Plant Cell, Vol. 10, 297-308, Copyright © 1998, American Society of Plant Physiologists
a-Tubulin Missense Mutations Correlate with Antimicrotubule Drug Resistance in Eleusine indica
Etsuo Yamamotoa,
Linghe Zeng1,a, and
W. Vance Bairda
a Department of Horticulture, Clemson University, Clemson, South Carolina 29634
Correspondence to:
W. Vance Baird, vbaird{at}clemson.edu (E-mail), 864-656-4960 (fax).
 |
ABSTRACT |
|---|
Dinitroaniline herbicides are antimicrotubule drugs that bind to tubulins and inhibit polymerization. As a result of repeated application of dinitroaniline herbicides, highly resistant and intermediately resistant biotypes of goosegrass (Eleusine indica) developed in previously wild-type populations. Three 
-tubulin cDNA classes (designated TUA1, TUA2, and TUA3) were isolated from each biotype. Nucleotide differences between the susceptible and the resistant (R) -tubulins were identified in TUA1 and TUA2. The most significant differences were missense mutations that occurred in TUA1 of the R and intermediately resistant (I) biotypes. Such mutations convert Thr-239 to Ile in the R biotype and Met-268 to Thr in the I biotype. These amino acid substitutions alter hydrophobicity; therefore, they may alter the dinitroaniline binding property of the protein. These mutations were correlated with the dinitroaniline response phenotypes (Drp). Plants homozygous for susceptibility possessed the wild-type TUA1 allele; plants homozygous for resistance possessed the mutant tua1 allele; and plants heterozygous for susceptibility possessed both wild-type and mutant alleles. Thus, we conclude that TUA1 is at the Drp locus. Using polymerase chain reaction primer-introduced restriction analysis, we demonstrated that goosegrass genomic DNA can be diagnosed for Drp alleles. Although not direct proof, these results suggest that a mutation in an -tubulin gene confers resistance to dinitroanilines in goosegrass.
|
|
INTRODUCTION |
|---|
Goosegrass (Eleusine indica) is considered one of the most troublesome weeds in the world and occurs extensively in croplands of cotton and soybean and in turf in the southeastern United States (Mudge et al. 1984 
; Vaughn et al. 1990 ). This weed has been effectively controlled by preemergence application of dinitroaniline herbicides. However, repeated application in monocrop production has "selected" for a highly resistant (R) biotype from previously susceptible (S) biotype populations (Mudge et al. 1984 ). Although not as common, an intermediately resistant (I) biotype also has been reported (Mudge et al. 1984 ; Vaughn et al. 1990 ).
Antimitotic dinitroanilines interfere with microtubule assembly, forming a tubulin–dinitroaniline complex that disrupts polymerization and microtubule stability (Morejohn et al. 1987 ; Hugdahl and Morejohn 1993 ). In the unicellular alga Chlamydomonas reinhardtii, dinitroanilines shorten flagella and inhibit flagellar regeneration (Quader and Filner 1980 ). In the parasitic protozoa Leishmania mexicana and Trypanosoma brucei, dinitroaniline apparently binds to protozoan tubulins and inhibits proliferation of these parasites (Chan and Fong 1990 ; Chan et al. 1993 ). In higher plants, dinitroanilines work by affecting root growth, especially the development of secondary roots in susceptible plants (Hacskayko and Amato 1968 ). In the presence of the herbicide, mitosis in the goosegrass S biotype is arrested at prometaphase, cell plates are misshapen or absent, and cells enlarge isodiametrically (Vaughn 1986 ). In contrast, cells of the R biotype are indistinguishable from those of controls, with the cortical and spindle microtubules remaining intact (Vaughn 1986 ). As studies in other systems suggest, binding of the dinitroaniline to - or ß-heterodimer tubulins inhibits microtubule polymerization and also destabilizes existing microtubules (Morejohn et al. 1987 ; Hugdahl and Morejohn 1993 ). Therefore, the resistance exhibited by the R biotype could result from an alteration in a tubulin, namely, a tubulin mutation that reduces dinitroaniline binding affinity or a tubulin mutation that produces hyperstable microtubules that resist dinitroaniline-induced depolymerization.
Two types of antimitotic herbicide-resistant mutants in Chlamydomonas were selected, and missense mutations were identified. In mutant upA12, the tua1-1 allele of the 1-tubulin gene had a missense mutation in which Tyr-24 was replaced by His; this alteration in the amino acid sequence correlated with the mobility shift observed for the variant -tubulin in two-dimensional gels (James et al. 1993 ). Similarly, mutants of the ß-tubulin genes colR4 and colR15 exhibited ß-tubulin variants shifted toward more acidic positions, and these mobility shifts correlated with amino acid substitutions: Lys-350 was replaced by Glu in colR4 and by Met in the colR15 (Lee and Huang 1990 ).
To identify the mode of inheritance of dinitroaniline resistance in goosegrass, we first exposed R, I, and S biotype seedlings to dinitroanilines and then assayed radicle growth. Based on inheritance of dinitroaniline resistance in F1 hybrids and subsequent F2 and F3 generations, it appears that dinitroaniline resistance in goosegrass is inherited as a single nuclear locus with three alleles, and resistance appears to be recessive to susceptibility (i.e., S > I > R) (Zeng 1997 ; Zeng and Baird 1997 ).
