Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by He, X.
Right arrow Articles by Smith, R. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by He, X.
Right arrow Articles by Smith, R. L.
Agricola
Right arrow Articles by He, X.
Right arrow Articles by Smith, R. L.
Related Collections
Right arrow Maize
Right arrow Other Forage Crops
Right arrow Cell Biology & Molecular Genetics
Published in Crop Sci. 43:2240-2251 (2003).
© 2003 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

GENOMICS, MOLECULAR GENETICS & BIOTECHNOLOGY

Improvement of Forage Quality by Downregulation of Maize O-Methyltransferase

Xu Hea, Mary Beth Hallb, Maria Gallo-Meagherc and Rex L. Smith*,c

a Dep. of Biological Sciences, Simon Fraser Univ., Burnaby, B.C. V5A 1S6, Canada
b Animal Sciences, Univ. of Florida, Gainesville, FL
c Agronomy Dep., Univ. of Florida, Gainesville, FL 32611

* Corresponding author (rls{at}mail.ifas.ufl.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Lignin is a complex, aromatic polymer that limits plant cell wall degradation by ruminants and reduces the nutritional value of forages. Genetic engineering, using an antisense strategy, offers the potential to modulate enzymes in the lignin biosynthetic pathway as a way to reduce lignin, thereby improving forage quality and animal performance. We investigated the effectiveness of expressing antisense sorghum O-methyltransferase gene (omt) to downregulate maize OMT and reduce lignin. Constructs contained a sorghum omt coding region in the antisense orientation driven by the maize ubiquitin-1 (Ubi) promoter (with the first intron and exon) along with bar, that confers glufosinate herbicide resistance, driven by the CaMV 35S promoter. Twenty-eight T0 plants regenerated from 17 herbicide-resistant callus lines from 13 independent bombardments expressed the brown midrib phenotype. O-methyltransferase activity was significantly lower in T1 transgenics compared with controls, with some plants showing a 60% reduction. Those T1 transgenics with downregulated OMT averaged 20% less lignin in stems and 12% less lignin in leaves compared with controls. On a whole-plant basis, lignin was reduced by an average of 17% with the greatest reduction being 31%. Digestibility was significantly improved in transgenic plants by 2% in leaves and 7% in stems. Mean whole-plant digestibility increased from 72 to 76%. This research demonstrates that genetic engineering has the potential to improve forage grass digestibility. This could be important, especially in tropical forage species, which generally have lower quality than temperate species.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
LIGNIN IS A COMPLEX, phenolic polymer found in plant cell walls and is essential for mechanical support, water and mineral transport, and defense in vascular terrestrial plants. However, lignin limits the utilization of plant biomass, making paper pulping difficult and reducing forage digestibility. Lignin has long been recognized as a major "antiquality" factor in forage grasses, especially in tropical and subtropical grasses, which generally have lower digestibility than temperate grasses, and often lack sufficient quality for adequate animal performance. Cloning of lignin biosynthetic pathway genes has offered opportunity to reduce lignin content and/or modify its composition to improve utilization of the plant cell wall through genetic engineering approaches. Of the genes targeted for downregulation of lignin, the caffeic acid 3-O-methyltransferase (OMT) gene has been favored because the maize (Zea mays L.) natural brown midrib bm3 mutants, which are a result of either a deletion or an insertion in the OMT gene (omt), have reduced lignin content and altered lignin composition (Vignols et al., 1995). In addition to maize, other mutants characterized by a brown midrib phenotype with reduced lignin content and/or altered lignin composition have been identified in sorghum [Sorghum bicolor (L.) Moench] and pearl millet [Pennisetum glaucum (L.) R. Br.]. Utilization of these brown midrib mutants has been considered an efficient way to improve forage quality (Cherney et al., 1991; Barriere and Argiller, 1993).

Reports have focused on OMT downregulation in either transgenic tobacco (Nicotiana tabacum L.) (Ni et al., 1994; Vailhe et al., 1996; Sewalt et al., 1997) or poplar (Populus tremuloides Michaux.) (Dwivedi et al., 1994; Atanassova et al., 1995; Doorsselaere et al., 1995) where the endogenous OMT activity level was reduced from 30% to more than 90%. In general, lignin chemical composition [the syringyl (S)/guaiacyl (G) ratio] was significantly changed while lignin content was not significantly reduced (Dwivedi et al., 1994; Atanassova et al., 1995; Doorsselaere et al., 1995). Only Ni et al. (1994) and Sewalt et al. (1997) reported a moderate reduction of lignin content in transgenic tobacco plants.

In this research, we wanted to determine whether antisense strategy would downregulate OMT activity, decrease lignin content, and increase forage digestibility to improve monocot forage quality. We selected ‘Hi-II’ maize (Armstrong et al., 1991), as a model monocot system, and investigated the effects of transgenic sorghum antisense omt expression on the downregulation of endogenous maize OMT. We also measured the effects of downregulated OMT on reducing lignin content and increasing forage digestibility in transgenic maize.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Material and Culture Methods
Hi-II maize inbred lines "A" and "B," obtained from the Maize Genetics Cooperation Stock Center (University of Illinois at Urbana-Champaign, Urbana IL 61801-4798), were intercrossed to produce the hybrid Hi-II plants used for tissue culture. Callus was initiated from immature tassels isolated 4 to 6 wk after planting by aseptically recovering immature tassels less than 4.0 cm long, cutting them into 1.0- to 1.5-mm segments and culturing those explants on initiation medium (Songstad et al., 1992). Once the type-II callus was formed, it was transferred to the maintenance medium described by Armstrong (1994). Callus initiation and maintenance cultures were kept in the dark at 28°C and transferred to fresh medium every 14 d.

