbila was found in all zones, The spatial distribution of MSV and the percentage of infestation in maize fields surrounded by Setaria sphacelata was assessed at 2-week intervals using a quadrant and the efficiency of MSV trans- Table 2. Species composition and density of Cicadulina mission by different species of collected leafhop- pers determined. Each species was replicated 30 Site Species Density/m2 times to determine transmission rates. Results indicated that species from different ecological Lake Region Oyugis C. storeyi 0.2 zones might vary in transmission efficiency. C. chinai Mbita C. chinai 2.0 Results C. storeyi Cicadulina population density. Throughout the C. mbila sampling period, Cicadulina population density HomaBay C. storeyi 2.0 varied in different locations. But at all sites there Western Region was a population buildup in January, which also Kitale C. mbila 0.2 marks the end of short rains (except for the Mtwapa Amagoro C. mbila 0.7 site in the Coastal region, where the population Kimilili C. mbila 0.8 peaked in April 1998). Central region Mitunguu C. mbila 2.3 This deviation from the normal trend was attributed C. ghauri to increased rainfall in the region, which preceded C. bipunctata the El Niño rainfall. In Embu and Kagumo, the trend Jamaica C. mbila 2.9 was similar to another population peak in May C. bipunctata 1998 (1 to 1.3 leafhoppers/m2). The two areas also Kagumo C. mbila 2.1 experienced high rainfall. C. bipunctata Embu C. mbila 1.7 Nyeri C. mbila 0.6

14 while C. chinai and C. bipunctata were only found in Additional data on the alternative breeding hosts of the Lake region (Table 2). Cicadulina spp. indicated that they inhabit different wild grasses. Nymphs were recorded on several Samples from the Coast were not identified grasses, Setaria sphacelata, var. splendida, because only females were caught and they died Echinochloa spyramidalis (Lam), Hyparrhenia rufa, before reaching the laboratory (Table 2). The Setaria incrassata (Hochst), and Sorghum vulgare. mean population density for all species ranged from 0.2 hoppers/plant in the lake region (Table 2) Discussion to 2.9 hoppers/m2 in the central zone (Jamaica site) Population dynamics of Cicadulina leafhoppers (Table 2). C. mbila was the dominant species in all was dependent on the moisture available in other areas except the lake region, where C. chinai different agroecological zones. The apparent was dominant. decrease in leafhoppers at Mitunguu irrigated area during the usual wet season of April to May could The differing composition of species in all zones be explained by the drought between the pre- explains the different levels of MSV in different sumed long rains and the onset of irrigation. areas. The central region has the highest insect density and is a hot spot for MSV, which is not Results from Mwea rice irrigation area indicated surprising since high C. mbila density is associ- that the draining of water during harvest also ated with high disease incidence (Table 2). reduced leafhopper population. Wild grasses, which could serve as alternative hosts also dried Relationship between maize invasion up due to lack of paddy water. Rose (1972 and and alternative hosts. The spatial distribution of 1973) suggested that leafhoppers migrate at the MSV-infected plants and the infestation percent- onset of the dry season to areas with fresh age in maize fields surrounded by Setaria grasses. sphacelata was not correlated. The study also confirmed reports by Rose (1978) Efficiency of MSV transmission by vectors and Webb (1987) that C. mbila, C. chinai, and C. from different agroecozones. Results indicate bipunctata occur in Kenya. C. mbila, which is the that species from different ecologies might differ in most effective transmitter of MSV, formed the their transmission efficiency. It is evident that in all largest percentage except in the lake region, while tests conducted, the C. mbila and C. storeyi the central highlands had the highest density. C. collected from the lake region (Oyugis and Mbita) storeyi was the dominant species in the lake were more efficient in transmission than the C. region. mbila collected from the western region (Kitale). Which further explains the disparity observed in streak incidence between the two areas.

Field testing of F3 parental lines from crosses made using MSV- resistant and local MSV- susceptible germplasm and Cicadulina vector.

15 The average density was between 0.2 to 2.9 Okoth, V.A.O. 1985. Some characteristics of hoppers/m2. Cicandulina spp. population associated with maize streak virus in Nigeria. Ph.D. Future Plans for Vector Ecology and thesis. IITA. University of Reading, U.K. Epidemiology Okoth, V.A.O and Z.T. Dabrowski. 1987. Compara- Factors contributing to vector and MSV outbreaks tive biology of some Cicandulina species are not yet well understood, and there is a need for and population from various climatic zones more studies on the spatial and temporal distribu- in Nigeria. Bull. Entomol. Res. 77:1-8. tion of leafhoppers. More data are also needed on Okoth, V.A.O., M. Gethi, D.M. Kalengesa, and J.O. environment variables, cropping pattern, vector Obara. 1989. Cicandulina species compo- density, disease incidence, severity, and preva- sition and incidence of MSV disease in lence. selected maize growing areas of Kenya. ICIPE Report 1989. There is also a need to clearly establish the Onudi, B.O. 1996. Occurrence of wild grasses relationship between vector migration and disease geminivirus in Kenya and their importance outbreaks, while the use of DNA markers will help as a source of inoculum for maize. Ph.D. in detecting genetic variability. Future plans also thesis. University of East Anglia. 145 p. include detecting other viral diseases and vectors, Rose, D.J.W. 1972. Times and sizes of dispersal documentation of alternative reservoirs, and flight by Cicandulina spp. (Hemiptera: developing MSV prediction models. Cicanddelidae). Vector of maize streak disease. J. Anim. Ecol. 41:495-506. References Rose, D.J.W. 1973. Distance flown by Cicandulina spp. (Homoptera: Cicadellidae) in relation Asanzi, C.M., N.A. Bosque-Perez, I.W. to maize streak disease in Rhodesia. Bull. Buggenhagen, D.T. Gordon, and L.R. Entomol. Res. 62:477-495. Nault. 1994. Interaction among maize Rose, D.J.W. 1983. The distribution of various streak virus disease, leafhopper vector species of Cicandulina in different African population and maize cultivars in forest countries, frequency and their attack and and savannah zones of Nigeria. Plant impact on crop production. Proceedings Pathol. 43:145-167. 1st International Workshop on Leafhoppers China, W.E. 1926. A new genus and species of of Economic Importance. CIE Publ. 297- Jassidae injurious to maize in Kenya 304. colony. Bull. Entomol. Res. 17:43. Storey, H.H. 1925. The transmission of streak China, W.E. 1928. Two new species of Cicandulina disease of maize by leafhoppers from the Gambia West Africa. Bull. Bolcultha mbila. Ann. Appl. Biol. 39:1-30. Entomol. Res. 77:53-56. Dabrowski, Z.T. 1983. Identifying and collecting Cicandulina for maize streak resistance. 3.1.3 Discussions, Unanswered Questions IITA Res. Briefs 4:2-3. Dabrowski, Z.T. 1987. Two new species of Prof. Peter Markham gave two presentations: Cicandulina china. (Hemiptera: “Leafhopper or vector feeding behavior” and “MSV Euscelidae) from West Africa. Bull. Entom. transmission and agroinoculation.” His papers left Res. 77:53-56. participants with more questions than answers, Fajemisin, J.M., S.K. Kim, Y. Efron, and M.S. Alam. which indicates just how much more there is to 1982. Breeding for durable resistance in learn about the virus, its vector, and maize crops. tropical maize with special reference to MSV. FAO IITA Expert Consultation on In the first presentation he reminded the partici- Durable Resistance Breeding. p 30. pants that only a handful of insect species are Fajemisin, J.M., Z.T. Dabrowski, Y. Efron, and S.K. involved in disease transmission. Also, when it Kim. 1986. Weather factors associated with comes to MSV, which is indigenous to Africa, we recurring maize streak epidemics. Pro- may need to look at indigenous crops that are ceedings of Seminars on resistant to MSV, such as sorghum, to better agrometeorology and crop protection in understand resistance mechanisms. the lowland humid and subhumid tropics, Colonou, Benin. p 1.

16 He then described how the virus is transmitted. He said that when the project was conceived it MSV creates “ideal” conditions that facilitate its tried to do something more ambitious, something spread by the vector Cicadulina. It uses the insect sustainable. MSV is an African problem that as a vehicle, hijacking the bug for a lift from plant to Africans are going to solve. The infrastructure plant. In fact, MSV-infected maize even tends to benefits are too great and the potential technolo- have a yellowish color that attracts the vector: “It is gies too valuable for such research to be a “pack like the virus is calling the insect vector to come and go” type of activity. and pick it up. It is as if the insect is under the influence of the virus because it is delivered in the Finally, Dr. Markham praised the project’s attempts maize plant’s phloem where it can be easily to improve sustainable agriculture and regional circulated into plant cells or various appropriate collaboration. As he observed, when individuals tissues.” The insect picks up the virus from the work together they can obtain answers to shared mesophyll, which it then infects via the phloem. questions and find missing links in research more Within 6 h it is ready to be delivered. Everything quickly while avoiding the pursuit of dead-end happens so quickly that within 6 h the virus can be inquiries. Everyone benefits from less waste and transmitted again by the insect vector. The effects, faster research. Collaboration also makes it easier as everyone knows, are very bad: maize plants to seek more support or funding. He urged all to attacked at the sixth leaf or earlier have signifi- work carefully but boldly, to identify questions cantly reduced leaf area and plant height, while whose answers will make an impact, and to think of grain yields decline by 90%. what could be done with available methodologies. He explained that the vector picks up many MSV Much training and technology has been trans- viruses and not individual or single ones. Since ferred, and it must not go to waste. Although MSV does not replicate in the insect—as declining progress has been made, we still face the chal- viral titers indicate—the amount of virus taken up by lenge of predictably controlling MSV. Furthermore, the insect tends to last a lifetime. there is a large pool of viruses out there ready to mutate, and in the future even immune lines may The relationship between MSV and the host or break down. Our work in biotechnology training has insect vector, Cicadulina spp., is very complex and been a success, but this is no time to sit on our there is much more to learn. We must go forward in laurels. We must all strive to keep up with new a zigzag course and change our ideas when developments and opportunities. necessary. For now, not answers but questions structure the field. Why are MSV epidemics 3.2 Disease and Epidemiology increasing? What is tolerance? Is tolerance in other plants funneling the virus into maize plants? Should 3.2.1 Introduction and Overview we be looking for immunity? What obstacles are there to finding immunity? We know that maize did MSV disease is characterized by chlorotic lines not come from Africa, yet it has genes that are along veins of maize leaves. In highly susceptible resistant to MSV. Are there some overlapping varieties, chlorotic streaks tend to coalesce into resistance genes in cereals or grains? What can uniformity, which frequently results in death. The we learn about this in the last 60 years of plant disease has been observed in many weeds and research? Does it provide the information that crops belonging to the family Graminae. In a study breeders are looking for? entitled “Variation of MSV in Kenya,” Dr. Jackson Njuguna studied and assessed levels of MSV Dr. Markham explained that the Green Revolution resistance in maize germplasm for use in national left all genes for resistance to insects untouched maize breeding programs. and that there was considerable potential here. In the meantime, he suggested that we try to better Dr. Njuguna and his team studied the biological understand the insect vector. Can we stop the difference of MSV isolates in the highlands east of insect from feeding on maize crops? What is the the Rift Valley with bimodal rainfall, the highlands source of the insect? Can we predict an outbreak west of Rift Valley with a unimodal rain pattern, the or when it is preparing to migrate by looking for Lake Basin with its bimodal rain pattern, and the changes in its physiology, morphology, or chemis- Coastal lowlands with bimodal rainfall. Their work try? Which species move locally or migrate long showed that: distances? What are the prime ecological zones that the insect vector favors? • Both CIMMYT and CIRAD have maize germplasm with very high resistance to MSV.