As with tubulins of most higher organisms, the tubulin genes of goosegrass are organized into gene families: there are three to five -tubulin genes and four to seven ß-tubulin genes (Mysore and Baird 1995 ). However, no biotype-specific restriction fragment length polymorphisms are present between the S, I, and R biotypes (Mysore and Baird 1995 ). The tubulin isotypes of the R biotype were indistinguishable from those of the S biotype by protein gel blot analysis (Waldin et al. 1992 ), implying that resistance is not associated with a modified tubulin. However, in a patent application from the same group (Cronin et al. 1993 ), it was stated that the 1-tubulin from the R+ biotype (ELR2) was shifted toward higher molecular weight and was more acidic relative to that of the S biotype (ELS15) in two-dimensional gels. This mobility shift coincided with a C-to-T transition that results in substitution of Thr-239 in ELS15 to Ile in ELR2.
Considering the multiple forms and high levels of dinitroaniline resistance identified in goosegrass, determination of the basis of this resistance would make goosegrass a compelling model system to study the molecular and biochemical mechanisms of herbicide resistance. Analysis of this system would also improve our knowledge of the plant cytoskeleton and its role in plant morphogenesis. To identify the molecular basis for dinitroaniline resistance in goosegrass, -tubulin cDNAs were isolated and characterized by reverse transcriptase–polymerase chain reaction (RT-PCR) and rapid amplification of cDNA ends (RACE). Here, we report that the genetic results for dinitroaniline resistance (Zeng 1997 ; Zeng and Baird 1997 ) correlate with missense mutations in a specific -tubulin cDNA class.
|
|
RESULTS |
|---|
Goosegrass -Tubulin cDNAs
Three -tubulin cDNAs were each isolated from S, I, and R biotypes of goosegrass by RT-PCR, using degenerate oligonucleotides (Figure 1 Figure 2 Figure 3) (GenBank accession numbers
AF008120,
AF008121, and
AF008122, respectively). Two additional putative -tubulin cDNAs were also isolated by this method (data not shown). Therefore, four to five -tubulin genes may be expressed in goosegrass. Cronin et al. 1993 observed four -tubulin isoforms, and Mysore and Baird 1995 estimated three to five -tubulin genes. Within each isotype class, the cDNAs from the three biotypes were almost identical in sequence (Figure 1 Figure 2 Figure 3). Sequence identity between TUA1S1 (1693 bp) and ELS15 (1635 bp) (Cronin et al. 1993 ) was 99.6% in the overlapping region and that of TUA1R2 (1628 bp) and ELR2 (1576 bp) (Cronin et al. 1993 ) was 99.9%. However, the 5' and 3' noncoding sequences of TUA1, TUA2, and TUA3 exhibited significant differences compared with the coding sequences (data not shown). The length of 5' noncoding regions varied between the biotypes and among the biotype clones (the longest sequences are shown in Figure 1 Figure 2 Figure 3). Heterogeneity at the 5' noncoding regions was also reported in Chlamydomonas tubulin genes (Brunke et al. 1984 ); however, a conserved (GCAA[A/C]C) and C/ T clusters upstream of the initiator ATG codon (Brunke et al. 1984 ) were not detected in the goosegrass -tubulins. AG repeats were found upstream of the ATG codon in TUA2. A conserved polyadenylation signal, AATAAA, is present in the TUA2s (Figure 2). This sequence was also present in maize tua3 (Villemur et al. 1992 ). Polyadenylation sites varied not only between the biotypes but also among the biotype clones (Figure 1 Figure 2 Figure 3).
View larger version (62K):
[in this window]
[in a new window]
|
Figure 1.
Nucleotide Sequences of Goosegrass TUA1 cDNAs.
The deduced amino acid sequence of TUA1S1 is shown above the nucleotide sequence. The nucleotides of the other TUA1s are given on the lines below, and only the nucleotides that differ from those of TUA1S1 are shown. The translation termination codon is represented by an asterisk. The PCR primer sites are indicated with arrows, and the nucleotides are underlined.
|
|
View larger version (65K):
[in this window]
[in a new window]
|
Figure 2.
Nucleotide Sequences of Goosegrass TUA2 cDNAs.
The deduced amino acid sequence of TUA2S1 is shown above the nucleotide sequence. The nucleotide sequences of the other TUA2s are given on the lines below, and only the nucleotides that differ from those of TUA2S1 are shown. The translation termination codon is represented by an asterisk. The PCR primer sites are indicated with arrows, and the nucleotides are underlined. A conserved polyadenylation signal sequence is also underlined. Primers aIIGR4 and aIIR44 overlap each other; therefore, the nucleotides of TUA2R1 are underlined for aIIR44.
|
|
View larger version (73K):
[in this window]
[in a new window]
|
Figure 3.
Nucleotide Sequences of Goosegrass TUA3 cDNAs.
The deduced amino acid sequence of TUA3S1 is shown above the nucleotide sequence. The nucleotide sequences of the other TUA3s are given on the lines below the TUA3S1 sequence. The translation termination codon is represented by an asterisk. The PCR primer sites are indicated with arrows, and the nucleotides are underlined.
|
|
Identification of Missense Mutations
Several nucleotide differences were identified between the susceptible (wild-type) -tubulins and the resistant (mutant) -tubulins. One nucleotide change from C to T results in an amino acid substitution of Thr-239 by Ile in TUA1R1 and TUA1R2 (Figure 1); from T to C, results in substitution of Met-268 by Thr in TUA1I1 (Figure 1); and from A to G, results in substitution of Thr-340 by Ala in TUA2R1 and TUA2R2 (Figure 2). In the modified TUA1 (tua1) of the R biotype, the substitution of Thr-239 by Ile replaces a polar (hydrophilic) residue with a nonpolar (hydrophobic) residue. Four other nucleotide substitutions observed in TUA1R1 and TUA1R2 did not result in amino acid replacements.