Vector Construction and Plant Transformation
Two omt antisense vectors were constructed, using either a full-length sorghum omt cDNA (isolated by R.L. Smith and J.C. Seib, GenBank accession no. AF387790) or a 995-base pair (bp) 5'-fragment, that were assembled in the antisense direction driven by the maize ubiquitin-1 (Ubi-1) promoter (plus the first intron and exon) and terminated by the nos terminator from pAHC17 (Christensen et al., 1992). Those vectors also contained bar and nos terminator from pAHC20 (Christensen et al., 1992). The bar gene was driven by the CaMV 35S promoter from pBI121 (CloneTech Laboratories, Palo Alto, CA) (See Fig. 1). The coding region of the sorghum omt has 89% homology to the maize omt coding region.



View larger version (9K):
[in this window]
[in a new window]
  		Fig. 1. Schematic representation of an antisense sorghum omt vector used for maize transformation. The vector contained the CaMV 35S promoter to drive bar with the nos terminator; and the maize Ubi-1 promoter (plus first intron and first exon) to drive the sorghum omt full-length cDNA (1600 bp), or the 995-bp 5'-fragment, in the antisense orientation with the nos terminator. Restriction sites labeled are; H = HindIII, B = BamHI, and E = EcoRI. Digestion with EcoRI releases a 2.5-kb fragment with the full-length omt or a 1.9-kb fragment with the 995-bp omt vectors.

 
Plasmid DNA used for bombardment was purified with the Qiagen Plasmid Maxi Kit (Qiagen Inc., Valencia, CA) according to Qiagen's instructions and precipitated onto 1.0-µm gold particles (Bio-Rad Laboratories, Hercules, CA) by means of the protocol described by Lemaux et al. (1990). Bombardment of callus was performed with a Biolistic PDS-1000/He device (Bio-Rad Laboratories, Hercules, CA) using the following conditions: 4.5 MPa (650 psi), 5 to 8 µg DNA per shot, one or two shots per plate, osmotic treatment (0.4 M sorbitol added to maintenance medium) of callus 4-h prebombardment and 16-h postbombardment (Vain et al., 1993). Following the postbombardment osmotic treatment, callus was initially transferred to maintenance medium for two to six days (until callus growth resumed), and then transferred to selection medium containing 3 mg L-1 glufosinate. During selection, callus was cultured at 28°C in the dark, subcultured every 14 d and after 6 to 8 wk of selection, resistant calli could be distinguished.

Plants were regenerated according to the three-step protocol (R1, R2 and R3 media) described by Armstrong (1994) on 4-6 mg L-1 glufosinate. Rooted plants were transplanted to pots containing Metromix-220 (Scotts-Sierra Horticultural Products Company, Marysville, OH), conditioned several days and transferred to an air conditioned greenhouse.

OMT Activity Assays
Leaf or stem tissue was ground into a fine powder with a mortar and pestle in the presence of liquid N2 and crude protein was extracted as described by Bugos et al. (1991). The OMT activity assay was a modification based on Bugos et al. (1991), Collendavelloo et al. (1981), and Gowri et al. (1991). Diluted tritiated S-adenosylmethionine (SAM) was prepared by adding 20 µL of tritiated adenosyl-L-[methyl-3H] methionine (specific activity 76.0 Ci/mmol, 1 µCi µL-1, Amersham) to 280 µL of an unlabeled SAM.HCl solution (100 nmol). The OMT assay conditions were optimized and a standard OMT assay was established as follows: 50 µL of plant extract, 10 µL of caffeic acid (final concentration 1 mM), 137 µL of 200 mM Tris-HCl pH 8.0 buffer. After incubation at 37°C for 5 min, 3 µL of the diluted tritiated SAM (0.2 µCi nmol-1) was added. After 2 h of incubation at 37°C, 20 µL of 2 M HCl was added to stop the reaction. The product was extracted by adding 1 mL of a mixture of hexane:ethyl acetate (1:1), vortexing for 1 min, then recovering the organic phase after centrifugation for 2 min. Seven milliliters of Scintilene (Fisher Scientific, Pittsburgh, PA) cocktail was added to each sample and the radioactivity of the sample was measured in a liquid scintillation counter. Each sample was assayed in duplicate, one without adding caffeic acid as a control for the assay.

Since OMT is developmentally and spatially regulated, preliminary experiments were conducted to develop a sampling strategy to standardize the OMT activity data. Those experiments determined that the second youngest leaf at both the 2- and 7-wk stage was suitable to standardize sampling. Protein content of samples used for the OMT assay was determined by the Bradford method (1976) using the reagent from Bio-Rad (Hercules, CA) with bovine serum albumin (BSA) as the protein standard.

Initiation of the T1 Progeny
The primary (T0) OMT transformants were highly variable in vigor and growth, making them unsuitable for OMT activity and other comparisons. To remedy this problem, T1 progeny were produced from seed from self-pollinated T0 plants. Two groups of T1 progeny were produced. The first group (the A-group) was initiated with five seeds from each of 33 selected, transgenic, self-pollinated T0 plants producing 33 families. These families were segregating so transgenic plants within those families were identified by herbicide resistance. The transgenic families were compared with a control group of 10 nontransgenic plants (from seed of different, self-pollinated, regenerated, plants). The OMT assays were performed at the 2- and 7-wk growth stages. Family members of this group are designated with numbers A1 through A5.