17 CIRAD390 and OSU23I from CIMMYT demon- extremely good gene product may, for example, strated a high level of MSV resistance. These encapsulate all other viruses.” He stated that a were immune to MSV in most locations. The two, difference of 2.5% in the genome of different virus however, also succumbed to MSV in some strains was insignificant. locations. • There are at least two MSV biotypes in Kenya. 3.2.2 Studies on the Variation of Maize Streak The different MSV biotypes are not confined to Virus in Kenya (Dr. J.G.M. Njuguna) specific ecology but appear to be spread all over. The primary purpose of this study was to examine • PANNAR hybrid 5195 appeared to perform very the biological differences of MSV in Kenya’s well in most locations in its MSV resistance and different ecological zones. Four different ecologi- adaptability. cal zones were identified: • Glasshouse screening matched field results. 1. Highlands east of Rift Valley in a bimodal rainfall Dr. Njuguna recommended that farmers grow pattern PANNAR hybrid 5195 to avoid heavy losses (see 2. Highlands west of the Rift Valley with unimodal tables provided in Dr. Njuguna’s paper, section rainfall pattern 3.2.2). He also pointed out that there is 98% 3. Lake Basin with bimodal rainfall pattern. homology in the genomes of MSV isolates from 4. Coastal lowland ecology with a bimodal rainfall Nigeria, South Africa, and Kenya. But Dr. Martin pattern. Darren from the University Cape Town urged workshop participants to remember that in the MSV The other purpose was to assess levels of MSV replication process the best gene or gene prod- resistance in maize germplasm gathered from ucts can serve the whole community: “A virus does diverse breeding programs for potential use in not use only its own gene product. It also uses national maize breeding programs. those of all the other viruses in the cell, and one

KARI, CIMMYT, and Rockefeller Foundation representatives at a multilocation trial using MSV-resistant lines.

18 Materials and Methods serve as a differential among locations. This could Streak-resistant maize germplasm was obtained help reveal whether MSV varied in different from five different programs: the Summer Grains locations. Institute, CIMMYT, IITA, CIRAD, and PANNAR. One or more maize germplasm specimens from these The local check varied from location to location institutes were planted in a national multilocation and was usually a Kenyan commercial hybrid trial. A total of 18 maize types were planted in each popular in the specific trial area. These included of the 12 locations in the four agroecological zones the following hybrids: Hybrid 614, Hybrid 512, (Table 3). Of these genotypes, two were suscep- Hybrid 511, Hybrid 622, and Pwani Hybrid. tible to MSV. The remaining 16 were resistant or tolerant to MSV (see Table 4 for details about the All local checks were known to be streak-suscep- maize germplasm and their sources of origin). tible germplasm. The local check was also During the planning of the experiment, it was hoped planted around the experimental plots to serve as that a significant MSV epidemic would break out at guard and MSV spreader rows. The experimental each of the 12 locations and that one or more types design used in each location was a randomized of the streak-resistant maize germplasm would complete block design with two replicates. The experimental units were 3-m-long single rows, with 75 cm between rows and 30 cm within rows. Two Table 3. Description of multilocation trial sites (12) seeds were planted per hill and later thinned to one plant per hill. Diammonium phosphate (DAP) Name Description fertilizer was applied (250 kg/ha) during planting. Jamaica Cool wet highlands bimodal rainfall Fields were also topdressed with calcium ammo- Mucatha Cool wet highlands bimodal rainfall nium nitrate (CAN) at 250 kg/ha 7 weeks after Githunguri Cool wet highlands bimodal rainfall planting. Kagumo Cool wet highlands bimodal rainfall Kitale Cool wet highlands unimodal rainfall The crop was exposed to natural sources of MSV Kiminini Cool wet highlands unimodal rainfall inoculum and other diseases. Records were taken Amagoro Lake basin bimodal rainfall Oyugis Lake basin bimodal rainfall on the incidence and severity of MSV at each Mtwapa Hot humid coastal lowlands location. Notes were taken on the adaptability of Mtepeni Hot humid coastal lowlands the exotic streak-resistant germplasm and their Nyeri (FTC) Cool wet highlands reactions to other diseases.

A parallel experiment was set up in the glasshouse to gain a better understanding of the level of MSV resistance in various genotypes listed in Table 4. Viruliferous Cicadulina mbila was used to screen Table 4. Description of the 18 maize genotypes selected for the multilocation trial the exotic germplasm by determining MSV resis- tance levels in a controlled environment. Pedigree Description Source VHCY OPV S. Africa Nonviruliferous C. mbila was caged on healthy VHCW OPV S. Africa pearl millet (Pennisetum americunum) in an insect- AO89 Inbred line PANNAR proof glasshouse maintained at approximately 25 AO76 Inbred line PANNAR oC. Pearl millet was planted in a steam-sterilized HASR OPV Burundi potting mixture contained in plastic pots 12 cm in CML 202 Inbred line CIMMYT diameter and 12 cm in height. When the pearl millet CML 197[MSRXPOOL 9] Inbred line CIMMYT CIF2-205-1)osu23I) Inbred line CIMMYT dried up or became unsuitable for oviposition, it Pool 9A BSAF3 Population CIRAD was gradually replaced with younger seedlings, 4- [MSR/Pool 9A]C2F5 OPV CIMMYT 6 in tall. Maize seedlings for glasshouse screening Tzi3 Inbred line IITA were similarly raised. Five seeds were planted per KU144SR/(Tzi35) Population IITA plot and then inoculated soon after emergence. ACR TZE-MSR-W Population IITA TZE MSR-W-F Population IITA Acquisition access period (AAP). At 48 h, Embu 11 Population Kenya nonviruliferous leafhoppers (about 30-40) were CIRAD390 Inbred line CIRAD given AAP on MSV-infected maize or wild grasses. PANNAR 5195 Hybrid PANNAR Local check Hybrid Kenya In most cases, detached MSV-infected leaves, rather than an intact plant were used as the virus

19 source. The distal portions of leaves were inserted incidence was recorded at all sites, but the highest into a glass cage 2.5 cm in diameter and 20 cm in MSV disease incidence was recorded at Jamaica, height. The distal end of the glass cage was ranging from 88% in the local check to 0% in sealed with fine muslin cloth (secured with a rubber CIRAD 390. But disease incidences of 2% and 6% band) to allow air circulation. MSV-infected maize were recorded in the same genotype (CIRAD 390) leaves (Muguga isolate) were usually secured at at Kagumo (location #4) and Oyugis (location #8), the other end of the cage by a cotton wool plug. respectively. During this season MSV disease The basal parts of MSV-infected leaves were incidences of 1% and 4% were recorded on immersed in a beaker of water to maintain turgidity genotype [MSRxPooL 9]CIF2-205-1(OSU23I) at during AAP. Jamaica and Kagumo (Table 5).

Inoculation access period (IAP). Five viruliferous During the 1998 long rain season, the multilocation leafhoppers were given a 7-day IAP on each lot of experiment was planted at nine locations. There test maize seedlings. Cellulose acetate cages (10 was no MSV incidence in two locations (Table 5). x 30 cm) were used for IAP because they allowed While MSV disease incidence increased at five or more maize seedlings to be inoculated with Muchatha (location #2), it decreased in Jamaica. five leafhoppers. Records were made on the Disease severity in the 18 entries of Jamaica incidence and severity of MSV in each maize (Table 6) is typical of other locations, except that genotype 30 days postinoculation period (PIP). A CIRAD 390 was rated immune at this location with a graded scale of 0-5 was used to assess the score of 0 and incidence of 0%. In at least four severity in each maize genotype: 0 = no streaking, locations, however, CIRAD 390 succumbed to an 1 = very few streaks, 2 = light streaking, 3 = moder- MSV biotype, although MSV symptoms on CIRAD ate streaking, 4 = severe streaking on at least 60% 390 were very mild (a score between 1.5 and 2.0). of leaf area and 5 = severe streaking on 75% of Similarly, in three locations where [MSRxPooL leaf area or more plus stunting (Mesfin et al 1992). 9]CIF2-205-1(OSU23I) succumbed to the MSV biotype, symptoms were mild (scores between 1.5 Results and 2.0). Maize streak disease incidence was very low during the three seasons in which the experiment Glasshouse screening was routinely conducted was carried out. The exception was Jamaica using a Muguga isolate of MSV. In cases where the (location #1), one of four sites in the highlands east test material proved immune, MSV isolates from of the Rift Valley. Table 5 summarizes disease the Coast and Lake Basin regions were used to incidence in each of the 10 locations over the three reconfirm findings. Results of glasshouse screen- seasons. Of the 10 locations planted in the first ing presented in Table 7 appear to confirm the season (1997 long rain) there was streak disease data from multilocation trial site # 1 (Jamaica): the incidence at four sites but very little in four others. At two immune genotypes were not escapes but Mitunguu (location #11), disease incidence was indeed immune to the local MSV biotype. The recorded in only five genotypes. Reasonable MSV immune line from CIMMYT [MSRxPool 9]CIF2-205- disease was recorded in Jamaica; incidence 1(OSU23I) was later screened in the glasshouse ranged from 100% to 0% (Table 5). Although an with MSV isolates from Oyugis, Amagoro, and MSV incidence of 0% was recorded in the maize Mtwapa. The isolates from Oyugis and Mtwapa genotype CIRAD 390 at Jamaica (location #1), caused mild streak symptoms (score 1.5). The disease incidences of 3% and 1% were recorded isolate from Amagoro, however, did not induce in the same genotype at Muchatha (location #2) streak symptoms on this genotype. and Githunguri (location #3), respectively. Other diseases recorded in the multilocation While no MSV incidence was recorded in the experiment included sugarcane mosaic disease, genotype [MSRxPooL 9]CIF2-205-1(OSU23I) at common rust caused by Puccinia sorgi, Turcicum Jamaica (location #1), during the 1997 long rain blight caused by Exserohilum turcicum, common season, disease incidence of 1% was recorded in smut caused by Ustilago maydis, and head smut the same genotype at Mitunguu in the same caused by Sphacelotheca reiliana. Sugarcane season. mosaic virus (SCMV) was the most severe on C390 (CIRAD 390) in all locations. Common rust and During the 1997 short rain season, the multilocation polysora rust (in the coast) were common in most experiment was conducted in five locations. MSV genotypes except in CIMMYT inbred lines CML202