Dinitroaniline Herbicide Response Phenotype Locus Identification
The association of these mutations with the dinitroaniline herbicide response phenotype (Drp) was tested by sequencing TUA1 and TUA2 of F3 individuals of known phenotype and genotype derived from crosses between the S and R biotypes. In TUA1, plants homozygous for susceptibility (Drp/Drp) contained only the wild-type TUA1 allele sequence (Figure 1, TUA1S3). In contrast, plants homozygous for resistance (drp/drp) contained only the mutant tua1 allele sequence (Figure 1, TUA1R2). Significantly, however, the plant heterozygous for susceptibility (Drp/drp) contained both wild-type and mutant allele sequences in an ~1:1 ratio (n = 9 for S allele and n = 7 for R allele) (Figure 1, TUA1SR). Similarly, for TUA2, Drp/Drp plants contained the wild-type TUA2 allele sequence (Figure 2, TUA2S3), and drp/drp plants contained the mutant tua2 allele sequence (Figure 2, TUA2R2). However, the Drp/drp plant contained only the mutant tua2 allele sequence (n = 14) (Figure 2, TUA2SR). Because the plants were selected for Drp, we conclude that TUA1 is the gene at the Drp locus and that the association of TUA2 is random. Linkage between TUA1 and TUA2 loci was not tested in this work. The sequences of root TUA1 and TUA2 from the R and S biotypes were the same as those of leaves (data not shown).
Similar analysis of an F1 hybrid between the I and R biotypes using direct sequencing of RT-PCR products (both strands) yielded an identical conclusion for TUA1 and TUA2. The plant heterozygous for susceptibility contained both I and R allele types for TUA1 but only the mutant allele type for TUA2 (data not shown). The allele types of heterozygotes between the I and S biotypes were not examined; however, results of genetic studies (Zeng 1997 ) suggest identical outcomes.
Drp Allele Identification by PCR–Primer-Introduced Restriction Analysis
Goosegrass genomic DNA was diagnosed for Drp locus alleles by a simple yet sensitive genotyping method (Jacobson and Muskovits 1991 ). ClaI-digested S biotype (S3, wild type) and undigested heterozygote (SR) yielded uncut 182-bp fragments as expected (Figure 4). ClaI-digested R biotype (R2, mutant) yielded a shorter 157-bp fragment, and the heterozygote (SR) yielded both fragments. This analysis confirmed the presence of the two allele types in the SR hybrid (Figure 4). Direct sequencing of the PCR fragments also revealed that the heterozygote contains both allele types (data not shown).
View larger version (115K):
[in this window]
[in a new window]
|
Figure 4.
Drp Allele Identification by PCR–Primer-Introduced Restriction Analysis.
PCR was performed as given in the Methods. Lane 1, 25-bp DNA ladder; lane 2, S3 (wild type) digested with ClaI; lane 3, R2 (mutant) digested with ClaI; lane 4, SR (heterozygote) digested with ClaI; and lane 5, SR (heterozygote) undigested. Molecular lengths of expected fragments are indicated at right.
|
|
Expression of TUA cDNAs
Using the 3' noncoding regions, we examined expression of TUA1, TUA2, and TUA3 by ribonuclease protection assays. TUA1 transcripts were present at high levels in roots and inflorescences and absent in leaves. In contrast, TUA2 transcripts predominated in leaves, whereas TUA3 transcripts were present mainly in roots but at a lower level than TUA1 (Figure 5). TUA1 transcripts also predominated in roots in RNA gel blot analysis (data not shown). The predominant expression of TUA1 in roots and inflorescences may be due to high rates of transcription and/or stability of TUA1 mRNA in root tips and young inflorescences. These expression patterns agree with those previously described for -tubulin genes in maize (Montoliu et al. 1989 ). Maize tubulin genes are primarily expressed in meristematic tissues, such as root tips and coleoptiles, suggesting that their expression is correlated with cell division. Expression of these genes is reduced in mature leaves and root cortex.
View larger version (71K):
[in this window]
[in a new window]
|
Figure 5.
Expression of TUA1, TUA2, and TUA3 in Leaves, Roots, and Inflorescences.
Poly(A)+ RNA was hybridized with the 3' noncoding regions of TUA1, TUA2, and TUA3 and digested with ribonucleases. Exposure time for TUA1 and the control was half that of TUA2 and TUA3. R, root; L, leaf; I, inflorescence.
|
|
Characterization of Goosegrass -Tubulins
The predicted Mr and pI of the deduced -tubulins are 49,731 and 4.9 for TUA1, 49,823 and 5.1 for TUA2, and 49,613 and 4.9 for TUA3, respectively (http://expasy. hcuge.ch/ch2d/pi_tool.html). The sequence identities between TUA1 and ELS15, TUA2, and TUA3 were 100, 88, and 91%, respectively (alignments not shown). (There are four decoding errors in the sequences of ELS15 and ELR2: Val-202, Asn-249, Thr-257, and Ser-325 [ Cronin et al. 1993 ]. Alignments were performed after our corrections.)
