Tải bản đầy đủ - 0trang
Chapter 1. Cytogenetics and Genetics of Pearl Millet
PREM P. JAUHAR AND WAYNE W. HANNA
XI. Genetics of QualitativeTraits
XIII. Genetics of Quantitative Traits
xn! Conclusion and Perspectives
Pennisetum is one of the most important genera of the family Poaceae. It includes such important species as pearl millet, Pennisetum glaucum (L.) R. Brown
[ =Pennisetum typhoides (Bum.) Stapf et Hubb., Pennisetum americanum (L.)
Schumann ex Leeke] (2n = 14),a valuable grain and forage crop; and its tetraploid
relative Napier grass (I? purpureum Schum.) (2n = 4x = 28), prized for its fodder
grown throughout the wet tropics of the world. Pearl millet is widely cultivated in
different parts of the world. It is a multipurpose cereal grown for grain, stover, and
green fodder on about 27 million hectares, primarily in Asia and Africa (ICRISAT,
1996). In terms of annual production, pearl millet is the sixth most important cereal crop in the world, following wheat, rice, maize, barley, and sorghum. Among
the millets, it is second only to sorghum.
Pearl millet is the only cereal that reliably provides both grain and fodder on
poor, sandy soils under hot, dry conditions. It is remarkable that it produces nourishment from the poorest soils in the driest regions in the hottest climates. In the
drier regions of Africa and Asia, the crop is a staple food grain. In more favored
areas, however, pearl millet grain is fed to bullocks, milch animals, and poultry. In
areas where other types of feed are not available, stover provides feed for cattle
(ICRISAT, 1996). Pearl millet is also grown in several other countries. It was planted to almost 1 million hectares in Brazil in 1996. In the United States, it is grown
as a forage crop on an estimated half a million hectares. It is also grown as a forage crop in tropical and warm-temperate regions of Australia and several other
countries (Jauhar, 198la).
Pearl millet is an ideal organism for cytogenetic and breeding research. Several favorable features of its chromosome complement--e.g., the small number and
large size of chromosomes with distinctive nucleolar organizers-make pearl millet a highly suitable organism for cytogenetic studies. Because of its low chromosome number, pearl millet offers a particularly favorable material for aneuploid
analysis and thereby elucidation of its cytogenetic architecture. Moreover, its protogynous flowers and outbreeding system make it ideal for interspecific hybridization and breeding work, particularly heterosis breeding. Pearl millet has
also been found suitable for molecular studies.
Although pearl millet has great agricultural importance and is a favorable organism for cytogenetic and molecular studies, it has not received the attention it
deserves. Consequently, the information available on its genetics and cytogenetics is far less than that available for other agricultural crops. In a comprehensive
CYTOGENETICSAND GENETICS OF PEARL MILLET
review, Jauhar (1981a) compiled the available literature on the cytogenetics and
breeding of pearl millet and related species. The purpose of this article is to summarize the information on cytogenetics and genetics of pearl millet mostly since
the publication of Jauhar’s book (198 la).
Pearl millet originated in West Africa, where it grows in chronically droughtprone areas. Selection exercised by early cultivators within a variety of cultural
contexts resulted in a multitude of morphologically diverse forms. The protogynous flowers of pearl millet facilitated the introgression of characters from related wild species to cultivated annual species. Although researchers generally agree
that pearl millet is of African origin, pinpointing its specific region of origination
has been controversial. Vavilov (1949-1950) placed pearl millet in the Ethiopian
Center of Origin (particularly Abyssinia and Sudan), considering this the region
of maximum diversity. However, the center of diversity is not always the center of
origin (Harlan, 1971). In light of the great morphological diversity present in introductions from Central Africa, Burton and Powell (1968) inferred that pearl millet originated there.
Another method used to pinpoint its center of origin is the occurrence of B chromosomes. Because B chromosomes frequently occur in primitive varieties but not
in commercially bred cultivars, Muntzing ( 1958) suggested that their occurrence
might indicate a crop’s center of origin. Therefore, based on the occurrence of B
chromosomes in pearl millet collections, some researchers consider Sudan (Pantulu, 1960) and Nigeria (Powell and Burton, 1966; Burton and Powell, 1968) to be
the crop’s centers of origin. However, drawing conclusions on the basis of occurrence of B chromosomes may not be scientifically sound (Jauhar, 1981 a), because
several ecological and edaphic factors influence the occurrence of B chromosomes. In rye (Secafecereale), for example, the frequency of Bs is higher in rnaterial growing on acidic soils than on basic soils (Lee, 1966). Working on clonal
plants of rye grown under different regimes of soil, temperature, and humidity,
Kishikawa (1970) found that the frequency of Bs was lower in progeny derived
from plants grown under high temperatures or dry soil conditions.
Considering that the greatest morphological diversity of pearl millet occurs in
West Africa, south of the Sahara Desert and north of the forest zone, and that the
wild progenitor also occurs in the drier, northern portions of this zone, Harlan
(197 1 ) suggested that the center of origin lies in a belt stretching from western Sudan to Senegal. Based on present-day distributions, the Sahel region of West Africa
appears to be the original home of pearl millet (Brunken et al., 1977). The cultivated types show the highest level of morphological variability in this region
(Clegg et al., 1984).
