Tải bản đầy đủ - 0 (trang)
II. Characters Used to Fingerprint Cultivated Varieties

II. Characters Used to Fingerprint Cultivated Varieties

Tải bản đầy đủ - 0trang

FINGERPRINTING CROP VARIETIES



87



imperative to reach a consensus on the practicality and validity of using

morphological data for cultivar identification and associated uses. Such

consideration is especially important because new demands that are made

on cultivar description often go beyond that of identification alone. These

include a need to calculate the genetic distance between cultivars, information that cannot be provided by morphological data.

2. Constraints upon Morphology



a. A Narrowing of Morphological Diversity

Morphological descriptions can provide unique identification of cultivated varieties (Molina-Can0 and Elena Rossello, 1978). However, their

ability to provide reliably discriminating identification is at best cumbersome (Patterson and Weatherup, 1984). Increased numbers of genetically

related releases by plant breeders have made unique identification more

difficult to achieve. This problem is especially acute in crops with limited

levels of elite genetic diversity or when convergent selection toward similar

morphologies is practiced (Wagner and McDonald, 1981). This situation

has been further exacerbated in wheat with the use of nontraditional

breeding crosses that have disrupted the previous correlation between

grain morphology and quality (Lookhart and Bietz, 1990).

b. The Effect of Environment on Morphology

Morphologies reflect not only the genetic constitution of the cultivar, but

also the interaction of the genotype with the environment (G x E) within

which it is expressed (Gottlieb, 1977; Brown, 1978; Lin and Binns, 1984;

Patterson and Weatherup, 1984). For example, “Welsh Bearded Red

Rough Chaff’ rivet wheat (Triticum aestivum) has an ear color that “depends on the season and varies from dark red to pale yellowish red”

(Zeven, 1990). Due to G X E effects, it is clearly inappropriate to compare

morphological data for varieties that have been collected across different

years and/or locations. Attempts to allow comparisons to be made across

more varieties grown in different years and/or locations, or sets of locations, are made by including a list of standard check varieties at each

location site in each year. However, G x E interaction effects have been

found to cause aberrant adjusted means for traits such that morphological data collected in field plots can provide, at best, only an initial screen

of varietal identity or distinctiveness (Higgins ef al., 1988). Subsequent

repeated tests of individuality would thus still need to be made, requiring

further replicated test plots of varieties.

Recent investigations using digital image analysis techniques have shown

that objective analysis of the morphologies of roots, leaves, and particularly of cereal grains could provide some degree of varietal identification



88



J. S. C. SMITH AND 0. S. SMITH



(Keefe and Draper, 1986,1988; Neuman et al., 1987; Symons and Fulcher,

1988). However, when image analysis was tested for its ability to correctly

identify varieties on an individual grain basis for each of 14 cultivars, the

percentage of correct positive identifications ranged from a low of 15% to a

maximum of 96%. It was concluded that the ability to provide correct

identification of cereal varieties using grain morphology was confounded

by close genetic relatedness among cultivars and by environmental effects

on grain shape.



c. The Inability of Morphological Data to be Used for the Estimation of

Genetic Distance

Descriptions based on morphological data are fundamentally flawed in

their ability to provide reliable information for the calculation of genetic

distance or the validation of pedigrees. First, distances that reflect the

genetic constitution of cultivated varieties can only be obtained using

measurements from traits that adequately sample the genome. Second, the

efficiency of the sampling sites is greatly increased if there are numerous

variants that can be unambiguously and repeatedly scored. Third, there

needs to be a direct and unambiguous correspondence between the genotype and the phenotype. Morphological data fail to meet these criteria.

First, much morphological variation cannot be consistently measured because of the effects of G x E interaction and the small effects of numerous

quantitative genes. Second, much obvious morphological variation has

been eliminated with the consequence that most varieties outwardly

appear similar. Third, for most morphological traits, the genetic control is

unknown, although it is known that multiple genotypes can give phenotypes of similar outward appearance. Therefore, it is impossible to determine

how completely the genome is sampled by morphological descriptions or

the extent to which similar phenotypes reflect similar genotypes. We have

concluded that morphological differences cannot be interpreted to provide

accurate estimates of genetic differences (J. S. C. Smith and Smith, 1989;