Sequence comparisons of goosegrass TUA1 and other plant, protozoan, and mammalian -tubulins are shown in Figure 6. The sequence identities between goosegrass TUA1 and 10 other -tubulins range from 100 to 84%.
View larger version (150K):
[in this window]
[in a new window]
|
Figure 6.
Comparisons of the Amino Acid Sequences of the -Tubulins.
Shown are the sequences for E. indica (Ei), Zea mays (Zm), Hordeum vulgare (Hv), Arabidopsis thaliana (At), Prunus amygdalus (Pa), Pisum sativum (Ps), L. donovani (Ld), Trypanosoma cruzi (Tc), Mus musculus (Mm), Sus scrofa (Ss), and Homo sapiens (Hs); the accession numbers are AF008120, P14641, Y08490, P29511, P33629, U12589, U09612, M97956, P05213, P02550, and P04687, respectively. Sequence alignments were performed with PIMA maximal linkage clustering (http://dot.imgen.bcm.tmc.edu:9331/multi-align/multi-align.html) and viewed with BOXSHADE (http://ulrec3.unil.ch/software/BOX_form.html). Regions of identity (black) and similarity (shaded) are indicated. Asterisks and dots indicate the amino acids cited in the text.
|
|
Goosegrass -tubulins possess a number of regions in which amino acid sequences are conserved among a wide range of eukaryotes. Amino acid positions 95 to 102, 254 to 267, and 401 to 418 are believed to be involved in basic tubulin functions such as polymerization or dimer interaction and are highly conserved across broad evolutionary taxa (Figure 6) (Silflow et al. 1987 ). A putative trypsin cleavage site, Arg-339, and GTP binding regions, residues 142 to 148 (phosphoryl binding), 180 to 183 (ribose binding), and 242 to 246 (base binding), were also present in goosegrass -tubulins (Figure 1 Figure 2 Figure 3 and Figure 6). Although Ser-180 is present in all other plant -tubulins, including TUA1 (Figure 6), Ala was present at this position in TUA2 and TUA3 of goosegrass (Figure 2 and Figure 3). Because TUA1 appears to be the predominant tubulin transcript in goosegrass (Figure 5), the presence of a ribose binding site in TUA1 alone could be sufficient for microtubule polymerization in goosegrass. The Gly-rich cluster (residues 142 to 148), which is believed to form a loop between a region of ß-sheet (residues 135 to 139) and a region of -helix (residues 152 to 157) (Silflow et al. 1987 ), was also present in goosegrass (Figure 1 Figure 2 Figure 3 and Figure 6). All of the predicted proteins contain a C-terminal tyrosine residue, indicating that they may be subject to the post-translational detyrosination/tyrosination cycle described in animal systems (Raybin and Flavin 1977 ). Lys-40, known to be a substrate for acetylation (LeDizet and Piperno 1987 ), was present in goosegrass TUA1 and all of the other -tubulins (Figure 6) but was replaced by Thr in goosegrass TUA2 and TUA3. This post-translational modification may not be necessary for microtubule stability, or acetylation of TUA1 alone may be sufficient for goosegrass microtubule stability.
The amino acid substitutions of Ile for Thr-239 in tua1 of the R biotype and of Ala for Thr-340 in tua2 of the R biotype increase hydrophobicity (amino- and carboxy-ranges of nine) (Kyte and Doolittle 1982 ) from 187 to 239 in TUA1R1 and TUA1R2 (Figure 7A) and from -34 to -12 in TUA2R1 and TUA2R2 (data not shown). As a result, these alterations may change the dinitroaniline binding property of the proteins. In contrast, substitution of Met-268 by Thr in TUA1I1 decreases hydrophobicity from 140 to 114 (Figure 7A). Thus, the resistance mechanism of the I biotype may differ from that of the R biotype, as previously suggested by ultrastructural analysis (Vaughn et al. 1990 ). The predicted secondary structure (Garnier et al. 1978 ) around the Thr-239 mutation site changes from a coil to a ß-sheet in mutant tua1 protein of the R biotype. Yet, the mutations encoded by the tua1 gene of the I biotype and the tua2 gene of the R biotype are not predicted to change the secondary structure from a ß-sheet. These results also implicate TUA1 as the Drp locus.
View larger version (37K):
[in this window]
[in a new window]
|
Figure 7.
Hydropathy Index Plots of Goosegrass EiTUA1, Leishmania LdTUA, and Human HsTUA1.
(A) Goosegrass EiTUA1. Amino acid positions 239 and 268 are marked by asterisks.
(B) Leishmania LdTUA.
(C) Human HsTUA1. Amino acid positions 141 and 194 are marked by pluses.
|
|
The hydropathy index plot of a dinitroaniline-susceptible Leishmania -tubulin (Figure 7B) exhibited a similar pattern to that of goosegrass. On the other hand, the hydropathy index plot of a dinitroaniline-resistant human -tubulin (Figure 7C) exhibited a slightly different pattern compared with those of goosegrass and Leishmania.
|
|
DISCUSSION |
|---|
The work reported here identified an -tubulin gene that correlates with dinitroaniline resistance in goosegrass. We conclude that TUA1 is the Drp gene and that a missense mutation, Thr-239 to Ile or Met-268 to Thr, may confer dinitroaniline resistance.