The second T1 progeny group (the B-group) was initiated from six seeds from 10 primary (T0) plants (selected from within the 33 primary plants mentioned above) along with control plants as described above. These T1 progeny plants were grown to a more mature stage and flag leaves were sampled for OMT and other comparisons. Also forage was harvested five days after anthesis. Family members of this group are designated with numbers B1 through B6. All plants were grown randomly arranged in an air conditioned greenhouse.

Southern and Northern Blot Analysis
Genomic DNA was extracted from 1 g leaf tissue by the Dellaporta et al. (1983) protocol and purified by chloroform-phenol extraction. Ten micrograms of genomic DNA was digested with EcoRI restriction enzyme and separated on a 0.8% (w/v) agarose gel. For copy number determination, a 1C value (haploid DNA content of maize) of 2.75 pg was used (Arumuganathan and Earle, 1991). Plasmid DNA was digested with the same restriction enzyme used for genomic DNA digestion and diluted to the required copy concentration. Southern blot analysis was performed according to Sambrook et al. (1989). Total RNA was isolated by means of Trizol reagent (GIBCO, BRL, Rockville, MD) according to the manufacturer's instructions. Ten micrograms of total RNA was size-fractionated on 1% (w/v) agarose formaldehyde gels. Northern blot analysis was performed as described by Sambrook et al. (1989). Strand-specific northern analyses were conducted with RNA probes (riboprobes) following the manufacturer's instruction (Promega, Madison, WI).

Cell Wall and Lignin Analyses
Cell wall analyses consisted of neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL). Neutral detergent fiber was determined by extracting dry, finely ground plant material with neutral detergent consisting of 30 g L-1 sodium lauryl sulfate, 18.6 g L-1 disodium EDTA, 6.8 g L-1 sodium borate dehydrate, 4.6 g L-1 sodium hydrogen phosphate, and 10.0 mL L-1 2-ethoxyethanol. A 0.5-g plant sample was extracted for 1 h at boiling temperature. Residue was recovered by filtration through glass wool and washed with hot water and acetone. Oven-dry and ash weights were determined. NDF is expressed as a percentage of dry weight.

The determination of acid detergent lignin (ADL), which is the crude determination of lignin, requires that acid detergent fiber (ADF), consisting mainly of cellulose and lignin, be isolated first. ADF was isolated by extracting 0.5 g tissue according to Goering and Van Soest (1970). This residue was further extracted with 72% (v/v) sulfuric acid for 3 h, recovered and washed with hot water to remove the acid. Dry and ash weights were determined and ADL lignin was calculated. ADL is also expressed as a percentage of dry weight.

Phloroglucinol-HCl and potassium permanganate-HCl staining of lignin were performed on 1-mm-thick stem slices as described by Dean (1997). Stem slices were obtained from the top region of the seventh and tenth internodes (from the top) of control and transgenic plants harvested 5 d after anthesis.

In Vitro Organic Matter Digestibility and Plant Yield
In vitro organic matter digestibility (IVOMD) was determined by a modified method of the Tilley and Terry two-stage method (Tilley and Terry, 1963; Moore and Mott, 1974). Plant samples of 0.5 g were incubated in media inoculated with bovine rumen fluid for 48 h at 39°C under anaerobic conditions, then were further digested in acid pepsin for 48 h. Undigested residue was recovered by filtration through glass wool, oven dried, and weighed. Organic matter of the original samples was determined by ashing the sample at 500°C. IVOMD is expressed as a percentage of dry organic matter. The yield (5 d after anthesis) of each transgenic and control plant was determined as oven-dried plant weight.

Statistical Analysis
Statistical analyses including analysis of variance, correlation, and covariance were performed comparing transgenic plants to control plants according to Lyman-Ott (1993) with SAS software version 6.12 (SAS Institute Inc., Cary, NC). To compare the overall difference of OMT activity between nontransgenic and transgenic plants the general linear model procedure was used for performing the analysis of variance. Fisher's least significant difference test was used to compare the differences among transgenic families and nontransgenic controls. Plant growing days and dry matter yields were included as covariates in the general linear model for comparing IVOMD, yield, ADL, and NDF means of 16 transgenic plants from seven transgenic families to 10 control plants.


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Brown Midrib Phenotype
The bm3 mutants of maize, which arise from a mutation occurring in the gene encoding OMT, display a severe reduction of OMT activity (Grand et al., 1985; Vignols et al., 1995; Morrow et al., 1997). Bm3 mutants have a brown midrib phenotype at the four- to six-leaf stage. This pigmentation disappears from the leaf midribs as plants grow. However, this pigmentation can be found in the rind and vascular tissues of stem cross-sections (Cherney et al., 1991; Barriere and Argiller, 1993). Among 248 OMT antisense T0 plants, 28 plants (regenerated from 17 herbicide-resistant callus lines from 13 separate bombardments) showed the brown midrib phenotype at the four- to six-leaf stage. This phenotype was not observed among the regenerated control plants. The brown midrib color was more obvious on the abaxial side than on the adaxial side of the leaf. Unlike the bm3 mutants, not all midribs turned brown at this stage; only midribs of two or three leaves of each plant turned brown. Additionally, distribution of the brown color was not homogeneous, even on a single midrib. When these plants were harvested at silking, some the internodes showed discoloration (data not shown). These data suggest different spatial and temporal expression of the antisense omt.