J. S. C. Smith et al., 1991a).



B. PROTEIN

DATA

1. Characters That Are Selected and Methods Used to

Resolve Profiles

Proteins and DNA can be used to provide varietal profiles. They are in

popular usage because the variation for these markers is ubiquitous and

this variation can be understood in genetic terms. These characters are



FINGERPRINTING CROP VARIETIES



89



in routine usage and are widely accepted as a source of reliable data in

evolution, taxonomy, and genetics (Tanksley and Orton, 1983; Doebley,

1989; Crawford, 1990; Weir, 1990). Proteins are molecules with net electrical charges that are affected by pH. Proteins can be separated by electrophoresis on the basis of their net electrical charge, molecular weight [see

Cooke (1984) for a brief listing of methods], isoelectric point, or combinations of these criteria utilizing multidimensional separations (Figs. 1-4).

They can also be separated by chromatography, wherein the surface chemistry of the protein molecule determines relative retention times on an

absorption matrix (Bietz, 1986) (Fig. 5). Proteins can be visualized in

several ways. Variation in enzymatic proteins (isoenzymes) can be revealed by specific histochemical stains. General proteins can be revealed

by various staining procedures, the most sensitive of which employ silver

nitrate. Chemical stains are eclipsed in their detection power by the incorporation of radiolabeled amino acids into proteins and their subsequent

exposure on film. Proteins can also be used as antigens to generate antibodies. Similarities between varieties can then be estimated by measuring

the degree of response between antigen and antibodies raised to other

antigens (Skerritt et al., 1984; Esen et al., 1989). Serological methods have

not yet revealed unique profiles among numerous varieties. However,

some differences have been revealed (Zawistowski and Howes, 1990). If

greater specificity can be achieved, then serological techniques could provide very rapid and cost-effective means of identification.



2. Sampling of Tissue, Extraction, and Running Conditions

Although proteins are products of the primary transcripts of DNA,

environmental factors can affect qualitative and quantitative levels of protein in seed (Higgins, 1984). Proteins also interact with other compounds,

especially those that are found in nonseed organs such as roots, leaves, and

tubers. These interactions can detract from the reproducibility of protein

profiles.

Initially, a plant organ is selected as a source of protein. Then, extraction

protocols are chosen to minimize the interaction of protein with other

constituents that could be undesirable sources of variation in the protein profile. Also, a time period or stage of development must be identified during which the protein gives profiles that are stable for any given

cultivar.

Seeds provide a convenient source of tissue that is at a defined and stable stage of development (Ladizinsky and Hymowitz, 1979) and that is

free from many compounds that interfere with proteins. Seedlings provide a rich source of enzymatically active proteins and are often the



90



J. S. C. SMITH AND 0. S. SMITH



Figure 1. Gliadin electrophoretic profiles from individual caryopsis extracts of two wheat

varieties. Needles point to differences in banding profiles in the two lanes where the varieties

are adjacent.



Figure 2. Alcohol-soluble glutenin electrophoretic profiles from individual grains of a single wheat variety. Variation in banding

profiles from lane to lane shows evidence of residual variability, or “biotypes,” within the seed lot.



Figure 3. Isozyme profiles of the enzyme malate dehydrogenase separated by starch gel electrophoresis for five individuals each of

maize inbred lines and hybrids.



Figure 4. Protein profile of an inbred line of maize following a two-step, two-dimensional

electrophoretic separation of embryo proteins labeled with [35S]methioine. The proteins

range from approximately 20 to 100 kDa (bottom to top) and isoelectric points are from 5.0 to

7.0 (left to right). Proteins were separated first according to their isoelectric points. They were

then transferred to a second gel and separated according to the molecular weights of their

subunits in a dimension at right angles to the first dimension.