Microtubules are assembled from heterodimer subunits consisting of an - and ß-tubulin, each with an Mr of ~55,000 (Amos 1979 ). Both - and ß-tubulins are organized into two compact structural domains that are closely associated to form a globular structure, from which a short linear tail projects. The N terminus is found in the large globular domain, whereas the C terminus is found in the small domain (Sackett and Wolff 1986 ). Tubulin heterodimer formation occurs as a result of a bond between the large domain of -tubulin and the small domain of ß-tubulin, whereas polymerization of the heterodimer occurs between the large domain of ß-tubulin and the small domain of -tubulin (Kirchner and Mandelkow 1985 ).
Colchicine, an antimitotic poison, binds to tubulin and induces the unfolding of the C-terminal region of ß-tubulin (Sackett and Varma 1993 ). The tubulin–colchicine complex can bind to the growing end of microtubules and impedes subsequent addition of dimers (Sackett and Varma 1993 ). Dinitroaniline also binds to tubulin heterodimers and disrupts microtubule polymerization (Morejohn et al. 1987 ; Hugdahl and Morejohn 1993 ). If the tubulin–colchicine model is applicable, and because the -subunit of a heterodimer may be modified by dinitroaniline binding in the S biotype of goosegrass, the tubulin–dinitroaniline complex may not be able to bind to the growing ends of the microtubules. However, in the R and I biotypes of goosegrass, dinitroaniline may not bind to the -subunit of a heterodimer. Another less likely possibility is that binding of dinitroaniline to the -subunit may not modify the conformation of the heterodimer significantly and thus not affect binding to the growing ends of the microtubules.
In colchicine/dinitroaniline cross-resistant Chlamydomonas, much more colchicine than dinitroaniline was required to produce antimicrotubule effects (Schibler and Huang 1991 ). These mutants also exhibited supersensitivity to taxol. However, there was no evidence for differential sensitivity to taxol between the S and R+ biotypes of goosegrass (Cronin et al. 1993 ). Moreover, in ß-tubulins of the R biotype, Lys-350 was not mutated to any other amino acid (data not shown). Thus, the molecular basis for dinitroaniline resistance in goosegrass differs from that in Chlamydomonas.
Currently, there is no evidence for functionally distinct tubulin isoforms in plants. Expression of goosegrass TUA1 is higher than that of TUA2 and TUA3. Expression of TUA1 predominates in roots, whereas TUA2 is undetectable and TUA3 is only moderately expressed (Figure 5). Dinitroanilines are used as preemergent, soil-incorporated herbicides. Thus, the cellular damage or growth inhibition in root tips caused by dinitroaniline may be mitigated only by the presence of resistant isoform(s) of tua1.
-Tubulins are highly identical among different species. However, plant -tubulins contain nonconservative substitutions at amino acid positions 38, 139, 295, and 303, which may reflect "plant-specific" sequences that determine unique biochemical characteristics of plant tubulins (Silflow et al. 1987 ). Monocotyledons contain two unique amino acids, Asn-253 and Val-385. Also, monocotyledonous and dicotyledonous tubulins cluster separately in the dendrogram (data not shown), consistent with their phylogenetic relatedness. -Tubulins of parasitic protozoa, a trypanosome and Leishmania, are more similar to those of plants than to those of mammals. The predicted hydropathy index plot of Leishmania -tubulin is almost identical to that of goosegrass (Figure 7A and Figure 7B). There are two locations where goosegrass and Leishmania -tubulins differ significantly from those of mammals. Val-141 of goosegrass and Leishmania is replaced by Phe in mammals (Figure 6). This amino acid change reduces hydrophobicity from 159 in goosegrass and 143 in Leishmania to 109 in mammals (Figure 7), and the predicted secondary structure changes from a coil to a turn. Also, Leu-194 of goosegrass and Leishmania is replaced by Thr in mammals (Figure 6). This amino acid change reduces hydrophobicity from 76 in goosegrass and 41 in Leishmania to -15 in mammals (Figure 7), and the secondary structure changes from an -helix to a coil. Met-268 in plants (the mutation site of the I biotype) is replaced by Pro in mammals (Figure 6). Pro introduces a bend in a polypeptide chain. These unique characteristics of mammalian -tubulins could explain why dinitroaniline binds to protozoan and plant tubulins but not to mammalian tubulins (Morejohn et al. 1987 ; Chan and Fong 1990 ; Chan et al. 1993 ). Such characteristics observed in mammalian -tubulins were not detected in ß-tubulins (data not shown).
Neither mutation identified in goosegrass TUA1 corresponds to the sites in mammalian -tubulins discussed above. This could explain why the resistant varieties of goosegrass are still more susceptible to dinitroaniline than are mammalian species. It may be informative to determine whether mammalian -tubulin can confer dinitroaniline resistance in transgenic plants.
|
|
METHODS |
|---|
Plant Material
Goosegrass (Eleusine indica) seed (resistant [R] biotypes from Cherokee County, AL [R1], and Marlboro County, SC [R2]; susceptible [S] biotypes from Gibson County, TN [S1], Anderson County, SC [S2], and Orangeburg County, SC [S3]; and intermediately resistant [I] biotype from Florence County, SC [I1]) were germinated at 37°C and grown in the greenhouse for 21 to 28 days. Seed were derived from the second or third generation of inbred lines. F1 heterozygote plants (SR) were generated by crossing R2 and S3 (Zeng and Baird 1997 ), and F2 and F3 individuals of known phenotype and genotype were used. Response phenotypes are defined by the concentration of dinitroaniline (i.e., oryzalin) at which root length is reduced by 50% (Zeng 1997 ).