Downregulation of OMT Activity and Endogenous OMT mRNA in T1 Plants
To standardize sampling for downregulation of OMT in the transgenic plants, developmental and spatial distribution of OMT activity was examined at the 2- (data not shown) and 7-wk stages. Figure 2 shows the relative OMT activity and the relative OMT mRNA accumulation levels of different aged leaves (second to ninth youngest), internodes, and sheaths of 7-wk control plants. The second youngest leaf had the highest leaf OMT activity which decreased as the leaf age increased to the sixth youngest leaf where the activity stabilized at a low level. Of the internodes, the extending internode had maximum OMT activity. The sheaths also had high activity. The OMT activity was highly correlated with OMT mRNA accumulation (coefficient r = 0.96). Only one size of omt transcript was present among different plant parts at that sampling stage. In addition, no differential hybridization intensity was observed between sorghum and maize stem tissues (compare Fig. 2C, lane 10 with lane 12).



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 2. Endogenous OMT activities and OMT mRNA levels distributed within maize control plants. Maize plants were sampled 7 wk after planting in the greenhouse. Lanes 1 through 8 represent leaf samples from the second youngest to the ninth youngest leaf; Lane 9, young internodes; Lane 10, extending internodes; Lane 11, leaf sheaths; Lane 12, sorghum stem. (A) OMT activity was measured on each maize plant part and the highest OMT activity (0.76 pkat/mg protein) was considered as 100%. (B) Relative OMT mRNA steady state levels were determined by RNA quantification of the northern blot data. (C) Northern blot analysis of the OMT steady state levels using the maize omt cDNA probe. The blot was also hybridized with the 28S rDNA probe to standardize RNA loading and transfer.

 
Growth conditions, as well as developmental stage, influenced OMT activity (data not shown). Irregular growth and highly variable plant vigor and size of the T0 plants made their use impractical for comparative OMT analyses, so T1 progeny of 33 selected primary T0 transformants over 14 independent events were used for comparisons. Of those 33 T0 plants, 26 originated from 30 bombardments with the full-length vector and seven from 15 bombardments with the 5'-fragment antisense vector. Of the 14 independent events, four were from bombardments with the 5' omt antisense vector (events labeled 3, 4, 10, and 11 on Fig. 3). Although five seeds were used to initiate the T1 plants, they were segregating and the transgenic plants had to be identified by herbicide resistance. Because of that genetic segregation or possible silencing of bar, only 89 plants were herbicide resistant.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 3. OMT activity of 2-wk-old (5-leaf stage) plants: (A) Control plants, (B) A-group T1 transgenic plants. Plants initiated from five seeds resulting from self-pollination of 33 selected primary transgenic (T0) plants were established and grown in the greenhouse. The second youngest leaf from transgenic and control plants were assayed for OMT activity two weeks after planting. The mean of 10 control plants was considered as 100% (0.10 pkat/mg protein). The 33 plant families are labeled under the graph and individuals within a family were designated A1 to A5 sequentially from left to right (not labeled on the Fig.). A bm in the family designation indicates that the T0 parent had a brown midrib phenotype. Independent events are signified by nonconnecting lines below the family designations.

 
Those 89 herbicide-resistant, 2-wk-old T1 transgenic and control plants were assayed for their OMT activity (Fig. 3). The OMT activity of the transgenic plants assayed was highly variable. Compared with the mean of the controls, 62% of the transgenics showed an OMT activity reduction of more than 25%, and 20% of the transgenics had a reduction of greater than 50%. The most extremely downregulated omt transgenic plant had about 10% of the control OMT activity. As shown in Fig. 3, OMT activity variation within each transgenic family was as large as between transgenic families, so there were no significant differences among the transgenic families. However, the mean OMT activity of the transgenic plants was significantly reduced compared with the control plants (P = 0.007). The brown midrib phenotype was not correlated with reduced OMT activity at the 2-wk (five-leaf) stage. This could possibly be due to sampling the second youngest leaf while the brown midrib phenotype was observed on the two older, lower leaves and suggests different spatial and temporal expression of antisense omt.

A number of reports on antisense downregulation have indicated that only a few transgenics maintained low enzyme activity when sampled at a more mature stage (Atanassova et al., 1995; Piquemal et al., 1998). To investigate this problem, we sampled the third youngest leaf of 35 T1 plants with downregulated OMT (reduced 33% or more; Fig. 3) when those plants reached the 7-wk (15-leaf) stage. We found that a portion of the plants having low OMT activity at the 2-wk stage had an increase in activity to a level similar to the control, with some being even higher than the control. Among the 35, 7-wk (15-leaf) stage plants analyzed, only five plants (bm17-A1, bm8-A1, bm6-A1, 7-A3, and bm128-A3) had reduced OMT activity (<=50% of the control). Three of those five plants (bm 17-A1, bm8-A1, and bm6-A1), originated from bombardment with the full-length omt antisense vector, and two (7-A3 and bm128-A3), originated from the omt 5'-fragment antisense vector showing both vectors were capable of downregulating OMT.

To investigate OMT downregulation of the endogenous OMT mRNA steady state levels, northern blot analysis was performed with an omt antisense riboprobe with the third leaf of those five plants that maintained reduced OMT activity at the 15-leaf stage. As shown in Fig. 4A, the third leaf of all five plants showed a significant reduction of the maize OMT mRNA levels (1.7, kilobases, kb) compared with the control. An unknown RNA of 1.3 kb had strand-specific binding with the omt antisense probe. That RNA did not hybridize when either a cDNA probe or a sense-strand probe was used. The reduced maize OMT mRNA levels suggest the suppression of endogenous OMT mRNA accumulation by antisense OMT gene expression.