93



J. S. C . SMITH AND 0. S. SMITH



94



0.58



-



0.5



-



0.4



-



0.3



-



0.2



-



0.1



5



10



15



20



25



30



35



40



45



50



55



MINUTES



Figure 5. Chromatogram of a wheat variety showing its gliadin profile following separation by reversed-phase high-performance liquid chromatography.



preferred choice of tissue for isoenzyme analysis or for the incorporation

of radiolabeled compounds for detection purposes. Preliminary studies

sampling tissue in order to establish a time or development “window”

during which the protein profile remains stable and thus is potentially

reflective of genotype must be undertaken. Some examples can be found

in isoenzyme profiling for seedlings of maize (Zea mays) (Goodman and

Stuber, 1980), seedlings of barley (Hordeurn vufgare)and ryegrass (Lofiurn

spp.) (Nielsen, 1985), and leaf tissue of cocoa (Theobroma cacao) (Atkinson et a f . , 1986), apple (Mafus spp.) (Weeden and Lamb, 1985), and

cassava (Manihot esculenta) (Ramirez et a f . , 1987).

Critical analyses of the stability of protein profiles in relation to tissue

sampling and extraction are especially vital when the data are analyzed in

both quantitative and qualitative (presence or absence) terms. Preliminary

quantitative investigations to test the validity of two-dimensional protein

profiles are necessary. For example, in barley (Gorg et a f . ,1988) and maize



60



FINGERPRINTING CROP VARIETIES



95



inbred lines (Higginbotham and Smith, 1989; Higginbotham ef al.,

1989a,b), studies were carried out to measure the genetic components of

variation. Quantitative variation for protein in wheat lead Burbidge et al.

(1986) to conclude that it would be preferential to consider only the

qualitative aspects of wheat protein profiles. Similarly, Cross and Adams

(1983a) concluded that quantitative variation for a protein in globulin

profiles of maize could not be relied upon because this protein varied in its

resultant expression among cultivars and was also known to decline in

amount during seed maturation (Cross and Adams, 1983b).



3. Stability of Biochemical Profiles in the Face of

Environmental Effects

Studies that demonstrate the stability of protein profiles can be assigned

to two categories: first, a test of profile stability in the face of environmental change, and second, the demonstration of genetic control of profiles.

a. Variation in Protein Profiles of Varieties Grown in Different Locations and/or Years

Huebner and Bietz (1988) studied the stability of protein profiles for two

hard red spring wheat varieties grown in seven locations in the United

States and three Australian cultivars grown in soils of three different sulfur

levels. While there were baking-quality differences due to location affects,

no electrophoretic differences for gliadin profiles were found among different lots of flour for each variety. Likewise, Lookhart and Finney (1984)

found that frosting affected baking quality of wheat but it did not change

the gliadin profiles. Similarly, Clements (1987) found no environmental

effects on the gliadin electrophoretic profile of soft wheat. The electrophoretic similarities of profiles from immature and mature seed provided additional evidence of the intransigence of seed protein profiles to

factors other than genetic change (Clements, 1987). However, Huebner

et al. (1990) found small quantitative differences for gliadins and glutelins

according to maturity and spike position in respect to chromatographic

profiles; qualitatively, profiles were unaffected by these factors.

Lookhart and Pomeranz (1985a) detected slightly different gliadin electrophoretic profiles for two wheat cultivars grown in soils that were severely deficient in sulfur; no differences could be detected due to various

levels of soil nitrogen. Huebner and Bietz (1988) were unable to find

differences in gliadin electrophoretic profiles for seedlots grown in soils

with different sulfur levels. Quantitative variation was detected by reversed-phase high-performance liquid chromatography (RP-HPLC) but

qualitatively the chromatographic profiles were unchanged. A small degree



96



J. S. C . SMITH AND 0. S. SMITH



of quantitative variation was also observed for varieties grown in soils with



no known fertilizer deficiencies at different locations and in different years

(Huebner and Bietz, 1988). Subtle quantitative differences in peak areas

were only revealed by a computed difference plot and did not preclude

cultivar identification by RP-HPLC. Similar low levels of quantitative, but

not qualitative, variation for chromatographic profiles have been found for

seedlots produced in different environments (Marchylo and Kruger, 1984;

J. S. C. Smith and Smith, 1986, 1988b; Marchylo et al., 1990). In maize,

for example, variation due to environmental factors did not qualitatively

affect any peaks that contributed more than 2% to the total varietal profile;

93% of variation for peak area was between genotypes (J. .S. C. Smith

and Smith, 1986). For oats (Avena sativa), no qualitative differences due to

environmental effects could be detected for either electrophoretic or chromatographic profiles of avenin proteins (Lookhart, 1985b; Hansen et al.,

1988). In rice (Oryza sativa), Sarkar and Bose (1984) found environmental

effects to be insignificant for electrophoretic profiles of globulin proteins

for eight varieties grown at two locations during 2 yr. A litany of other

reports of the invariance of protein profiles with respect to environmental

or storage conditions has been made (Lee and Ronalds, 1967; Wrigley,

1970; Zillman and Bushuk, 1979a; Cooke, 1984).