RNA and DNA Isolation
RNA was isolated from leaves (all biotypes), roots (R1, R2, S1, and S2), and inflorescences (I1) (John 1992 ); poly(A)+ RNA was isolated from leaves of R1, R2, S1, and S2, roots of S1, and inflorescences of I1 by using an Oligotex mRNA kit (Qiagen, Chatsworth, CA) and a FastTrack 2.0 kit (Invitrogen, Carlsbad, CA). Genomic DNA was isolated from S3, R2, and SR (Dellaporta et al. 1984 ).
cDNA Isolation by Reverse Transcriptase–Polymerase Chain Reaction
Nested degenerate oligonucleotides for -tubulins (1TA5, 5'-ATCCACATCGGYCAGGCYGG-3'; 3TA5, 5'-GCMAAYGCSTGCTGGGAGCT-3'; and 2TA3, 5'-CCGACYTCYTCATARTCCT TCTC-3') were synthesized based on the sequences of corresponding genes from seven other species (rice, maize, Arabidopsis thaliana, pea, almond, Chlamydomonas reinhardtii, and Volvox carteri; accession numbers Z11931, U05258, M84699, U12589, X67162, M11447, and X12846, respectively), biasing for monocotyledon sequences. cDNA synthesis (reverse transcription) and subsequent DNA amplification (polymerase chain reaction [PCR]) were performed by one-tube reverse transcriptase (RT)–PCR method, using a GeneAmp RNA PCR kit (Perkin-Elmer, Foster City, CA) and following the manufacturer's protocol with minor modifications. Two sets of primers were used for each biotype: 1TA5 x 2TA3 and 3TA5 x 2TA3. Derived PCR fragments were gel purified by Wizard PCR Preps DNA purification system (Promega) and ligated into pGEM-T (Promega). Plasmid DNA was isolated by Wizard Minipreps (Promega), and both strands were sequenced using ABI PRISM Dye Primer or Terminator Cycle Sequencing kits and an ABI 373 automated sequencer (Perkin-Elmer). PCR and sequencing primers were synthesized by Integrated DNA Technologies (Coralville, IA). Sequence analysis was performed using GeneWorks (IntelliGenetics, Campbell, CA), and homology searches were performed using the BLAST server (http:www.ncbi.nlm.nih.gov/BLAST/nph-blast). Missing isotype cDNAs were isolated by synthesizing isotype (or locus)-specific PCR primers (TUA1, [aIF1] 5'-CATCCAGGCTGACGGTCAGATG-3', [aIR2] 5'-AGCAAGGTCCTCACGCGCCTCG-3', and [aIR22] 5'-ACACCTCGACGACACTGG-TGGA-3'; TUA2, [aIIF3] 5'-ATCCAGCCGGATGGCCTCAT-3', [aIIR4] 5'-AGCCAGATCCTCCCT TGCT TCT-3', and [aIIR44] 5'-TGAGAAGACCTCTGCCACGGCT-3'; and TUA3, [aIIIF5] 5'-CCTCGGTTGGCGTCGCACAT-3' and [aIIIR6] 5'-GCCAAGTCCTCACGGGCTTCT-3') and again performing RT-PCR. Also, putative nucleotide substitutions (mutations) or PCR-related cloning errors were confirmed or corrected, respectively, by isotype-specific RT-PCR. All RT-PCRs were performed starting with RNA. The sequences of missense mutations were confirmed by direct sequencing of both strands.
Rapid Amplification of cDNA Ends
The missing 5' and 3' ends of the cDNAs were isolated by the rapid amplification of cDNA ends (RACE) technique (Frohman et al. 1988 ), using a Marathon cDNA synthesis kit (Clontech, Palo Alto, CA) and following the manufacturer's protocol. Gene-specific primers for 3' RACE were (TUA1, [aIGF1] 5'-GTGCCTTGACCGCATCAGGAAGCTTGCC-3'; TUA2, [aIIGF3] 5'-CAGGTGCAGGGAAGCACGT TCCAA-GGGC-3'; and TUA3, [aIIIGF5] 5'-TGTGCCTGGACCGTGTCCGCAAGTTGG-3'), and gene-specific primers for 5' RACE were (TUA1, [aIGR2] 5'-GAACGCGCTGT TGGTGATCTCAGCCACG-3'; TUA2, [aIIGR4] 5'-GACCTCTGCCACGGCTGTGT TGT TGCTGA-3'; and TUA3, [aIIIGR6] 5'-CACAGCACGCTGGACCT T TGCCAGGTCA-3').