View larger version (73K):
[in this window]
[in a new window]
  		Fig. 4. Northern blot analysis with the strand-specific omt antisense riboprobe to detect endogenous OMT mRNA steady state levels. (A) The five A-group T1 plants that maintained low OMT activity through the 7-wk stage. Third youngest leaf was sampled. The 1.7-kb band is the endogenous OMT mRNA and is not visible in the transgenic plants but shows a strong band in the control. (B) B-group T1 plants sampled at the flag-leaf stage. The 1.7-kb band is strongly present in the controls and bm17-B3, which behaved as a control as it had lost the antisense omt. The endogenous 1.7-kb band is present at a reduced level in some transgenics and is not visible in others. The blot was sequentially hybridized with the strand-specific maize antisense omt riboprobe and the 28S rDNA probe. The reduced expression of the endogenous OMT suggests that antisense omt expression may promote instability of the endogenous OMT mRNA. The 1.3 kb unknown RNA has strand-specific binding with the omt antisense probe. That RNA did not hybridize when an omt cDNA probe or the sense riboprobe was used. A bm in the plant designation indicates that T0 parent had the brown midrib phenotype. Each plant family is generated from an independent event.

 
Since our major objective was to investigate the potential of forage quality improvement, sampling at a more mature stage (flag-leaf stage) would better reflect the effect of OMT downregulation at that utilization stage. To do that, the second group of T1 progeny was initiated using seeds from each of 10 selected T0 plants reflecting nine independent events and selected because of reduced family OMT activity from the 33 families shown in Fig. 3. The young flag leaves of those T1 plants and the R1 control plants were assayed for OMT activity. The results indicated that OMT activity was less variable within each transgenic family (Fig. 5) and was significantly different among transgenic families (P < 0.01). In transgenic families bm8, bm128, bm17, and 7, OMT activity was significantly reduced compared with OMT activity of the nontransgenic control plants.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 5. OMT activity of the B-group T1 plants sampled at the flag-leaf stage. Plants were initiated using seed from 10 self-pollinated primary transgenic plants selected on the basis of reduced family OMT activity (Fig. 3). Family members are designated B1 through B6. The young flag leaves from transgenic and control plants of these 7-wk-old plants were assayed for OMT activity. The mean of 10 control plants was considered as 100% (0.32 pkat/mg protein). A bm in the plant designation number under the graph indicates the T0 parent had a brown midrib phenotype. Bm17-B3 could be viewed as a control as it lost the antisense omt. Families are generated from independent events, except that families bm8 and bm9 are not independent.

 
The downregulation of the endogenous mRNA steady state levels of this second group of transgenic plants sampled at the flag-leaf stage is shown in Fig. 4B. That northern analysis with the omt antisense riboprobe also shows reduced OMT mRNA steady state levels in the transgenics, however with those plants, the reduction is variable as shown by the intensity of the 1.7-kb band varying from not visible (208-B1) to the control plant intensity. Plant bm17-B3 could be viewed as a control because the antisense omt transgene was not detectable by Southern analysis (see Fig. 6).



View larger version (110K):
[in this window]
[in a new window]
  		Fig. 6. Southern blot analysis of the B-group T1 plants. Genomic DNA (10 µg) was digested with EcoRI to release the 2.5- or 1.9-kb fragment of the sorghum omt antisense expression cassette (full-length or 5'-fragment of the sorghum cDNA, respectively) and hybridized with a sorghum omt cDNA probe. The sorghum omt antisense expression cassette includes part of the Ubi-1 promoter first intron region (about 600 bp), sorghum antisense omt (full-length or 5'-fragment of the sorghum omt cDNA), and nos terminator (see Fig. 1). {lambda}-DNA (HindIII) and {phi}X174 (HaeIII) were used as DNA size markers. Bm designations indicate the T0 parent had brown midrib phenotype. Families were generated from independent events.

 
Southern blot analysis was performed to determine the integration of the antisense omt in the genome of transgenic plants (six independent events). Genomic DNA was digested with EcoRI to release a 2.5-kb (from the full-length omt antisense vector) or a 1.9-kb fragment (from the omt 5'-fragment antisense vector). Those fragments did not contain the entire expression unit (Ubi-1-antisense omt-nos) but did contain the sorghum omt antisense region, about 600 bp of the Ubi-1 promoter sequence and the nos terminator (see Fig. 1). Hybridization was performed with the sorghum omt cDNA probe under high stringency conditions. Interestingly, the hybridization results (Fig. 6, lanes 15–16) showed that the maize Hi-II control plants had three hybridization bands (about 9.4, 6.5, and 4.4 kb, respectively) rather than a single band, which could be interpreted as RFLPs, since maize has been characterized as having a single omt gene. As indicated in Fig. 6, most of the herbicide-resistant plants contained a DNA fragment that hybridized to omt but differed from the control; some contained the 2.5- or 1.9-kb fragment as well as larger fragments. Antisense omt in plant bm8-B4 (Fig. 6, lane 12) segregated as compared with the other siblings (Fig. 6, lanes 9–11), which was consistent with the segregation of bar (data not shown), suggesting that the transgene was integrated at more than one locus. One herbicide-resistant plant (bm17-B3, lane 14, Fig. 6) was similar to the control when hybridized with omt, which suggested the loss of the omt transgene, probably due to recombination. However, genotypes such as 7-B1, bm5-B3 and bm8-B4 can be used for further investigation of stability of antisense omt inheritance and downregulation.