20 Table 5. Disease incidence (%) in 12 locations Entry 1 2 3 4 5 6 7 8 9 10 11 12

1997 long rains VHCY 82.8 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 VHCW 76.0 0.0 0.0 - 0 0 0 0 0.0 - 3.0 0 AO89 98.3 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 AO76 88.4 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 HASR 91.7 0.0 0.0 - 0 0 0 0 1.3 - 0.0 0 CML202 91.3 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 CML197 100.0 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 [MSRxPool 9]CIF2-205-1 (OSU23I) 0.0 0.0 0.0 - 0 0 0 0 0.0 - 1.3 0 Pool 9A BSAF3 6.0 0.0 0.0 - 0 0 0 0 0.0 - 1.3 0 [MSR/Pool 9A]C2F5 48.0 0.0 0.0 - 0 0 2 0 0.0 - 0.0 0 Tzi3 94.0 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 KU144SR/(Tzi35) 100.0 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 ACR TZE-MSR-W 91.1 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 TZE MSR-W-F 93.6 0.0 0.0 - 0 0 0 0 0.0 - 3.0 0 Embu 11 98.0 0.0 0.0 - 0 0 0 0 0.0 - 1.1 0 CIRAD 390 0.0 3.0 1.0 - 0 0 0 0 0.0 - 0.0 0 PANNAR 5195 93.8 0.0 0.0 - 0 0 0 0 0.0 - 0.0 0 Local check 98.5 0.0 0.0 - 0 0 0 0 2.0 - 0.0 0 1997 short rains VHCY 15.3 0.0 - 14.5 - - 1.4 1.3 - - 0.0 - VHCW 19.7 0.0 - 3.6 - - 0.0 9.2 - - 0.0 - AO89 11.5 0.0 - 11.2 - - 0.0 4.6 - - 0.0 - AO76 21.5 0.0 - 5.7 - - 3.1 0.0 - - 0.0 - HASR 31.6 0.0 - 16.6 - - 3.9 3.0 - - 0.0 - CML202 18.8 0.0 - 2.2 - - 1.4 3.1 - - 0.0 - CML197 59.5 0.0 - 1.8 - - 5.0 9.2 - - 0.6 - [MSRxPool 9]CIF2-205-1 (OSU23I) 1.1 0.0 - 3.7 - - 0.0 0.0 - - 0.0 - Pool 9A BSAF3 3.0 0.0 - 0.0 - - 0.0 0.0 - - 0.0 - [MSR/Pool 9A]C2F5 17.2 0.0 - 0.0 - - 9.0 2.7 - - 0.0 - Tzi3 16.7 4.1 - 6.9 - - 0.0 3.3 - - 0.0 - KU144SR/(Tzi35) 27.2 0.0 - 13.3 - - 0.0 2.8 - - 0.0 - ACR TZE-MSR-W 20.2 0.0 - 27.3 - - 0.0 3.0 - - 0.0 - TZE MSR-W-F 31.5 0.0 - 9.2 - - 0.0 7.5 - - 3.0 - Embu 11 85.1 1.2 - 12.6 - - 7.1 10.4 - - 0.0 - CIRAD 390 0.0 3.0 - 2.0 - - 0.0 5.7 - - 0.0 - PANNAR 5195 10.0 0.0 - 6.2 - - 0.0 3.1 - - 0.0 - Local check 88.2 0.0 - 15.2 - - 9.5 11.0 - - 0.6 - 1998 long rains VHCY 7.7 0.9 - 0.0 0 0 0.0 0 3.9 - 0.0 - VHCW 6.1 0.0 - 1.1 0 0 0.0 0 0.0 - 2.3 - AO89 8.1 48 - 0.0 0 0 4.5 0 0.0 - 2.0 - AO76 3.1 2.9 - 0.0 0 0 1.4 0 0.0 - 0.0 - HASR 6.5 0.0 - 3.0 0 0 4.3 0 0.0 - 0.0 - CML202 0.0 1.0 - 3.7 0 0 1.4 0 0.0 - 0.0 - CML197 4.1 0.93 - 16.9 0 0 0.0 0 39.2 - 0.0 - [MSRxPool 9]CIF2-205-1 (OSU23I) 0.0 0.0 - 0.0 0 0 0.0 0 0.0 - 2.5 - Pool 9A BSAF3 0.0 0.0 - 0.0 0 0 0.0 0 0.0 - 2.5 - [MSR/Pool 9A]C2F5 2.8 1.9 - 2.1 0 0 1.5 0 0.0 - 0.0 - Tzi3 8.9 14.7 - 5.9 0 0 8.9 0 0.0 - 0.0 - KU144SR/(Tzi35) 17.0 0.0 - 3.9 0 0 0.0 0 11.8 - 0.0 - ACR TZE-MSR-W 12.6 1.0 - 7.6 0 0 1.3 0 5.1 - 0.0 - TZE MSR-W-F 13.1 1.0 - 0.0 0 0 0.0 0 9.8 - 3.0 - Embu 11 13.2 12.6 - 23.9 0 0 4.0 0 62.5 - 1.1 - CIRAD 390 0.0 0.0 - 2.2 0 0 0.0 0 0.0 - 0.0 - PANNAR 5195 6.6 0.96 - 28.3 0 0 0.0 0 31.5 - 0.0 - Local check 23.3 59.0 - 1.6 0 0 10.1 6.8 28.6 - 0.0 -

21 Table 6. Maize streak disease incidence and severity on 18 maize genotypes planted in Jamaica, 1997 long rains

Entry Source Plants established % MSV Severity (no.) incidence score VHCY S. Africa 58 82.8 3.0 VHCW S. Africa 68 76.0 3.0 AO89 PANNAR 59 98.3 3.0 AO76 PANNAR 43 88.4 2.0 HASR Burundi 60 91.7 1.5 CML202 CIMMYT 46 91.3 1.5 CML197 CIMMYT 35 100.0 2.5 [MSRxPool 9]CIF2-205-1 (OSU23I) CIMMYT 70 0.0 0.0 Pool 9A BSAF3 CIRAD 66 6.0 1.0 [MSR/Pool 9A]C2F5 CIMMYT 75 48.0 2.5 Tzi3 IITA 67 94.0 3.0 KU144SR/(Tzi35) IITA 56 100.0 3.5 ACR TZE-MSR-W IITA 76 91.1 2.0 TZE MSR-W-F IITA 63 93.6 2.0 Embu 11 Kenya 64 98.0 5.0 CIRAD 390 CIRAD 43 0.0 0.0 PANNAR 5195 PANNAR 64 93.8 3.0 H614 Kenya 68 98.5 5.0

Table 7. Glasshouse screening of 18 maize genotypes intended for multilocation trial using the Muguga isolate of MSV

Entry Source Plants established % MSV Severity (no.) incidence score VHCY S. Africa 24 8.3 4.0 VHCW S. Africa 24 42.6 3.0 AO89 PANNAR 21 80.9 3.0 AO76 PANNAR 14 71.4 3.0 HASR Burundi 11 81.8 2.5 CML202 CIMMYT - - - CML197 CIMMYT 17 100.0 3.0 [MSRxPool 9]CIF2-205-1 (OSU23I) CIMMYT 20 0.0 0.0 Pool 9A BSAF3 CIRAD 23 34.8 2.0 [MSR/Pool 9A]C2F5 CIMMYT 20 35.0 2.5 Tzi3 IITA 19 100.0 2.0 KU144SR/(Tzi35) IITA - - - ACR TZE-MSR-W IITA 17 29.4 3.0 TZE MSR-W-F IITA 16 93.8 3.0 Embu 11 Kenya - - - CIRAD 390 CIRAD 60 0.0 0.0 PANNAR 5195 PANNAR - - - H614 Kenya - - -

and [MSRxPool 9]CIF2-205-1(OSU23I). The Discussion and Conclusions PANNAR hybrid was well adapted in most loca- The data indicated that MSV disease was actually tions. Perhaps the most important finding is that all absent in many of the 12 locations. However, the streak-resistant maize germplasm from all five conclusions can be drawn from Jamaica (location sources were indeed resistant to MSV inKenya, but #1) and two others. Both maize genotypes CIRAD their level of resistance varied from one genotype 390 and [MSRxPOOL 9]CIF2-205-1(OSU23I) to the other. demonstrated high levels of MSV resistance. They