Drp Allele Identification by PCR–Primer-Introduced Restriction Analysis
To distinguish Drp locus alleles of TUA1, PCR–primer-introduced restriction analysis (PIRA) was performed (Jacobson and Muskovits 1991 ). A forward primer for PCR (5'-GCTTGTTTCTCAGGTCATTTCATCATCGA-3') was synthesized from the sequence of TUA1 at positions 782 to 810 of TUA1S1 (Figure 1); a reverse primer (5'-TCAGCCACGGACAGCTGCTC-3') was synthesized from 182 bases downstream of the forward primer. Two bases, TC, in the forward primer (underlined) were modified from CT in the original sequence to create a ClaI restriction site (ATCGAT) for mutated alleles. Genomic DNA (100 ng) from S3, R2, and SR plants was used as a PCR template. PCR conditions were denaturation at 95°C for 2 min; five cycles of 95°C for 30 sec, 60°C for 30 sec, and 72°C for 30 sec; 30 cycles of 95°C for 30 sec, 58°C for 30 sec, and 72°C for 30 sec; and extension at 72°C for 3 min. DNA fragments of the expected size were agarose gel purified and resuspended in 50 µL of H2O. One microliter of the resuspended DNA was used for the second PCR. PCR conditions were the same as those of the first reaction, except that the denaturing time was reduced to 20 sec. PCR products were digested with ClaI and fractionated on a 4% NuSieve GTG agarose gel (FMC, Rockland, ME).
Ribonuclease Protection Assays
Ribonuclease protection assays were performed using an RPA II kit (Ambion, Austin, TX) and following the manufacturer's protocol. The 3' noncoding regions of TUA1, TUA2, TUA3, and a partial coding region of TUA3 (between aIII9' and 2TA3 primers, used as control) were PCR amplified and cloned into pGEM-T Easy (Promega). PCR primers are as follows: TUA1, (TUA1F) 5'-CGAGTACTAGATGAATCTACCGCTTC-3' and (TUA1R) 5'-TGATAACCCAGTAGCAAAGCGG-3'; TUA2, (TUA2F) 5'-AGTACTCTGATCTGGATGCATGGT-3' and (TUA2R) 5'-GAGCACAACCATCATGCCAG-3'; and TUA3, (TUA3F) 5'-GGGAGAAGACTACTGAGTAGCTGGTT-3', (TUA-3R) 5'-TGAACAATCATGACCAACCGG-3', and (aIII9') 5'-CATCAAGACGAAGAGGACTGTC-3'. Antisense probes were synthesized by in vitro transcription using MAXIscript kit (Ambion) and following the manufacturer's protocol. The labeled probes were excised from denaturing polyacrylamide gels (Novex, San Diego, CA) and eluted. The probes (4 x 104 cpm for TUA1, TUA2, and TUA3; 104 cpm for the control) were hybridized with poly(A)+ RNA (1.2 µg for TUA1, TUA2, and TUA3; 0.3 µg for the control) at 42°C for 14 hr. Hybridized TUA1, TUA2, and TUA3 were digested with ribonucleases and fractionated on denaturing polyacrylamide gels, and the hybridized control was fractionated on a nondenaturing polyacrylamide gel. Gels were exposed to x-ray films at -70°C.
|
|
FOOTNOTES |
|---|
1 Current address: U.S. Department of Agriculture Salinity Laboratory, Riverside, CA 92507. 
|
|
ACKNOWLEDGMENTS |
|---|
We thank Drs. Beth Krizek (University of South Carolina, Columbia) and Brenda Shirley (Virginia Polytechnic Institute, Blacksburg) for technical advice. This project was supported by grants from the U.S. Department of Agriculture (No. 95-37315-2152) and the South Carolina Agricultural Experiment Station (Technical Contribution No. 4366).
Received July 21, 1997; accepted December 8, 1997.
|
|
REFERENCES |
|---|
Amos, L.A. (1979). Structure of microtubules. In Microtubules, K. Roberts and J.S. Hyams, eds (New York: Academic Press), pp. 1–64.
Brunke, K.J., Anthony, J.G., Sternberg, E.J., and Weeks, D.P. (1984) Repeated consensus sequence and pseudopromoters in the four coordinately regulated tubulin genes of Chlamydomonas reinhardtii.. Mol. Cell. Biol. 4:1115-1124
[Abstract/Free Full Text]
Chan, M.M., and Fong, D. (1990) Inhibition of leishmanias but not host macrophages by the antimicrotubulin herbicide trifluralin. Science 249:924-926
[Abstract/Free Full Text]
Chan, M.M., Grogl, M., Chen, C.-C., Bienen, E.J., and Fong, D. (1993) Herbicides to curb human parasitic infections: In vitro and in vivo effects of trifluralin on the trypanosomatid protozoans. Proc. Natl. Acad. Sci. USA 90:5657-5661
[Abstract/Free Full Text]
Cronin, K.E., Hussey, P.J., Ray, J.A., and Waldin, T.R. (1993). Herbicide resistant plants. World Intellectual Property Organization International Publication No. WO 93/24637.
Dellaporta, S.L., Wood, J., and Hicks, J.B. (1984). Maize DNA miniprep. In Molecular Biology of Plants: A Laboratory Manual, R. Malmberg, J. Messing, and I. Sussex, eds (Plainview, NY: Cold Spring Harbor Laboratory Press), pp. 36–37.