Transgene Expression in T1
Expression of bar and antisense omt was investigated by northern blots sequentially hybridized with omt, bar, and 28S rDNA probes. Ten T1 progeny from eight independent, T0 plants were analyzed. As indicated in Fig. 7, levels of bar transcript varied within the same transgenic family (Fig. 7B) as well as between independent events (e.g., lane 4 and lane 7 in Fig. 7A). Antisense omt transcripts could be detected in only a few samples (Fig. 7A, lanes 1, 4 and 7, and 7B, lanes 1 and 2) as compared with the 1.7-kb endogenous OMT mRNA transcript. In plants bm6-B1 and bm6-B3 (Fig. 7B, lanes 1 and 2), the larger size of the antisense transcripts appeared to be due to recombination (compare with Fig. 6, lanes 1 and 2). In some cases, the level of endogenous omt transcript was lower than the control (Fig. 7A, lanes 1 and 3), while in other cases, the endogenous OMT mRNA level was higher than the control (Fig. 7A, lanes 4–8, and 7B, lanes 1 and 2). The reduced endogenous OMT mRNA level suggests the stimulation of RNA turnover by antisense transcripts.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 7. Northern blot analysis of the T1 transgenic plants. Total RNA (10 µg per lane) was loaded. The blotted membrane was sequentially hybridized with the sorghum omt, bar, and 28S rDNA probes. (A) A-group T1 plants were sampled at the 2-wk-old stage (5-leaf stage) and (B) B-group T1 plants were sampled at the flag-leaf stage. The approximate sizes of the transcripts are marked. The 1.7-kb transcript is the maize endogenous OMT mRNA. The 1.6- and 1.0-kb transcripts represent the full-length, and 5'-fragment of the sorghum antisense omt vector, respectively. Families were generated from independent events, except that families bm8 and bm9 are not independent.

 
Histochemical Lignin Staining
Two histochemical staining methods, commonly used to roughly estimate lignin content, were used to stain free-hand sections taken from the mid region of the seventh and tenth internodes (from the top) of transgenic plants with reduced OMT activity or from nontransformed control plants. With phloroglucinol-HCl staining, the epidermis and vascular tissue of transgenic plants showed less red staining than control plants (Fig. 8A). With potassium permanganate-HCl staining, the reduced intensity of color and the change of brown-red to brown color were also observed in the transgenics (Fig. 8B). These results indicate that both lignin content and the composition of lignin, i.e., subunit ratios, have been changed in the transgenic plants that displayed reduced OMT activity.



View larger version (94K):
[in this window]
[in a new window]
  		Fig. 8. Lignin histochemical staining. Shown are the seventh and tenth internodes (from the top), of bm128-B3 transgenic and control plants. Plants were sliced into 1-mm free-hand sections from the top region of the internodes and subjected to (A) phloroglucinol-HCl, and (B) potassium permanganate-HCl staining. Note the more intense staining of the control sections.

 
Digestibility, Plant Yield, Lignin Content, and Other Cell Wall Parameters
Digestibility (in vitro organic matter digestibility, IVOMD), dry matter yield, ADL (acid detergent lignin), and NDF (neutral detergent fiber) percentage values for leaves, stems and whole plants of 16 T1 transgenic plants with downregulated OMT activity and 10 control plants (harvested five days after anthesis) are presented in Table 1. Whole-plant data were reconstructed mathematically from plant-part weights and percentages. Leaf sheaths were also analyzed separately, but because they represent only a small portion of the whole plant, their data are not presented separately, but are included in whole plant data.


View this table:
[in this window]
[in a new window]
  		Table 1. In vitro organic matter digestibility (% of organic matter), dry matter yield (g), ADL (%) and NDF (%) of leaf, stem and total plant tissue harvested five days after anthesis from transgenic and control plants. (ADF and sheath values not shown).

 
The leaf mean IVOMD percentage of transgenic plants was 74.7 ± 0.4 (range 77.9–72.8), whereas that of the control plants was 73.3 ± 0.3 (range 73.5–72.1), means significantly different (P = 0.046). The stem mean IVOMD percentage of transgenic plants was 76.7 ± 0.7 (range of 81.2–71.6), whereas that of the control was 71.4 ± 1.0 (range was 75.6–67.0), means significantly different (P = 0.0003). The whole plant mean IVOMD percentage for transgenic plants was 76.0 ± 0.4 (range of 78.2–74.1), while that of the control plants was 72.4 ± 0.4 (range 74.0–70.0), means significantly different (P = 0.0001).

Acid detergent lignin (ADL), which is widely used by animal nutritionists to measure lignin in forages, is presented in Table 1 as percentage of dry matter. The leaf mean ADL content of the transgenic plants was 1.57 ± 0.03 and that of the control plants was 1.78 ± 0.04, which represents a 13% lignin reduction for the transgenic plants compared with the controls. The stem mean ADL% for the transgenic plants was 1.98 ± 0.10 and that of the control plants was 2.47 ± 0.15, which represents a 20% lignin reduction for the transgenic plants compared with the controls. The ADL difference between the transgenic and control plants was highly significant for both leaves and stems (P = 0.0003 and 0.003 for leaves, stems, respectively). Whole plant mean ADL% for the transgenic plants was 1.72 ± 0.06 and 2.06 ± 0.06 for controls, which is a 16% reduction in lignin.