22 were immune to MSV in most locations. This Mesfin, T., N.A. Bosque Perez, I.W. Buddenhagen, agrees with data from glasshouse screenings G. Thottapilly, and S.O. Olojede. 1992. using a Muguga isolate of MSV. Since these two Studies of maize streak virus isolates from genotypes succumbed to MSV in some locations grass and cereal hosts in Nigeria. Plant but not in others, it can be concluded that at least Dis. 76:789-795. two biotypes of MSV occur in Kenya, one more Mullineaux, P.M., J. Donson, B.A.M. Morris- virulent than the other. This observation is similar to Krisimich, I.M. Boulton, and J.W. Davids. that of Bock (1975) and Njuguna (1996). The 1984. The nucleotide sequence of MSV glasshouse screening appears to agree with the DNA. EMBO J. 3:3063-3068. field data, suggesting that MSV isolates from some Ngwira, P. 1988. Serological differentiation of MSV locations were more virulent than others. For Isolates and epitope characterization of example, both Amagoro and Oyugis in the lake the MSV capsid protein. MS thesis. The ecology had MSV isolates of differing virulence. Ohio State University. 79 p. But contrary to earlier reports by these two authors, Njuguna, J.G.M. 1996. Epidemiology of maize the different MSV biotypes are not confined to streak disease in Kenya. Ph.D. thesis. The specific regional ecologies but appear throughout Ohio State University. 152 p. Kenya. 3.3 Breeding and Biotechnology Both CIRAD and CIMMYT germplasm can be made available to the national maize breeding 3.3.1 Overview program at no cost to MSV resistance breeding. This will be our best source of resistant This session’s presentations also attracted much germplasm. Finally, with regard to its MSV resis- discussion. Maize breeder Dr. J. Ochieng spoke tance and adaptability, PANNAR hybrid 5195 on “Breeding for MSV resistance as a priority in appeared to perform very well in most locations. Kenya.” He argued that reducing losses in maize For this reason, Kenyan farmers should be encour- productivity through MSV-resistant/tolerant variet- aged to grow this variety to avoid heavy losses ies should be the farmers’ first line of defense. He associated with the MSV epidemic. reported that the Kenyan team has developed a strategy that harnesses multidisciplinary synergies References to produce multiple disease-resistant varieties of Book, K.R, E.J. Guthrie, G. Meredith, and T. maize. This strategy relies on such biotechnolo- Ambetsa. 1975. Maize streak virus. gies as marker-assisted selection (MAS), Records of research for 1974. East African agroinfection, and maize genome transformation. Agricultural and Forestry Research Organization. p 118. Spanning a decade, their long-term plan is to use Briddon, R.W., P. Lunness, L.C. Chamberlin, and genome mapping and transformation techniques P.G. Markham. 1994. Analysis of the to produce MSV-resistant populations for future genetic variability of maize streak virus. inbred lines. Their medium plan (5-10 years) uses Virus Genes 9(1):93-100. agroinoculation to produce MSV-resistant open- Damsteegt, V.D. 1983. Maize streak virus: 1 Host pollinated varieties (both synthetics and compos- range and vulnerability of maize ites). In the short term, the project will use MAS on germplasm. Plant Dis. 67:734-737. MSV-resistant VC, DC, TWC, TC, and SC hybrids. Efron, Y., S.K. Kim, J.M. Fajemisin, J.H. Mareck, The team is awaiting the results of mapping C.Y. Tang, Z.T. Dabrowski, H.W. Rossel, studies to implement MAS and other supportive G. Thottappilly, and L. Buddenhagen. techniques. 1989. Breeding for resistance to maize streak virus A multidisciplinary team Dr. Ochieng listed several questions for which approach. Plant Breed. 103:1-36. scientific answers are urgently needed. Why has Howell, S.H. 1984. Physical structure and genetic MSV infection escalated over the years? Are more organization of the genome of MSV virulent types linked to mutation, recombination, or (Kenyan isolate). Nucleic Acid Res. migration? Is the emergence of more virulent types 12:7359-7375. linked to the increase in the number of susceptible Luwarowitz., S.G. 1988. Infectivity and complete maize varieties being grown? Dr. Ochieng con- nucleotide sequence of the genome of cluded by praising the efforts of his research team South African Isolate of MSV. Nucleic Acid and by reminding the audience of how desperately Res. 16:229-249.

23 Africa’s farmers need MSV-resistant maize variet- Resistance against MSV has been identified in ies (for Dr. Ochieng’s complete study results, see various sources. In the early 70s, IRAT and KARI section 3.3.3). researchers discovered that the cultivar “Revolu- tion” from Reunion Island possessed MSV resis- Workshop participants also eagerly listened about tance. Today, however, breeding programs have the work of Dr. Jane Ininda and her team on “DNA access to and use different sources of resistance. polymorphism among sources of MSV resistance At CIMMYT’s Harare station, for example, resis- and disease expression in their F2/F3-derived tance has been developed from a population ZSR lines.” Several participants went on a field trip to 923 BULK, while the West African-based IITA relies see her work, especially at the Muguga Research on TZ-Y. France’s CIRAD breeders use the CVR- Center. Dr. Ininda explained that eight varieties of C3 population that goes back to IRAT 297, and in MSV-resistant maize were obtained. CIMMYT South Africa an unpublished source of resistance provided CML202, CML197, and OSU23I. IITA has been used to develop various MSV-resistant provided Tzi3 and Tzi35; PANNAR Seeds pro- hybrids. vided A076. CIRAD supplied C390 and the Grain Crops Institute supplied VHCY. Our work takes advantage of new technologies to begin to map and characterize different sources of Her results showed that the highest percentage of resistance. Various parental sources of resistance highly resistant lines was in the F3 lines that origi- have been acquired, and F3 populations have nated from CIMMYT’s OSU23I. At least 46% of been derived from them using molecular-mediated OSU23I had high resistance to MSV vs. 26% for mapping. CIRAD’s C390. Only 2% of IITA’s Tzi3 lines was found to be highly resistant. This dropped to 0.01% Materials and Methods for CML202 lines from CIMMYT. Eight sources of MSV resistance were supplied by various institutions. CIMMYT provided CML202, Other highlights of Dr. Ininda’s work included the OSU23I, and CML197. IITA supplied Tzi3, and findings that 69% of OSU23I lines was resistant vs. Tzi35; CIRAD supplied C390. The Grain Crops 62% of C390 lines. In addition, 42% of CML02 and Institute provided VHCY and PANNAR Seeds 40% of Tzi3 lines were grouped as MSV resistant. donated AO76. These lines were chosen because Finally, trials at Alupe in the Lake Victoria region they were traced back to different sources of MSV showed that 38% of OSU23I lines, 28% of C390 resistance. In addition, Kenya provided the sus- lines, 14% of CML202 lines, and 4% of Tzi3 lines ceptible check EM11-133 (Table 8). were highly resistant (see section 3.3.2 for Dr. Ininda’s compete study results). In September 1997, initial crosses were made between EM11-133 and each of the sources of MSV resistance. In May 1998, parents were grown 3.3.2 DNA Polymorphisms among Sources of in glasshouses at the JIC for DNA extraction. MSV Resistance and Disease Expression

in F2 and F3-derived Lines (J. Ininda, DNA extraction and RFLP. Fragments produced J. Snape, and R.O. Opondo) after DNA digestion with restriction enzymes in any one locus vary in size or length for different indi- Introduction viduals. These differences are designated RFLP. The basic objectives of this study are (1) to In this experiment, equal quantities of leaf tissue evaluate restriction fragment length polymorphism from each inbred line were harvested and then (RFLP) and microsatellite polymorphism among lyophilized at –80 oC. Plant genomic DNA was sources of MSV and (2) to evaluate the expression extracted using the CTAB technique. All parental lines were analyzed for 36 RFLP probe-enzyme of MSV disease symptoms among F3 lines of different parental sources. It is hoped that identify- combinations using three restriction enzymes ing MSV-resistant maize lines will assist Kenya’s (EcoR1, EcoRV, and HindIII) and 64 pairs of breeding efforts to fight MSV. This is a top priority microsatellite primers (see Tables 9 and 10). RFLP program because the country’s most popular probes were obtained from the University of maize hybrid lacks resistance to MSV. Missouri, Columbia, and were initially chosen to map four chromosomes of the maize genome. Separate digests were performed with EcoR1, EcoRV, and HindIII and then separated on 0.7%

24 Table 8. Stage of inbreeding and agronomic descriptions of 10 maize lines used to assess RFLP and microsatellite polymorphism

Source Line Generation MSVa Rusta Maturity Grain color

CIMMYT CML202 S 4 T T Late White

CML197 S 4 T R Late White

OSU23I S 4 I T Mid White

IITA Tzi3 S 2 T R Mid Yellow

Tzi35 S 2 T R Late White PANNAR A076 -b T T Mid White GRAIN VHCY - T R Early Yellow CIRAD C390 - I R Early Yellow

Kenya EM11-133 S 4 S R Late White

Kenya E12-210 S 4 S R Mid White a T = tolerant, S = susceptible, I = immune, and R = resistant. b Inbreeding level not available.

Table 9. RFLP probes used in assessing polymorphism

Probe (locus name) Source type Insert size(bp) Polymorphism bn18.01 UMCa Genomic 2400 abcb Csu3 UMC CDNA 1200 abc Csu21 UMC CDNA 600 abc Npi262 UMC Genomic 490 abc Npi268 UMC Genomic 670 abc Npi297 UMC Genomic 1000 abc Npi304 UMC Genomic 1000 abc Umc10 UMC Genomic 1100 abc Umc11 UMC Genomic 890 abc Umc29 UMC Genomic 990 abc Umc105 UMC Genomic 610 abc Umc167 UMC Genomic 620 abc

aUMC, University of Missouri, Columbia. babc means polymorphism was observed between EM11-133 and CML202, Tzi3, and OSU23I for at least one enzyme.

Agarose gel. Southern blotting, filter hybridization least 100 pairs of SSR reverse and forward with 32P-labeled probes, and autoradiographing primers were obtained from Research Genetics, were described by the Cereals Department at JIC. Inc. in Huntsville, Alabama (Table 10). The SSR Data were collected for all the probe-enzyme procedure was a 20 ul PCR reaction mix, contain- combinations and for those bands that both ing 20 ng of genomic DNS, 0.1 uM of each primer, yielded a clear hybridization signal and showed a 1x reaction buffer, 1 u Boe Taq DNA polymerase, clear detection of a band. RFLP bands in the and 2 mM each of dGTP, dTTP, dATP, and dCTP. autoradiographs were scored visually on a scale Samples were covered with 1 drop of mineral oil of 1 or 0, where 1 = presence of an allele and 0 = and subjected to 30 thermal cycles. They were absence of an allele. then denatured (1 min, 94 oC), annealed (1 min, 56 oC), and extended (1 min, 72 oC), which was fol- Microsatellite (SSR) analysis. Microsatellites lowed by a final extension step (4 min, 72 oC) using are short DNA repeats of two or three base pairs. a DNA thermal cycler. PCR products were dena- The alleles in the different genotypes are amplified tured for 5 min at 95 oC and then separated on 6% by PCR and differentiated by the size of the repeat. denaturing acrylamide gel with 8M urea at 80W of The size of the microsatellite is a function of copy constant power for 1 h and 15 min. The silver number of the basic repeat. A search was con- stunning protocol was used and autoradiographs ducted for maize microsatellite markers, and at were exposed and scored.