PREM P. JAUHAR AND WAYNE W. HANNA
Traditionally, characterization of genetic resources of crop plants has been accomplished through a combination of morphological and agronomic traits, e.g.,
growth habitat, earliness, and disease and pest resistance. Biochemical and molecular markers have also been used to obtain additional information on a crop
plant’s center of domestication, the effect of domestication on genetic diversity,
and potential gene flow between wild and cultivated types (Gepts and Clegg,
1989). However, using restriction fragment length polymorphisms (RFLPs)
among chloroplast, nuclear ribosomal RNA, and alcohol dehydrogenase (ADH)
sequences in a group of 25 wild and 54 cultivated accessions of pearl millet, Gepts
and Clegg (1989) could not identify the precise pattern of its domestication.
Brunken et al. (1977) hypothesized the existence of several independent domestications of pearl millet in the southern fringe of the Sahara. Based on polymorphisms in 12 genes coding for 8 enzymes in 74 cultivated samples and 8 wild
samples from West Africa, the 82 samples were classified into three groups: (1)
wild types, (2) early maturing cultivars, and (3)late cultivars (Tostain et al., 1987).
The early maturing cultivars were found to have the highest enzyme diversity,
whereas cultivars from Niger showed the most diversity. The high diversity of the
early maturing group and its extensive divergence from West African wild millets
further suggest multiple domestications.
III. TAXONOMIC TREATMENT
Pearl millet is the most important member of the genus Pennisetum in the tribe
Paniceae. It has received a variety of taxonomic treatments, and its scientific binomials have been frequently shuffled by a variety of taxonomists. Consequently,
it has had many Latin names, perhaps more than any other grass. In the post-Linnaean period from 1753 to 1809, pearl millet was treated as a member of at least
six different genera, namely, Panicum, Holcus, Alopecuros, Cenchrus, Penicillaria, and Pennisetum (see Jauhar, 1981a,c).
At the beginning of this century, pearl millet was commonly referred to as Pennisetum typhoideum, Penicillaria spicata, Panicum spicatum, and Pennisetum
alopecuroides (Chase, 1921). By the mid-19th century, however, pearl millet was
generally called Pennisetum typhoideum L. C. Rich, but this nomenclature was not
widely accepted. The Latin name Pennisetum americanum given by K. Schumann
(1895)-apparently based on the first name “Panicum americanum L.” used by
Linnaeus (1753bwas accepted by Terrell (1976) and hence used by several
American workers. However, this name is inappropriate and misleading because
it inadvertently implies the American origin of pearl millet (Jauhar, 1981a,c).
CYTOGENETICSAND GENETICS OF PEARL MILLET
Stapf and Hubbard (1933, 1934) gave the name Pennisetum fyphoides (Bum.)
Stapf et Hubb., which was accepted by several modem taxonomists, including Bor
(1960), and used by most pearl millet workers outside the United States. In the
1960s, American workers joined the rest of the world in calling pearl millet Pennisetum ophoides (Burton and Powell, 1968).The name Pennisetum glaucum (L.)
R. Br., based on Panicum glaucum (L.) R. Br., was adopted by Hitchcock and
Chase (195 1) in Manual of the Grasses of the United States. Consequently, American scientists currently engaged in research on pearl millet use this name.
All annual and perennial members of the section Penicillaria fall under the x =
7 group. They have typically penicillate anther tips. Whereas most penicillarias are
diploid with 2n = 14 chromosomes, one, viz., Napier grass, is a perennial
Of the 32 species described by Stapf and Hubbard (1934) in the section Penicillaria of the genus Pennisetum, only two have been found outside Africa. There
is considerable variation in seed and other characters both between and within different cultivars or races. Such variation could be attributed to independent domestications and migrational events resulting in geographical isolations. The protogynous nature of pearl millet and its intercrossabilitywith its wild relatives must
have generated much of the existing genetic diversity. Meredith (1955) described
four taxa, which he called “allied species,” closely related to pearl millet: Pennisetum americanum, I? nigritarum, I! echinurus, and I? albicauda. Since these
are interfertile with pearl millet, they were merged into a single species with pearl
millet (Brunken et al., 1977). However, for the sake of convenience, Brunken subdivided the morphologically heterogeneous pearl millet species he called “Pennisetum americanum” into three subdivisions: ( 1) ssp. americanum encompasses
the wide array of cultivated pearl millets; (2) ssp. monodii includes all the wild and
semiwild diploid races that are fully fertile with pearl millet and therefore form a
single reproductive unit with it; and (3) ssp. stenostachyum is morphologically intermediate between the two preceding species.
Amoukou and Marchais (1993) found some evidence of a partial reproductive
bamer between wild and cultivated pearl millets. Crosses between 16 cultivated
accessions (f? glaucum ssp. glaucum) (as female parents) and 11 wild accessions
(f? glaucum ssp. monodii), from the whole range of diversity of the species,
showed certain degrees of seed malformation and reduced 1000-grain-weightand
germination ability. These are manifestations of a genetic imbalance between the
cultivated and the wild groups, probably resulting from reproductive barriers that
developed during the domestication process.