The relationship between IVOMD and lignin (ADL) content was measured by Pearson correlation coefficients. IVOMD was negatively correlated to leaf ADL (r = -0.47, P = 0.01), and to stem ADL, with stem having the strongest negative correlation (r = -0.70, P = 0.0001). The whole plant IVOMD was negatively correlated to ADL (r = -0.69, P = 0.0002). Flag-leaf OMT activity was correlated to leaf ADL content (r = 0.62, P = 0.0007) and the stem OMT activity was correlated to ADL content (R = 0.58, P = 0.0019). In general, lower OMT activity correlated with lower ADL.

Dry matter yields of the leaf, stem, and whole plants (Table 1) were highly variable and were not statistically different. Little can be concluded from this data as to whether transformation with antisense omt affects plant dry matter yield. NDF, which is an estimate of cell walls, is also summarized in Table 1. However, since ADF is an estimate of both lignin and cellulose, that data is not presented. Transgenic leaves appeared to have significantly reduced NDF compared with the control plants (P = 0.012). In stems, NDF was not significantly reduced in transgenic plants relative to the controls. Correlations between IVOMD and NDF were not significant.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Transgene Expression and OMT Downregulation by Antisense Strategy
Several factors affect transgene expression, such as the DNA integration site within the genome (position effect), the intactness of the transgene expression cassette, and transgene inactivation. Northern analysis data indicated that the expression levels of bar and antisense omt varied. We were able to detect antisense transcripts in only a few transgenics and, in those cases, with the antisense transcript present, the endogenous OMT mRNA steady state levels were not reduced. This could possibly be explained by: (1) different localization of the endogenous OMT mRNA and the antisense OMT mRNA molecules within the cell preventing them from forming a duplex and triggering degradation, or (2) the detected antisense RNA molecules possibly being transcribed in different cell types. In most transgenics, the antisense transcript was not detected and the endogenous OMT mRNA was reduced, which may suggest that an RNA degradation mechanism was operating. In those transgenics with a high endogenous OMT mRNA level and undetectable antisense transcripts, it is possible that either the antisense transcriptional level was lower than the detectable level, or the antisense gene was silenced. Further experiments with transcriptional run-on assays would be necessary to answer whether gene silencing occurred.

It is well documented that transgene expression varies among independent transformation events, individual plants from the same transformation event, and progeny of a transgenic plant (Gordon-Kamm et al., 1990; Spencer et al., 1992; Walter et al., 1992; Gallo-Meagher and Irvine, 1996; Zhang et al., 1996). We decided to screen progeny from many primary transgenic plants rather than focus on the progeny from only select plants. Initially, we chose progeny of 33 primary transgenic plants based on the presence of the brown midrib phenotype and independent transformation events. Our finding was that OMT activity was as highly variable within a transgenic family as between transgenic families. Among these analyzed plants, no correlation was observed between brown midrib phenotype and OMT activity, which may reflect the spatial and temporal expression patterns of the antisense omt. Considerable somatic instability of transgene expression was observed when we sampled 35 transgenic plants, that previously had reduced OMT activity, at the more mature 7-week (15-leaf) stage and found that only five plants maintained the depressed OMT activity (at about 50% of the control).

Northern analysis with the antisense riboprobe showed that the endogenous OMT mRNA levels of those five plants were severely reduced. To investigate forage quality improvement at a more practical utilization stage, a more mature stage was sampled. Young flag leaves of selected transgenic and control plants were sampled for OMT activity and OMT mRNA analyses. The data showed that both OMT activity and OMT mRNA levels were reduced in some of those transgenic plants and endogenous OMT mRNA expression was correlated with OMT activity levels at both the 15-leaf and flag-leaf stages.

Two vectors were used for bombardment, the full-length sorghum omt antisense vector was used for 30 bombardments and the 995 bp 5'-fragment omt antisense vector was used for 15, or 33% of the bombardments. Of the 33 T0 plants used in the OMT analyses, 24 were from the full-length vector and seven, or 21% of those T0 plants were from the 5' omt antisense vector bombardments. Of the 14 independent events analyzed, four or 29% of them were from the 5' omt antisense vector. T1 plants with reduced OMT, reduced lignin, and increased digestibility represented families of three T0 plants (7, bm5 and bm128) resulting from 5' omt antisense vector and families of four T0 plants from full-length vector bombardments. Both vectors were capable of downregulating OMT and were not statistically different in that capability suggesting they are equally effective.

It is reported that generally, 10 to 50% of the transgenic plants expressing an antisense gene show suppression of the target gene, and in some cases the antisense gene may have no effect at all (De Lange et al., 1993; Arndt and Rank, 1997). Recently, it was found that placing a gene fragment both in sense and antisense directions in the same construct leads to much higher efficiency in silencing virus and endogenous gene expression, especially when an intron is included between the gene fragments (Smith et al., 2000; Waterhouse et al., 1998; Wesley et al., 2001).