25 Degree of polymorphism and levels of het- Table 10. Microsatellite primers used in analyzing erozygosity. All probes and microsatellite parental polymorphism, their expected primers were initially screened for all 10 parents to sizes, and annealing temperatures detect the degree of polymorphism between Locus Expected Anneal Polymorphisma sources of MSV resistance and susceptible size cbp) (oC) parents. MAC.E00BO3 148 56 ab MAC.T03B03 158 56 x Mapping populations and assessment of MAC.E01CO8 123 56 x disease severity. Of the eight resistant parents, MAC.E01F06 131 56 ab MAC.T02B10 107 56 a- four were chosen to develop mapping populations. MAC.T02H12 138 56 - By selfing individual F2 plants, 104 F2:F3 lines were MAC.1A01 110 56 xb obtained from the cross EM11-133/OSU23I and MAC.E01C01 95 56 ab 156 lines from EM11-133/Tzi3. All the F3 lines were MAC.E0G01 137 56 ab planted in three locations in Kenya. Artificial MAC.T02B08 112 56 ab infestations with viruliferous leafhoppers carrying MAC.T02E01 147 56 ab MAG.1A03 104 56 ab the Coast virus isolate were done at Mtwapa, MAC.E01A03 167 - 56 ab Alupe, and Muguga. Bngl107 - 58 - Bngl108 - 58 ax MSV severity was rated on 10 individual plants per Bngl119 - 58 xx plot using a scale of 1-5, where 1 = no symptoms Bngl127 - 61 - Bngl131 - 61 - on leaves, 2 = light streaking on young and old Bngl143 - 61 - leaves, 3 = moderate streaks on leaves, 3-4 = Bngl149 - 61 - moderate streaks and some stunting, 4 = severe Bngl153 - 61 - streaking on 60% of leaf area, 4-5 = severe streak- Bngl161 - 55 ab ing of 75% of leaf area, 5 = severe streaking on Bngl166 - 55 ab >75% of leaf area. Bngl176 - 55 xx Bngl182 - 55 ab Bngl236 - 55 - Results and Discussion Bngl238 - 55 ab Bngl240 - 55 ab RFLP and Microsatellite Polymorphisms Bngl279 - 55 xx among Sources of Maize Streak Virus Bngl603 - 55 - Resistance Bngl609 - 58 xx Bngl653 - 58 xx Bngl107 - 58 ax Genetic variation for RFLPs among parents. Bngl391 - 58 - An autoradiograph was obtained after analyzing Bngl426 - 56 ab the parents for polymorphism with three radioactive Bngl434 - 58 ab labeled probes. Polymorphism was observed for Bngl572 - 58 - Bngl657 - 58 xx at least one probe/enzyme combination for any Bngl119 - 58 - two parents. Some probe-enzyme combinations bngl669 - 58 a- showed more RFLP variants than others. For bngl127 - 58 ax example, umc167 showed only 4 alleles while bngl1292A - 58 - umc10 showed polymorphism between EM11-133 bngl430 - 58 - and CML202, Tzi3, C390, and OSU23I (Fig. 1). This bngl469A - 58 a- bngl619 - 58 - suggests that the use of EcoR1 may expose bngl153 - 58 - greater polymorphisms among parents. bngl236 - 58 - bngl640 - 58 a- Genetic variation for SSRs among bngl420 - 58 - the parents. Genetic variation for microsatellites bngl667 - 58 ab (Figs. 2 and 3) shows two extreme patterns ob- bngl389 - 58 ab bngl565 - 58 ab tained for two different loci amplified. A total of 54 bngl572 - 58 - microsatellite primers were assayed, and 35 gave bngl589 - 58 xx detectable PCR (polymerase chain reaction) a a = polymorphism was observed between CML202 and EM11- products and/or polymorphisms (Table 10). 133, b = polymorphism was observed between Tzi3 and EM11- Unfortunately, difficulties in establishing the right 133, x = not polymorphic, - = did not amplify. PCR conditions for DNA amplifications in our genotypes limited work progress. But our study

26 Fig. 1. Pattern of RFLP bands obtained from 10 parents by labeling probes with radioactive 32P

1 = A076, 2 = C390, 3 = CML197, 4 = Tzi3, 5 = CML202, 6 = EM11, 7 = EM12, 8 = OSU23I, 9 = VHCY, and 10 = Tzi35 a: Nearly monomorphic bands of umcl167 for DNA digested with Hindlll, b: Polymorphic bands of umc67 for digests of EcoR1, EcoRV, and Hindlll, c: Polymorphic bands of umc10 for DNA digested with Hindlll

27 Fig. 2. Pattern of microsatellites showing polymor- phism among 10 parents by using silver staining Fig. 3. Pattern of microsatellite locus showing nearly monomorphic bands in 5 lines 1 = A076, 2 = C390, 3 = CML 197, 4 = Tzi3, 5 = CML202, 6 = EM11-133, 7 = EM12-210, 2 = C390, 3 = CML197, 4 = Tzi3, 5 = CML202, 8 = OSU23I, 9 = VHCY, and 10 = Tzi35 and 7 = EM12

also showed that there is adequate polymorphism for both RFLP and microsatellite markers between Table 11. Percentage of lines in each disease at least two sources of MSV resistance (CIMMYT category for each mapping population and IITA) and the susceptible Kenyan inbred lines MSV EM11- EM11- EM11- EM11- of EM11-133 to enable screening of the mapping Score/ 133/ 133/ 133/ 133/ population for DNA markers. Class C390 OSU23I CML202 Tzi3 Muguga Expression of Disease Symptoms among

F3 Lines of Different Parental Sources 1.0-2.0 26.3 46.2 0.01 2.2 of MSV Resistance 2.1-3.0 35.3 23.1 42.0 38.0 3.1-4.0 26.3 14.4 46.4 55.5 Results indicate that the probability of obtaining 4.1-5.0 12.1 16.3 11.6 4.3 highly resistant MSV lines is higher when the Alupe source of resistance used for conversion pos- a sesses immunity to MSV (Table 11). The highest HR 28.2 37.3 13.7 4.4 R 24.3 32.5 73.9 42.2 percentage of highly resistant lines was found in F 3 S 47.5 29.2 11.8 53.3 lines originating from OSU23I. In this population, 46% of the lines had high resistance to MSV at aHR = highly resistant, R = resistant, and S = susceptible. Muguga, while lines possessing C390 as a source of resistance had 26% highly resistant lines. In addition, 69% of the lines from OSU23I, 62% from C390, and 42% from CML202 were classified as

28 resistant. CML202 and Tzi3 with high resistance to fore? Or was it a case of more MSD-susceptible MSV were fewer: 0.01% and 2.2%, respectively. But maize varieties being grown by farmers? If it is a a greater percentage of their lines fell in the middle case of new MSV strains, are these pathotypes the category. Lines with the OSU23I and CML202 result of (gene) mutation, migration, or genetic sources were found to be adapted and resistant to recombination? It is hoped that research scientists other common diseases, particularly leaf rust, gray elsewhere have unraveled these intricacies. If not, leaf spot, and turcicum blight at Muguga (data not then Kenyan breeders will have to figure out the shown). This suggests that OSU23I or CML202 causes of the haphazard reactions of maize have stronger resistance to MSV. genotypes (varieties) to MSD over time and space. We urgently need this information to plan The same pattern of segregation was observed at more viable breeding programs that will develop Alupe using the Lake region isolate. A larger useful products (cultivars) for our target farmers. percentage of lines from OSU23I and C390 were in the highly resistant class compared with the other Breeding strategies. Past research efforts to two sources (Table 11). However, 74% of the lines develop MSV-resistant maize varieties in Kenya derived from the MSV source CML202 fell in the were based on crossing and backcrossing using resistant or tolerant category. Furthermore, 42% of “Hickory King” and “Reunion Revolution” varieties the lines from the Tzi3 source also fell in this as putative sources of resistance. But there were category. In both Muguga and Alupe, the largest little tangible benefits from this early research effort. number of lines that showed susceptibility to MSV Recent reports show an increasing incidence of were from Tzi3: 56% of the lines were susceptible MSVD in Kenya, with infection rates as high as 70- to MSV at Muguga and 53% were susceptible at 80% in the Central highlands (Theuri and Njuguna Alupe. 1988).

3.3.3 Breeding for MSV Resistance as a A conventional breeding solution is envisaged in Priority in Kenya (J.A.W. Ochieng) the Coordinated Ecosystem Breeding (CEB) strategy launched by the maize breeding team of A survey conducted on maize farms in 34 districts KARI in 1999 (Table 13). The breeding plan is in Kenya by Ray Louie in 1978 revealed low and based on backcrossing, followed by selection. It sporadic incidence of maize streak virus disease uses reputed exotic sources of MSV resistance in (MSVD). A repeat survey 10 years later by Theuri, crosses with adapted breeding stocks and however, indicated a remarkable (4-80-fold) capitalizes on genetic complementarity. The CED increase in MSD incidence in five districts of strategy is designed to harness multidisciplinary Kenya’s Central Province (Table 12). Similarly, a synergy to develop multiple disease-resistant 1998 survey of 10 farms in southwest Kenya varieties of maize. showed a high incidence of 80-100%. Even biotechnological approaches, such as using Some pertinent questions to ask about these marker-assisted selection, agroinfection via changes are: Why has there been an escalation in Agrobacterium tumefaciens, and maize genome MSD incidence over the years? Was it due to new transformation are seen as candidate methodolo- and more virulent MSV pathotypes coming to the gies. Supporting research thrusts include MSD epidemiology, characterization of MSV isolates from various parts of Kenya, and studies on insect feeding behavior. Susceptibility reactions among Table 12. Distribution of maize streak virus (MSV) disease (% incidence), Kenya in 1988 maize genotypes will aid in crafting viable breed- ing plans. The overall goal is to incorporate resistance to MSV and other economically impor- Province District 1977 1978 1988 tant disease pathogens into maize varieties with Site desirable agronomic traits. 12 Central Nyandarua 0.0 0.0 - - So far, the MSV Project team has assembled both Nyeri 11.6 8.2 36 1 a large array of very good sources of MSV resis- Kirinyaga 1.0 7.3 80 0 tance and genetic stocks (accessions) with good Muranga - 1.3 82 5 Kiambu 1.6 9.6 87 13 agronomic traits. These are now ready to be filtered into the mainstream breeding program. In Source: J.M. Theuri (1988).

29 Table 13. Genetic breeding strategy for developing maize streak virus disease varieties

Short-term<5 yr Medium-term: 5-10 yr Long-term: >10 yr Germplasm available Inbred lines: Breeding populations Populations/Landraces (Local and exotic) Methodology envisaged Recycling (Conversion):- Backcross-cum-recurrent Recurrent selection schemes · Crossing /Backcrossing selection · Pedigree selection Formation of synthetics and Formation of composites composites · Sister line development Commercial (products) (Marker-assisted selection) (Agroinoculation) (Genome mapping and transformation) MSV-resistant VC, DC, TWC, MSV-resistant OPVs: MSV-resistant populations as TC, and SC hybrids (synthetics, composites) source of future inbred lines

the meantime, results of mapping studies are Current Commercial Maize Varieties and their being obtained so that MAS and other supportive Survival with MSV techniques can be applied. Crosses have been There are no MSV-resistant maize varieties in the made between the exotic MSV resistance sources highlands of Kenya. This includes the four varieties and adapted local breeding stocks. In doing this, H614, H625, H626, and H627 that were developed caution has been taken to “respect” heterotic for the highland region (over 1,700 m asl), where patterns. Whatever breeding plan is put into action, MSV was formerly rare. The two hybrids—H622 and breeders should not lose sight of the need to sort H623—which are suited for the medium- to high- out heterotic patterns and develop hybrids (for the altitude zones (1,200-1,700 m), are also not resis- formal seed sector) and open-pollinated varieties tant to MSV. (for the informal seed sector). Two hybrids, H511 and H513, are common in the Reference mid-altitude zone. H511 is susceptible to MSV, Njuguna, J.G.M., C.J. Kedera, L. Muriithi, S. Songa, although H513 is fairly tolerant. Furthermore, DH01, and B. Odhiambo. 1990. Overview of DH02, and the newly recommended release— maize disease in Kenya. In: Proceedings DH194 (DH03)—are partially resistant to MSV. In the of a Workshop on Review of the National lowlands there are CCM, PHI, and PH4 maize Maize Research Program. KARI/ISNAR varieties. PHI and PH4 are fairly resistant to MSV, Management Training Linkage Project. 19- but CCM is susceptible. 23 November 1990. Kakamega, Kenya. Managing Maize Streak Virus 3.4 Private Sector–Maize Seeds and MSV The private sector has developed long- and short- term strategies for controlling MSV. The long-term goal is to develop MSV-resistant varieties, which 3.4.1 Kenya Seed Company and MSV can be achieved by breeding for multiple disease (F. Ndambuki) resistance. In the short term, the goal is to integrate cultural with chemical practices to decrease MSV Introduction losses. MSV infection is very high in Kenya’s Central Province, particularly in the Kiambu district. MSV An off-season nursery in Sigor, West Pokot, where occurs in Nyanza, the Coast region, and isolated the temperature oscillates around 28 oC, provides parts of the Eastern Province. In addition to MSV, an excellent site for MSV screening. To increase other diseases such as lowland blight (H. maydis), the number of leafhoppers, either finger millet or lowland rust (P. polysora), ear rot, striga, and maize bulrush millet is planted around the maize plot. stalk borer are rampant in the coastal lowlands Field data collection, particularly for MSV, was and around Lake Victoria. prevented during variety testing by the erratic