The most discriminatory technique available for the separation of proteins is two-dimensional electrophoresis. This technique provides both

quantitative and qualitative data. It allows the most exhaustive test of the

stability of protein profiles for seedlots grown in different environments.

Gorg et al. (1988) showed sufficient repeatability of profiles for 13 varieties of barley from two cultivations to provide discrimination. Similarly,

Higginbotham et al. (1989a,b, 1991a,b) identified proteins from maize

embryos that were indicative of genetic differences.

b. Profile Stability Shown through the Demonstration of Genetic Control

of the Profile

The second means of establishing clear independence of protein profiles

from environmental effects is to demonstrate genetic control of proteins. It

has been possible to characterize genetically isoenzyme variants for many

enzyme systems in numerous species of cultivated plants (Tanksley and

Orton, 1983; Nielsen, 1985). Only in esterases has the variability in banding profiles been found to be so dependent upon developmental control

that, for some loci, and for some crop species, the genetic control cannot

readily be established (Nielsen, 1985; Rebordinos and Perez de la Vega,

1990). The genetic control of isoenzymes, in particular, has been thoroughly researched in many species. In maize, for example, all of the

current practical applications are dependent upon basic genetic studies



FINGERPRINTING CROP VARIETIES



97



carried out by Drs. Goodman and Stuber and their colleagues in work that

spanned a period of approximately 15 years. During that time, thousands

of crosses involving F1 and F2 crosses and back-crosses were made and tens

of thousands of individual progeny were assayed in the laboratory in order

to determine the details of the genetics of these markers [see Stuber et al.

(1988) and numerous references cited therein]. Even in crops that are not

so amenable to genetic analysis, for example, apple, it is possible to

compare isoenzymatic profiles of progeny trees with those of their parents

and to partially verify genetic control (Weeden and Lamb, 1985).

The usage of isozymes to routinely test purity in tens of thousands of

seedlots representing more than 100 different maize hybrids, (Smith and

Weissinger, 1984; Smith and Wych, 1986; Brink et al., 1989) provides

additional evidence of the freedom of isozyme expressions from environmental effects. Seedlots have been produced in the United States, Canada,

South America, and Europe and in no instance known to us has environment been found to qualitatively affect the isozyme profiles.

General protein detection either by stains or spectroscopic absorbence

methods often reveal numerous bands or peaks per individual plant or

variety-bands that may exhibit close or overlapping gel migration or

column retention times. This level of complexity in banding patterns has

prevented the establishment of genetic control for albumin and globulin

proteins (Cross and Adams, 1983a), although these profiles appear to be

strongly variety dependent (Koranyi, 1989a,b). Profiles of alcohol-soluble

seed proteins are frequently utilized for varietal description and the genetic

control of these proteins is well established, at least for some species of

cultivated plants (Shewry et al., 1983; Soave and Salamini, 1983; Wilson,

1986; Wilson et al., 1989; Graybosch and Morris, 1990; Gupta and

Shepherd, 1990).

There is abundant evidence to show that protein profiles can be obtained

for all crop species of major importance and that these profiles are independent of environmental or storage conditions. They are thus reflective of

genotype. In some cases, quantitative variation can occur due to environmental effects, but this contribution to variation is small and can be taken

into account when considering intervarietal comparisons. It is imperative

that data having a quantitative component, for example, those derived

from RP-HPLC or two-dimensional electrophoresis, be carefully examined

in order to be able to evaluate their degree of significance in discriminating

between qualitatively similar profiles. Environmental effects have only

been found to impinge upon varietal identification in conditions of extreme

sulfur deficiency; such circumstances are extremely rare and can be tested

for. Protein profiles of cultivated varieties, therefore, satisfy one prerequisite of providing a fingerprint; they are stable descriptors of genotype.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

II. Characters Used to Fingerprint Cultivated Varieties

Tải bản đầy đủ ngay(0 tr)

×