Downregulation of OMT, Increased Digestibility, Reduced, and/or Changed Lignin
The natural, brown-midrib mutant bm3, caused by a deletion or a B5 transposon insertion in the omt coding region (Vignols et al., 1995; Morrow et al., 1997), caused a reduction in lignin content and a change in lignin subunit composition measured by the sinapyl/guaiacyl (S/G) ratio (Cherney et al., 1991). Therefore, OMT has been considered a good downregulation target in transgenic tobacco (N. tabacum) and poplar trees (P. tremuloides). However, most of the studies of transgenic tobacco and poplar with downregulated OMT activity have shown that the lignin S/G ratio was significantly reduced, but the lignin level of the transgenic plants was similar to the controls, even when the OMT activity was reduced more than 90% (Dwivedi et al., 1994; Atanassova et al., 1995; Doorsselaere et al., 1995; Vailhe et al., 1996; Tsai et al., 1998; Lapierre et al., 1999). Contrary to those reports, Ni et al. (1994) and Sewalt et al. (1997) reported a moderate decrease of lignin content in their OMT downregulated transgenic tobacco plants. More recently, Guo et al. (2001) reported that downregulation of caffeic acid 3-O-methyltransferase (COMT) reduced lignin content and altered lignin composition in transgenic alfalfa (Medicago sativa L.).

An average lignin reduction of 16.5% was measured in transgenic maize plants being reduced to an average of 1.7% of plant dry weight compared with 2.1% in the control. Additionally, in the transgenic plants, the brown midrib phenotype and expression of modified color in stems stained with the phloroglucinol-HCl and potassium permanganate was observed which strongly suggests that lignin composition was modified. Along with this, and most likely because of reduced and modified lignin, digestibility (IVOMD) was improved in transgenic whole plants to 76% from 72.4% in controls (a 4.9% increase). Stems of transgenic plants had a greater lignin reduction (19.8%) compared with leaves (11.8%) and stems had a higher increase in digestibility (7.4%) compared with leaves (2.0%). This large increase in digestibility of stems relative to that of the control made the transgenic maize stems equally or more digestible than leaves. Although yield was measured in this study, the individual, greenhouse-grown plants were highly variable, giving no significant mean differences between transformed and control plants. In this study, NDF was lower in transgenic plants but was not significantly correlated to digestibility.

This work has shown that genetic engineering using antisense technology has potential to improve forage quality in grasses and that antisense technology can be used to downregulate OMT activity, which in turn can reduce lignin biosynthesis and modify the lignin molecule. Even though whole-plant digestibility improvement was modest, major improvement occurred in the stems. In most forage grasses, the digestibility of the leaves is much higher than the stems. Digestibility improvement of forage grass stems, as with this study, could make the less desirable stems much more acceptable and could improve forage intake and utilization. We observed brown midrib phenotype at the four- to six-leaf stage, and vascular tissue discoloration at a mature stage which indicate changed lignin composition. These phenotypes were similar to the natural bm3 mutants and were not observed in those control plants and progeny lacking the antisense gene. Our lignin histochemical staining of stem sections indicated that the transgenic plants with downregulated OMT activity had modified lignin composition. Both the ADL and histochemical measurements indicated that downregulated OMT also downregulated lignin. All above-mentioned experiments suggest that OMT may play an important role in the monolignol biosynthesis of monocots. Our data suggest that genetic engineering with antisense technology to downregulate OMT has the potential to reduce lignin biosynthesis and thereby improve the utilization efficiency of forage grasses.


   ACKNOWLEDGMENTS
 
We thank Dr. Raymond C. Littell, Statistics Department, University of Florida, for assisting with the statistical analyses. We also thank Dr. Christine D. Chase, Horticultural Department, University of Florida, and Dr. Robert G. Shatters, USDA, ARS, USHRL, Fort Pierce, FL, for technical advice and assistance. We also thank Jeffrey C. Seib, Douglas Manning, Robert Querns and Richard Fethiere, Agronomy Department, University of Florida, for assistance. Research was supported by the University of Florida Experiment Station and a USDA Special Grants/Tropical-Subtropical grant. Florida Agriculture Experiment Station Journal Series No. R08163.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Part of a dissertation submitted by the senior author in partial fulfillment of the requirements of the Ph.D. degree at the University of Florida, Gainesville, FL.

Received for publication January 9, 2002.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Plant Physiol.Home page
S. Guillaumie, H. San-Clemente, C. Deswarte, Y. Martinez, C. Lapierre, A. Murigneux, Y. Barriere, M. Pichon, and D. Goffner
MAIZEWALL. Database and Developmental Gene Expression Profiling of Cell Wall Biosynthesis and Assembly in Maize
Plant Physiology, January 1, 2007; 143(1): 339 - 363.
[Abstract] [Full Text] [PDF]

Home page
 J Exp BotHome page
S. Lunkenbein, E. M. J. Salentijn, H. A. Coiner, M. J. Boone, F. A. Krens, and W. Schwab
Up- and down-regulation of Fragariaxananassa O-methyltransferase: impacts on furanone and phenylpropanoid metabolism
J. Exp. Bot., July 1, 2006; 57(10): 2445 - 2453.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. Ralph, T. Akiyama, H. Kim, F. Lu, P. F. Schatz, J. M. Marita, S. A. Ralph, M. S. S. Reddy, F. Chen, and R. A. Dixon
Effects of Coumarate 3-Hydroxylase Down-regulation on Lignin Structure
J. Biol. Chem., March 31, 2006; 281(13): 8843 - 8853.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
J. Poerschmann, A. Gathmann, J. Augustin, U. Langer, and T. Gorecki
Molecular Composition of Leaves and Stems of Genetically Modified Bt and Near-Isogenic Non-Bt Maize--Characterization of Lignin Patterns
J. Environ. Qual., August 9, 2005; 34(5): 1508 - 1518.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
J. F. Pedersen, K. P. Vogel, and D. L. Funnell
Impact of Reduced Lignin on Plant Fitness
Crop Sci., March 28, 2005; 45(3): 812 - 819.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (12)
Right arrow Citing Articles via Google Scholar
Google Scholar