30 occurrence of the disease, which greatly depends • Intercropping, which makes it more difficult for on the population dynamics of the vector male and female leafhoppers to mate. Cicadulina spp. In some years, damage is insignifi- • Chemical control. The Company has decided to cant; in others, epidemics wipe out farmers’ crops. seed-dress certified seed maize sold in MSV- prone areas and maize seed produced in MSV- Collaboration prone sites with Gaucho. Working with KARI provides the Kenya Seed Company with both useful information and 3.4.2 Seed Production and MSV germplasm. We use KARI’s Muguga facilities to in Zimbabwe screen varieties in National Performance Trials Seed production is mostly conducted in high (NPT). By screening maize lines from CIMMYT and rainfall areas with optimum environments for both other sources, we are able to use some of them summer and irrigated winter crops. Seed growers into our breeding program. Advanced trials of very are farmers with irrigation facilities and good maize promising hybrids will be entered into NPTs next production skills. To maximize their annual in- year. comes, most of the seed growers also grow irrigated winter wheat. During the dry winter, Sources of Resistance irrigated wheat is usually the only green crop, and Efforts to introduce MSV resistance into commer- so it harbors high populations of insect pests such cial varieties have not been completely success- as the MSV vector, Cicadulina mbila. Furthermore, ful, but there are hopes that some varieties may the winter wheat crop usually dries up just as soon prove useful. These include IB32, which irrigated, early planted maize crops germinate. belongs to CIMMYT’s population lines. It is from Since winter is a dry season in Zimbabwe, these these CIMMYT populations (TZMSR, 8549SR, early-planted maize crops are the only green 9725SR, 8730SR, etc.) that we extracted inbred plants in the field once the wheat crops have been lines to combine with local germplasm. Other removed. Naturally, the MSV vector and other sources include HASR from Burundi, which unfortu- insects migrate to early planted maize crops. nately is not very stable, and CML lines, which are proving useful for resistance. Because of this scenario, MSV infection is com- mon and widespread in Zimbabwe. About 10% of The work has faced several constraints, such as: the early planted crops are seed maize crops, and • Difficulties in obtaining uniform field infestation, yield losses from MSV infection are reported to be which results in unreliable field screening. as high as 43% (Mzira 1984). The MSV vector • Impossibility of detecting field escapes. feeds on the sap of young maize plants, and the • Lack of cheap, uniform screening methods. critical period for infection is the first 8 weeks from • Glasshouse and leafhopper rearing facilities are planting. Generally, the earlier the stage of infection expensive. A virus infection method that does not the higher the yield losses. In addition to the winter require the vector would be helpful. wheat crops, other conditions favorable to the • Lack of suitable MSV-resistant germplasm/lines development of MSV infection include: close with stable resistance to other diseases. rotation of graminaceous crops, poor control of • Available sources for resistance have grain feed and susceptible volunteer plants, a late yield potentials that are too low to be useful, preceding rainy season, and neighboring irrigated particularly in the highland region, which has a pastures or graminaceous crops. 10-15 t/ha yield potential. Control Measures Cultural MSV Crop Management Control options include seed dressings with The short-term approach combines various insecticides and avoiding the vector by not recommended aspects of crop management: planting seed crops adjacent to winter cereal crops, not planting too early or too late, and/or • Early planting in all areas to avoid vector buildup clearing a 10-m-wide perimeter around the crop. later in the season. Removing the source of infection requires roguing • Crop rotations with a nonhost crop where and out infected plants, removing weeds and alterna- when possible to reduce disease incidence. tive hosts, and eradicating volunteer plants. • Avoiding relay planting of maize. Resistant parentals are most desirable. But • Roguing out diseased plants, which is important farmers’ only MSV-resistant options are C5051 and because the major source of the virus is infected products from CIMMYT Harare, for direct use in maize. crosses and backcrosses to elite lines.

31 4. Achievements, Results, and Recommendations

Research results and achievements during the first B. Odhiambo. This work continues in phase II. phase of the MSV project are divided into four • MSV isolates of different biological and molecu- areas: lar properties were detected, which means that mixed MSV populations do exist. This is impor- 4.1 Studies on MSV-Resistant Maize Lines tant information for resistance breeding. • Results show that the highest percentage of • Zones or regions prone to MSV outside of the peak periods of MSV disease infestation were highly resistant lines is in F3 lines originating from OSU23I provided by CIMMYT. At least 46% of identified. OSU23I had high resistance to MSV. • Preliminary evidence for virulence variability in • At least 26% of CIRAD’s C390 had high resis- MSV populations within zones and maize crops tance to MSV. Only 2% of Tzi3 lines from IITA was established. was found to be highly resistant. • 69% of OSU23I lines, 62% of C390 lines, 42% of 4.4 Collaborative Achievements and CML02 lines, and 40% of Tzi3 lines are resistant. Benefits • Trials at Alupe in the Lake Victoria region • African scientists received advanced training showed that 38% of OSU23I lines, 28% of C390 and learned sophisticated biotechnology skills lines, 14% of CML202, and 4% of Tzi3 are highly in DNA diagnostics and agroinoculation, such as resistant. the development of appropriate molecular • Considerable information was gathered from markers for resistance breeding. field nursery materials and bulking seeds about • Member African countries suffering similar MSV agronomic traits of interest. This was shared problems were provided information and during the workshop. research results. In addition, institutions that • Jane Ininda studied molecular markers at the offered MSV-resistant germplasm, such as IITA, John Innes Centre, and reports of her consider- CIMMYT, CIRAD, PANNAR, and others, gained able, well-documented achievements were from the information and data generated by the made available. project. • Technology transfer and training improved 4.2 Studies on the MSV Vector Kenya’s national capacity to breed and adapt • Aerial densities of trapped Cicadulina indicate MSV-resistant maize. two annual population peaks: July-August and • The project’s technical advisors to Kenya, Dr. November-December. Peter Markham and Prof. Ed Rybicki with their • C. chinai was found to be the most efficient in assistants, Dr. Jina Banks (JIC) and Martin MSV transmission (70%), followed by C. mbila Darren (UCT), visited as planned and provided (60%) and C. storeyi (40%). But species in project guidance. different ecologies differed in their transmission • Closer links were created between specialists— rates. plant breeders, pathologists, and entomologists— • The progenies of Cicadulina breeding in maize who have been used to working in isolation even crops are not important sources for disease at national research institutes like KARI. spread to maize planted later. • As part of the project’s training component, Dr. • An ICIPE/KARI team began epidemiological Jane Ininda and Dr. Benjamin Odhiambo went to work that provides valuable information, espe- JIC, and Dr. Jackson Njuguna and Benjamin cially about the rearing of more aggressive Odhiambo went to UCT. A joint KARI/ICIPE vector species to challenge resistance in the workshop on new molecular technologies for the field. project was also held at ICIPE. • KARI received much-needed information through 4.3 Studies on MSV regional networking with other countries involved in similar or related kinds of research. This • MSV isolates in Kenya and the relevant gene increased KARI’s opportunities for collaborating constructs were analyzed at UCT and transferred with others in the public and private sectors at to JIC for agroinfection studies. The system is regional and global levels. used for preliminary screening to challenge MSV • Regional maize breeding efforts were refocused resistance from inbred lines. Agroinoculation/ from improving yields to disease and pest infection studies were started and optimized at resistance. JIC for future work at KARI by both Dr. Ininda and

32 • Experts from JIC and UCT continue to visit MSV- pathology of MSV and of the genetic basis of infected maize fields in Kenya, which reflects their resistance. Phase II is essentially a more focused commitment to the project’s success and allows effort to determine the location of MSV resistance for continued technology and information ex- genes. According to KARI’s Dr. Jane Ininda, change. mapping to determine MSV resistant loci in five maize varieties ended in August. The workshop was part of an effort to ensure that Africa reaps the benefits of new advances in 4.5 Recommendations agricultural technology. It focused on maize The participants agreed on the need for a stan- because effectively controlling or eliminating the dard MSV Sample Data Sheet. The Sheet should various factors that reduce maize yields in Africa include the following information: date, location, will help contribute to greater self-sufficiency and ecozone, streak incidence, genotypes, stage of stability in the region. infection, streak rating, and the area’s infection history. Other notes could indicate comments The MSV project will improve the household food about grass samples (infected or harboring MSV). security and socioeconomic status of resource- poor farmers, especially women, who produce Participants also agreed that the exchange of 75% of food crops. The project will make its germplasm should include appropriate and basic impact, moreover, through sustainable agricultural information, such as inbred lines, population, S2 practices. Institutions from the North and South— lines, etc. Seed companies should also be including the private sector—are working together to appropriately involved in product development combat MSV and to stabilize maize yields and and promotion by assisting with various aspects of prices. Through this collaboration, KARI has intellectual property rights or licensing agreements. already developed a multidisciplinary team of researchers who have access to some of the latest Finally, researchers emphasized the need to focus biotechnology techniques: agroinoculation, more on farmers and their needs, as well as on the molecular markers, DNA polymorphism analysis, constraints or limitations of their agricultural prac- and other related skills. This technology and tices. expertise is providing a better understanding of the

MSV-resistant germplasm nursery; sources include IITA, CIMMYT, CIRAD, PANNAR- South Africa, the Grain Crops Institute, and KARI.