Frohman, M.A., Dush, M.K., and Martin, G.R. (1988) Rapid production of full-length cDNA from rare transcripts: Amplification using a single-gene specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85:8998-9002
[Abstract/Free Full Text]
Garnier, J., Osguthorpe, D.J., and Robson, B. (1978) Analysis of the accuracy and implications of simple methods for predicting the secondary structure of globular proteins. J. Mol. Biol. 120:97-120
[CrossRef][Web of Science][Medline]
Hacskayko, J., and Amato, V.A. (1968) Effect of trifluralin on roots of corn and cotton. Weed Sci. 16:513-515
Hugdahl, J.D., and Morejohn, L.C. (1993) Rapid and reversible high-affinity binding of the dinitroaniline herbicide oryzalin to tubulin from Zea mays L. Plant Physiol. 102:725-740
[Abstract]
Jacobson, D.R., and Muskovits, T. (1991) Rapid, nonradioactive screening for activating ras oncogene mutations using PCR primer introduced restriction analysis (PCR-PIRA). PCR Methods Appl. 1:146-148
[Medline]
James, S.W., Silflow, C.D., Stroom, P., and Lefebvre, P.A. (1993) A mutation in the 1-tubulin gene of Chlamydomonas reinhardtii confers resistance to antimicrotubule herbicides. J. Cell Sci. 106:209-218
[Abstract]
John, M. (1992) An efficient method for isolation of RNA and DNA from plants containing polyphenolics. Nucleic Acids Res. 20:2381
[Free Full Text]
Kirchner, K., and Mandelkow, E.-M. (1985) Tubulin domains responsible for assembly of dimers and protofilaments. EMBO J. 4:2397-2402
[Web of Science][Medline]
Kyte, J., and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132
[CrossRef][Web of Science][Medline]
LeDizet, M., and Piperno, G. (1987) Identification of an acetylation site of Chlamydomonas -tubulin. Proc. Natl. Acad. Sci. USA 84:5720-5724
[Abstract/Free Full Text]
Lee, V.D., and Huang, B. (1990) Missense mutations at lysine 350 in ß2-tubulin confer altered sensitivity to microtubule inhibitors in Chlamydomonas.. Plant Cell 2:1051-1057
[Abstract/Free Full Text]
Montoliu, L., Rigau, J., and Puigdoménech, P. (1989) A tandem of -tubulin genes preferentially expressed in radicular tissues from Zea mays.. Plant Mol. Biol. 14:1-15
Morejohn, L.C., Bureau, T., Molé-Bajer, J., Bajer, A.S., and Fosket, D.E. (1987) Oryzalin, a dinitroaniline herbicide, binds to plant tubulin and inhibits microtubule polymerization in vitro.. Planta 172:252-264
[CrossRef][Web of Science]
Mudge, L.C., Gossett, B.J., and Murphy, T.R. (1984) Resistance of goosegrass (Eleusine indica) to dinitroaniline herbicides. Weed Sci. 32:591-594
Mysore, K.K., and Baird, W.V. (1995) Molecular characterization of the tubulin-related gene families in herbicide resistant and susceptible goosegrass (Eleusine indica). Weed Sci. 43:28-33
Quader, H., and Filner, P. (1980) The action of antimitotic herbicides on flagellar regeneration in Chlamydomonas reinhardtii: A comparison with the action of colchicine. Eur. J. Cell Biol. 21:301-304
[Web of Science][Medline]
Raybin, D., and Flavin, M. (1977) Modification of tubulin by tyrosylation in cells and extracts and its effect on assembly in vitro.. J. Cell Biol. 73:492-504
[Abstract/Free Full Text]
Sackett, D.L., and Varma, J.K. (1993) Molecular mechanism of colchicine action: Induced local unfolding of ß-tubulin. Biochemistry 32:13560-13565
[CrossRef][Medline]
Sackett, D.L., and Wolff, J. (1986) Proteolysis of tubulin and the substructure of the tubulin dimer. J. Biol. Chem. 261:9070-9076
[Abstract/Free Full Text]
Schibler, M.J., and Huang, B. (1991) The colR4 and colR15 ß-tubulin mutations in Chlamydomonas reinhardtii confer altered sensitivities to microtubule inhibitors and herbicides by enhancing microtubule stability. J. Cell Biol. 113:605-614
[Abstract/Free Full Text]
Silflow, C.D., Oppenheimer, D.G., Kopczak, S.D., Ploense, S.E., Ludwig, S.R., Haas, N., and Snustad, D.P. (1987) Plant tubulin genes: Structure and different expression during development. Dev. Genet. 8:435-460
[CrossRef]
Vaughn, K.C. (1986) Cytological studies of dinitroaniline-resistant Eleusine.. Pestic. Biochem. Physiol. 26:66-74
Vaughn, K.C., Vaughan, M.A., and Gossett, B.J. (1990) A biotype of goosegrass (Eleusine indica) with an intermediate level of dinitroaniline herbicide resistance. Weed Technol. 4:157-162
Villemur, R., Joyce, C.M., Haas, N.A., Goddard, R.H., Kopczak, S.D., Hussey, P.J., Snustad, D.P., and Silflow, C.D. (1992) -Tubulin gene family of maize (Zea mays L.): Evidence for two ancient -tubulin genes in plants. J. Mol. Biol. 227:81-96
[CrossRef][Web of Science][Medline]
Waldin, T.R., Ellis, J.R., and Hussey, P.J. (1992) Tubulin-isotype analysis of two grass species resistant to dinitroaniline herbicides. Planta 188:258-264
Zeng, L. (1997). Genetic and Molecular Characterization of Dinitroaniline Herbicide Resistance in Goosegrass (Eleusine indica). PhD Dissertation (Clemson, SC: Clemson University).
Zeng, L., and Baird, W.V. (1997) Genetic basis of dinitroaniline herbicide resistance in a highly-resistant biotype of goosegrass (Eleusine indica). J. Hered. 88:427-432
[Abstract/Free Full Text]
ABSTRACT
INTRODUCTION