PREM P. J A W AND WAYNE W. H A N N A
RELATIVESOF PEARL MILLET
Elephant or Napier grass, Pennisetum purpureum (2n = 4x = 28), is a perennial relative of pearl millet (see Section V). It has typically penicillate anthers. Native to Africa, it is a robust perennial with creeping rhizomes. It was introduced
into the United States in 1913. It is extensively grown in the humid tropics throughout the world.
N.CHROMOSOMES, KARYOTYPE, AND MEIOSIS
OF 5,7,8, AND 9
The genus Pennisetum is a heterogeneous assemblage of species with chromosome numbers as multiples of 5 , 7, 8, and 9, for example, P. ramosum (2n = lo),
P. ryphoides (2n = 14) and P. purpureum (2n = 28), P. massaicum (2n = 16,32),
and P. orientale (2n = 18, 36, 54). The chromosome morphology is diverse and
substantial size differences exist. A notable feature is that species with lower chromosome numbers have larger chromosomes. Thus, pearl millet (2n = 14) and P.
ramosum (2n = 10) have relatively large chromosomes, larger than those of other members of the tribe Paniceae. In contrast, species with higher chromosome
numbers, e.g., I? orientale (2n = 18), have strikingly smaller chromosomes than
those of pearl millet (2n = 14) (Fig. 2C).
A characteristicfeature of perennial species of Pennisetum is the occurrence of
chromosomal races or cytotypes, e.g., P. orientale L. C . Rich. (2n = 18, 27, 36,
45, 54) and F! pedicellatum Tin. (2n = 36,45,54). However, no such cytotypes
occur in the annual cultivated or wild pearl millets, all of which have 2n = 14 chromosomes.
Rau (1929) was the first to determine the somatic chromosome number of pearl
millet as 2n = 14, and he mentioned these chromosomes as being large. The chromosomes have median to submedian centromeres; the shortest chromosome pair
is satellited, and during meiosis the shortest bivalent is associated with the nucleolus. The chromosomes of diploid taxa of the section Penicillaria are similar to
those of pearl millet. Thus, I? ancylochaete, P. gambiense, I! maiwa, and I? nigritarum have 2n = 14 chromosomes, and their chromosome morphology is similar
to one another and to chromosomes of pearl millet (Veyret, 1957). Not surpris-
CYTOGENETICS AND GENETICS OF PEARL MILLET
ingly, therefore, these taxa are interfertile with pearl millet, and there is no barrier to gene flow across these taxa.
Pennisetum violaceum and R mollissimum, the two close wild relatives that
form a primary gene pool with pearl millet, and I? schweinfurthii (a representative
species of tertiary gene pool) were assessed for their genomic organization, using
in situ hybridization with rDNA probes on somatic metaphase spreads and interphase nuclei (Martel et al., 1996). These studies showed chromosomal similarity
of rDNA sequence locations in the three taxa in the primary gene pool.
Pearl millet regularly forms seven bivalents at meiotic metaphase I. A characteristic feature is the rapid terminalization of chiasmata, such that at diakinesis
mostly loose ring bivalents with two terminalized chiasmata each are observed.
The annual, semiwild taxa also have regular meiosis with 7 11. They all have the
genomic constitution AA.
Recently, Reader et al. (1996) used fluorescence in situ hybridization (FISH) to
characterize the somatic complement of pearl millet. A metaphase spread was hybridized with Fluorored-labeled rDNA (derived from plasmic clone pTa71; Gerlach and Bedbrook, 1979) and then stained with DAPI. In that double exposure.
two large and two small NOR loci were observed.
Napier grass is a perennial relative of pearl millet. Burton (1 942) determined its
somatic chromosome number as 2n = 28 chromosomes. It is an allotetraploid (2n
= 4x = 28) with diploidlike meiosis (see Jauhar, 1981a). It is genomically represented as AABB, the A genome being largely homologous to the A genome of
pearl millet (see Section V).
Researchers generally believe that several crop species have evolved from
species with lower basic chromosome numbers, with increase in chromosome
number occurring by means other than straight polyploidy. Evidence supporting
this view has been found by RFLP studies of maize (Helentjaris et al., 1986;
Whitkus et al., 1992), brassicas (Slocum et al., 1990; Kianian and Quiros, 1992),
and sorghum (Hulbert et al., 1990; Whitkus et al., 1992; Chittenden et ul., 1994).
Based on cytogenetic evidence, Jauhar (1968, 1970a, 1981a) hypothesized that x
= 5 may be the original basic number in Pennisetum and that pearl millet (2n =
14) may be a secondary balanced species as a result of ancestral duplication of
chromosomes. If duplication of a part of the original genome occurred during the
evolution of pearl millet, some duplicate loci should be observed in the present
genome. Liu et al. (1 994) indeed detected several duplicate loci in their RFLP linkage map of the pearl millet genome. However, further studies are needed to fully
characterize the duplicated regions of the genome.