33 Participants

1. Chumo J.J. 10. Darrin Patrick Martin Kenya Seed Company University of Cape Town P.O. Box 553 Kitale Cape Town Private Bag 7000 Tel. 254-0325-20941-6 Tel. 02 6855413 E-mail: [email protected] 2. Florence Wambugu ISAAA AfriCenter 11. Abdulrazak Yusuf P.O. Box 25171, Nairobi Alemoya University Tel. 254-2-632054 P.O. Box 138 Dire Wara E-mail: [email protected] Tel. 261-05-112374

3. Susan Kimani Njogu 12. Alpha Oumar Diallo ICIPE CIMMYT, Kenya P.O. Box 30772 Nairobi P.O. Box 25171 Tel. 254-2-802501/3/9 Nairobi E-mail: [email protected] Phone 254-2-522878 E-mail: [email protected] 4. Wesley S. Chivatsi KARI Mtwapa 13. James Theuri P.O. Box 16 Mtwapa, Kenya KARI Muguga Tel. 254-011-485839 P.O. Box 30148 Nairobi 5. J.A.W. Ochieng KARI Headquarters 14. Peter G. Markham P.O. Box 57811 John Innes Centre Nairobi Corney Norwich Tel. 254-2-583301-20 NR4 UH UK Tel. 44(0) 1603-452571 6. Francis Ndambuki E-mail: [email protected] Kenya Seed Company P.O. Box 553, Kitale 15. Vincent A.O. Okoth Tel. 254-0325 20941 Development Consultant Institution (DCI) P.O. Box 1108 Kampala 7. Lewis Machida Tel. 041-344815 Monsanto/Cargill E-mail: [email protected] P.O. Box 5398 Harare 16. Sam Oyewole Ajala Tel. 263-4336628/55 IITA Nigeria PMB 5320, Ibadan 8. Dickson Ligeyo Tel. 234 2 2412626 KARI, Kitale E-mail: [email protected] P.O. Box 450 Kitale 17. Otula Owuor Tel. 254-0325-31818 African Sciences P.O. Box 76336 9. Patricia Ngwira Nairobi Department of Agricultural Research Tel. 254-2-577935/561635 P.O. Box 158 E-mail: [email protected] Lilongwe Tel. 265-767222/225 18. Duncan T. Kirubi KARI Embu P.O. Box 27 Embu Tel. 254-0161-20116/20873

34 19. Arnold G.O. Okech 27. Michael Njuguna KARI Headquarters ISAAA P.O. Box 57811 P.O. Box 25171 Nairobi Nairobi Tel. (0161) 20116/20873 Tel. 254-2-632054 E-mail: [email protected] 20. Johnson Thaiya Monsanto Kenya Limited 28. Benjamin Odhiambo P.O. Box 47686 KARI Kabete Nairobi P.O. Box 14733 Tel. 254-2-719567-74 Nairobi Tel. 254-2-440113 21. Kiarie Njoroge E-mail: [email protected] KARI Katumani or [email protected] P.O. Box 340 Machakos 29. Macharia Gethi KARI Embu 22. Joseph D. Devries P.O. Box 27, Embu The Rockefeller Foundation Tel. 254-0161-20116 P.O. Box 47543 E-mail: [email protected] Nairobi Tel. 254-2-228061 30. John S. Wafula E-mail: [email protected] KARI Headquarters P.O. Box 57811 23. Jackson G.M. Njuguna Nairobi KARI Muguga Tel. 254-2-583343 P.O. Box 30148 E-mail: [email protected] Nairobi Tel. 254-0154-32885 31. Kahiu Ngugi KARI Katumani 24. Samwel Njongoro Njihia P.O. Box 340 KARI Muguga Machakos P.O. Box 30148 Tel. 25400145-21077 Nairobi Tel. 254-0154-32885 32. George Orinda KARI Headquarters 25. Robert Opondo Odhiambo P.O. Box 57811 KARI Muguga Nairobi P.O. Box 30148 Tel. 254-2-58301-9 Nairobi Tel. 254-0154-32885

26. Jane Ininda KARI Muguga P.O. Box 30148 Nairobi Tel. 254-0154-32885

35 36 Advances in Maize Streak Virus Disease Research in Eastern and Southern Africa

Workshop Report 15-17 September 1999 KARI and ISAAA AfriCenter, Nairobi, Kenya

Edited by

Florence Wambugu and John Wafula

Published in collaboration with the Kenya Agricultural Research Institute (KARI)

No. 16-2000 37 Published by: The International Service for the Acquisition of Agri-biotech Applications (ISAAA). Ithaca, New York.

Copyright: (2000) International Service for the Acquisition of Agri-biotech Applications (ISAAA)

Reproduction of this publication for educational or other noncommercial purposes is authorized without prior permission from the copyright holder, provided the source is properly acknowledged.

Reproduction for resale or other commercial purposes is prohibited without the prior written permission from the copyright holder.

Citation: Advances in Maize Streak Virus Disease Research in Eastern and Southern Africa, Workshop Report, 15-17 September 1999, KARI and ISAAA AfriCenter, Nairobi, Kenya. ISAAA Briefs No. 16. ISAAA: Ithaca, NY. 43 p.

ISBN: 1-892456-20-6

Publication Orders: Please contact the ISAAA SEAsiaCenter or write to [email protected] ISAAA SEAsiaCenter c/o IRRI MCPO Box 3127 Makati City 1271 The Philippines

Info on ISAAA: For information about ISAAA, please contact the Center nearest you: ISAAA AmeriCenter ISAAA Afri Center ISAAA Euro Center ISAAA SEAsia Center 260 Emerson Hall c/o CIP c/o John Innes Centre c/o IRRI c/o Cornell University PO 25171 Colney Lane MCPO Box 3127 Ithaca, NY 14853 Nairobi Norwich NR4 7UH Makati City 1271 USA Kenya United Kingdom The Philippines

or write to [email protected]

Electronically: For Executive Summaries of all ISAAA Briefs, please visit www.isaaa.org The full versions of ISAAA Briefs are also published electronically on behalf of ISAAA by CABI Publishing throughAgBiotechNet at: http://www.agbiotechnet.com Price: Cost US$ 25 per copy, including postage. Available free of charge to developing countries.

ii38 Contents

Foreword...... iv Preface...... v Keynote Address...... vi Acknowledgments...... viii

1. Overview...... 1 1.1 Introduction...... 1 1.2 Maize Streak Virus (MSV)...... 3 1.3 Collaborating to Combat MSV...... 3 1.4 Objectives of the MSV Workshop...... 4 1.5 Workshop Summary: Achievements of Phase I of the MSV Project...... 6

2. Status of MSV in Africa...... 6 2.1 Activities and Roles of International Organizations...... 6 2.1.1 Introduction...... 6 2.1.2 CIMMYT and MSV...... 6 2.1.3 West Africa: Breeding for MSV Resistance–IITA Experience...... 8 2.2 Status of MSV in Various African Countries...... 9 2.2.1 MSVD Occurrence, Development and Distribution in Malawi...... 9 2.2.2 Country Report: Uganda...... 10 2.2.3 MSV in Ethiopia...... 11

3. The Ecology of MSV...... 12 3.1 Vectors...... 12 3.1.1 Cicadulina spp. and MSV...... 12 3.1.2 Cicadulina Population Dynamics and Species Distribution...... 12 3.1.3 Discussions, Unanswered Questions...... 16 3.2 Disease and Epidemiology...... 17 3.2.1 Introduction and Overview...... 17 3.2.2 Studies on the Variation of Maize Streak Virus in Kenya...... 18 3.3 Breeding and Biotechnology...... 23 3.3.1 Overview...... 23 3.3.2 DNA Polymorphisms among Sources of MSV Resistance and

Disease Expression in F2 and F3-derived Lines...... 24 3.3.3 Breeding for MSV Resistance as a Priority in Kenya...... 28 3.4 Private Sector–Maize Seeds and MSV...... 30 3.4.1 Kenya Seed Company and MSV...... 30 3.4.2 Seed Production and MSV in Zimbabwe...... 31

4. Achievements, Results, and Recommendations...... 32 4.1 Studies on MSV-Resistant Maize Lines...... 32 4.2 Studies on the MSV Vector...... 32 4.3 Studies on MSV...... 32 4.4 Collaborative Achievements and Benefits...... 32 4.5 Recommendations...... 33

Participants...... 34

39iii Foreword

The Maize Streak Virus Workshop marked the nology techniques as agro-inoculation, molecular culmination of the first phase of a collaborative marking, and DNA polymorphism, microsatellite effort spearheaded by ISAAA to gain a better analysis provided through the proprietary technol- understanding of maize streak virus (MSV) dis- ogy of Novartis in France to Kenya Agricultural ease. Launched in 1996 in response to Kenya’s Research Institute (KARI) scientists, is a remark- need to stabilize yields, the project is providing able achievement. And the collaborative efforts national research groups in Africa with the data and required to locate MSV-resistant germplasm information they need to combat MSV through involved not only institutions in industrialized improved resistance breeding. Prior to this project, countries, such as CIRAD (France) and Pannar scientists lacked basic information about how the (South Africa), but also institutions in developing virus was spread, what strains of the virus were countries such as KARI (Kenya), CIMMYT (Mexico), most virulent, and what types of MSV resistance IITA (Nigeria), and others. Such South/South were most effective. To answer these questions, collaboration helped lay the groundwork for the several MSV-resistant types of germplasm from current regionalization of the project; in fact, the Kenya and elsewhere were tested in multilocation Workshop Report already includes MSV status trials for disease incidence, disease severity, and reports from Ethiopia, Malawi, and Uganda. Such ecological adaptability. Using proprietary tech- regional cooperation builds confidence, increases nologies donated by the John Innes Centre in the the speed of research, and uses resources more UK and the University of Cape Town in South efficiently. Africa, scientists from Kenya were also able to characterize virus isolates and identify the best We are pleased to disseminate the information maize lines to use for breeding MSV-resistant presented at this Workshop to a larger audience. varieties. This success has moved the project into Both KARI and ISAAA hope that the scientific its second phase, and work is now being directed information contained here will be useful to others towards developing high-yielding, MSV-resistant and that the collaborative structure of this project hybrids for the region’s small-scale farmers. Local may serve as a model for similar efforts. We and international seed companies operating in believe that partnerships between the North and Africa have also been involved for greater impact. South, between the private and public sectors, are the most effective way to transfer the benefits of Held in Nairobi, Kenya in September 1999, the agri-biotechnology to those who need them most: Maize Streak Virus Workshop brought together the world’s resource-poor farmers. Finally, we African scientists from different countries involved would like to express our special thanks to The with MSV and related maize improvement pro- Rockefeller Foundation for funding this project and grams, policymakers, and other key players from the Maize Streak Virus Workshop. around the world. Its accomplishments, further- more, testify to the power of the international partnerships that made its work possible. The Florence Wambugu donation of technology and training by institutions Director in developed countries, for example, provided ISAAA AfriCenter, Nairobi, Kenya Kenyan scientists with such cutting-edge biotech-

iv40 Preface

Diseases, whether human or plant, have no fixed region before embarking on phase II. boundaries. They do not stop at a nation’s bor- The project has developed an interdisciplinary ders. That is why I am pleased with the presence at research approach that fosters team building within this workshop of participants from this region and and between institutions. By strengthening the beyond. Institute’s research capability through training and the acquisition of technology—particularly The objectives of the Kenya National Food Policy agroinoculation techniques—the project has made Paper No.1 of 1994 and Kenya Agricultural Re- important advances in understanding MSV dis- search Institute are to meet the need for food self- ease. sufficiency and to raise the quality of life of Kenyans. Maize is the staple food throughout most Ladies and gentlemen, your scientific efforts to of Kenya, and we all know that in Africa a lack of manage this disease should focus on finding grain, especially maize, means famine. Yields of relevant, practical solutions that are farmer friendly. the crop are dangerously low. The current yield If our people’s basic food, health, and education gap in Kenya and other African countries is wide: needs are not met, we will not join the ranks of 2.3 to 4.7 tons lower per hectare than the crop’s developed nations. Real development requires potential yield. This gap is exacerbated by a not just independence and democratic participa- population growth rate that is even higher than the tion but also solutions to poverty, hunger, and cereal crop production growth rate. A concerted environmental destruction. These are some of the effort is needed to reduce this yield gap, which we areas in which you scientists must play a vital role can achieve by eliminating or reducing the factors in. This practical vision of improving your fellow that are contributing to low yields. citizen’s lives and protecting the environment should be your inspiration. Never forget that your Socioeconomic and infrastructure constraints are work on MSV is making a real difference. It has the the primary issues that afflict farmers in Africa. potential to double current maize yields through However, there are also important biotic and new cultivars that are tolerant to MSV, other maize abiotic factors, such as pests, diseases, weeds, diseases, and pests. and soil fertility problems. One such factor, maize streak virus (MVS), is endemic to Africa, appearing I am therefore pleased to announce that phase II of in Southern, Eastern, and Western Africa. Reports the MSV project has been developed and ap- have indicated that in some instances MSV proved for funding. I wish to express my sincere disease causes yield losses of 100%. thanks to the Rockefeller Foundation for having supported this project and for agreeing to finance Here at KARI, tremendous achievements were its second phase. We at KARI are indeed very recorded during the first phase of the MSV project, appreciative of the support that the Rockefeller which focused on understanding the mechanism of Foundation has given and is still giving to this and MSV resistance in maize lines from Kenya and other KARI projects. I thank them for their commit- other parts of Africa. The project acquired several ment to promote agricultural research in Africa and MSV-resistant types of germplasm and tested specifically for sponsoring this workshop. them in multilocation trials for disease incidence, disease severity, and ecological adaptability. We Finally, it gives me great pleasure to express have detected variability in the MSV genome and appreciation to all the participants, especially in the virulence of the disease in different popula- those from other countries in the region. Let me tions. take this opportunity to thank all the other collabo- rators: ISAAA, for their tireless facilitation of phase I; Transmission of the disease involves Cicadulina and International Center for Insect Physiology and vectors, and we have identified various hosts. The Ecology (ICIPE), the University of Cape Town, and team also acquired agroinoculation technology to the John Innes Centre, UK, for providing our screen for the disease, and they were trained in researchers with facilities to do molecular studies. vector taxonomy and in the molecular (RFLP) Lastly, I thank the MSV team for their good work. comparison of maize genotypes. Due to the importance of some of their findings, the project Cyrus N’Diritu found it necessary to share some of their experi- Director, KARI ence and information with other scientists in the Nairobi, Kenya

41v Keynote Address

I am indeed pleased with the kind of science that resistance to MSV for Kenyan farmers. Finally, it is the MSV project has promoted. The project has our hope that eventually the project will develop extended into Eastern and Southern Africa, and its maize varieties that are resistant to MSV. benefits will expand even further to include others seeking to develop similar projects of their own. It A higher level of resistance to pests and diseases is exciting to see this regional impact. The in maize is one of the most critical objectives for Rockefeller Foundation is also eager to see this crop improvement teams in eastern and southern project develop genetic controls for MSV that can Africa. Resistance to pests and diseases contrib- be deployed by small-scale farmers. We believe utes to higher and more stable yields and is that improved crops are going to become increas- particularly relevant to poor farmers, who can ill ingly significant in the future, and our push to afford either investments in crop protection or include developing countries in the benefits such losses that occur as a result of these constraints. crops can provide is an important part of our vision. Because grain yield is easier to evaluate and Such initiatives will require much collaboration, and compare among varieties being tested for release it is gratifying to see the MSV project’s collabora- criteria, however, resistance traits have long been tive efforts. neglected in crop improvement throughout the world, including Africa. The Rockefeller Foundation formerly focused more on such issues as soil fertility for small-scale The project, “Understanding the mechanisms of farms, but now a full portfolio of genetic improve- maize streak virus resistance of maize in Kenya” ments—including Striga resistance in maize—exists. aimed to contribute to rectifying this bias. Results to Striga, as we all know, is another major limiting date have been encouraging. By importing, testing, factor in African maize production. Weevil resis- and employing new sources of resistance to maize tance is also an urgent need. In cassava, the streak virus (and other important diseases) in Rockefeller Foundation is keen to map genes with developing new breeding populations and lines, resistance to cassava mosaic virus (CMV). There the project has considerably broadened the is also an increased need for ecosystem breeding genetic variation available to maize breeders in programs. These are programs that encourage a Kenya. Indeed, it has been especially encouraging multidisciplinary approach and avoid duplication to see the results of this work being rapidly inte- of research efforts among ecosystems that share grated into the core maize breeding program at the same biotic and abiotic constraints. KARI. As the results of these actions become better known, there is reason to hope that this This MSV project integrated new techniques and knowledge will spread to other countries of the methodologies, such as agroinoculation and region, contributing further to a decline in yield mapping resistance genes, which end up in marker losses in the fields of farmers who desperately R selection. Marker R selection or marker-assisted desire to increase their productivity levels. selection is the selection for resistant genes. It incorporates novel, valuable components to An added benefit of the project has been a deeper advance its effectiveness, including the recognition and more widely shared understanding of the of hardworking individuals, the regional improve- disease itself. With this understanding has come a ment of scientific capacity, new positive inputs into greater willingness among researchers of differing KARI’s maize breeding program, and the combina- backgrounds and views to work together and tion of biotechnology breeding with other methods. develop better solutions for farmers. Equally The project presents numerous opportunities to important, several researchers at KARI have train scientists in such advanced research institu- gained vital exposure to molecular techniques of tions as the John Innes Centre, the University of maize genetics research. This experience has Cape Town, Novartis Seeds, other entities that are broadened their vision for what is possible in future exploring the frontiers of science, and the interna- phases of crop improvement in the country. tional agricultural research centers (IARCs). These investments in training are helping to build capacity The future holds many challenges for this and in agri-biotechnology throughout Africa. The related initiatives. Maize streak virus resistance is project has also resulted in a national screening now known to be controlled by one major gene with program designed to discover the best sources of several modifiers. As researchers at CIMMYT and

vi42 elsewhere continue to identify closer-linked Foundation therefore salutes the collaborators of molecular markers for the major gene located on this initiative and wishes them well in their efforts. chromosome 1 of maize, it will become possible to routinely screen for this trait in early generation In the meantime, we expect that the second phase evaluations of genotypes. If similar progress can will result in a fully equipped, developed, and be made with other diseases important to the operational laboratory dealing with R marker region, the possibility of rapidly “stacking” resis- systems and other maize diseases. This operating tance genes in adapted materials will become a laboratory will allow scientists from various disci- reality. plines to collaborate on developing a successful concept of virus/host interactions, which can help The benefits of this research to poor and provide information on various isolates and marginalized African farmers are obvious. The whether resistance really does break down. Thank “dual track” approach employed by this project you all and I am glad to be one your collaborators. (represented by the equal emphases given to conventional, product-oriented work, and longer- term, molecular aspects of crop improvement), Joe DeVries moreover, could perhaps be proposed as a Rockefeller Foundation model for future interventions. The Rockefeller Nairobi, Kenya

43vii Acknowledgments

The Kenya Agricultural Research Institute (KARI) We deeply appreciate the assistance and editorial and International Service for the Acquisition of help of Anatole F. Krattiger, David Alvarez, Otula AgriBiotech Applications (ISAAA AfriCenter) wish to Owuor, and Michael Njuguna in the publication of thank the staff members of both organizations and this brief. those of the International Center for Insect Physiol- ogy and Ecology (ICIPE), John Innes Center (JIC), We also wish to acknowledge the contribution of Novartis University of Cape Town (UCT), and Novartis Seeds for letting the project and Kenyan part- Seed for their contributions to the success of the ners use their patented agroinoculation tech- workshop and the publication of this proceedings. nology through their Molecular Marker Laboratory at Many KARI scientists were directly involved in the Toulouse, France, and for backstopping KARI scientists. organization of the workshop. Macharia Gethi Dr. Jane Inida’s short-term training on the use of modern played an instrumental role in heading the commit- technology in gene and gene marker identification through tee that organized the workshop. Ben Odhiambo microsatellite analysis of MSV-resistance linkage maps and Jane Inida worked tirelessly in arranging the from different parent lines have helped identify genes that workshop. We thank Michael Njuguna of ISAAA confer MSV resistance in five mapping populations from AfriCenter who coordinated the invitation of all different sources used in phase I of the project. Thus, workshop speakers and participants. The joint selection, combined with targeted breeding, is expected efforts of Florence Wambugu, John Wafula, and to enhance the development of MSV-resistant varieties Michael Njuguna have resulted in the timely during phase II. Without this modern technology, the pro- publication of this brief. cess would have taken more years than it will. Finally, we are especially grateful to the Thanks are also due to all distinguished speakers, Rockefeller Foundation for funding both the project session chairs, rapporteurs, and all participants and the workshop. John Lynam and Joe DeVries who were involved in the presentations and who of the Foundation have been instrumental in joined in various stimulating discussions in the guiding the successful implementation of the different workshop sessions. Their inputs ensured project; their deep knowledge and dedication to a high level of discourse and helped to provide this project are gratefully acknowledged. useful guidance on complex issues. This work- shop would not have been successful without the support of the KARI secretariat that coordinated most activities, including the field trip to Muguga station.

viii44 Advances in Maize Streak Virus Disease Research in Eastern and Southern Africa

Workshop Report 15-17 September 1999 KARI and ISAAA AfriCenter, Nairobi, Kenya

Edited by Florence Wambugu and John Wafula

Published in collaboration with the Kenya Agricultural Research Institute (KARI)

No. 16-2000 45 US$ 25.00 ISBN: 1-892456-20-6

46