| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
USDA-ARS Plant Science Res. Unit, Dep. of Agronomy and Plant Genetics, 411 Borlaug Hall, 1991 Upper Buford Circle, Univ. of Minnesota, St. Paul, MN 55108
* Corresponding author (jungx002{at}umn.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Abbreviations: ADF, acid detergent fiber • ADL, acid detergent lignin • IVDMD, in vitro dry matter digestibility • IVNDFD, in vitro neutral detergent fiber digestibility • NDF, neutral detergent fiber • NIRS, near-infrared reflectance spectroscopy
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
As alfalfa stems develop, lignification and concentration of cell wall material increase (Buxton and Brasche, 1991). The primary cause for this increase in cell wall material is deposition of thick-walled, lignified xylem tissue by cambial activity after elongation of internodes is complete (Jung and Engels, 2002). It is clear that reductions in stem cell wall degradability result from this deposition of xylem tissue. Cell wall degradability by rumen microorganisms was high and virtually complete for elongating internodes because all tissues were nonlignified, other than protoxylem vessels (Jung and Engels, 2002). Stem cell wall degradability declined when alfalfa internodes entered post-elongation development, but several tissues (epidermis, collenchyma, chlorenchyma, secondary phloem, protoxylem parenchyma, and cambium) did not lignify and these tissues remained highly and rapidly degradable (Engels and Jung, 1998; Jung and Engels, 2001).
Selection for improved forage quality has been successful for increasing protein concentration and in vitro dry matter digestibility (IVDMD), and decreasing NDF and acid detergent lignin (ADL) concentration of alfalfa herbage (Coors et al., 1986; Hill, 1981; Hill and Barnes, 1977; Shenk and Elliott, 1971). However, because these selection studies were done on whole herbage, the observed shifts in forage quality may have resulted from inadvertent selection for altered leaf-to-stem ratio. Quality of alfalfa leaf and stem material are sufficiently different that any shifts in relative proportions of leaf and stem results in significant changes in herbage quality (Sheaffer et al., 2000). In the case of selection for reduced ADL concentration in alfalfa herbage (Hill, 1981), it was shown that the resulting divergence in herbage ADL concentration was due primarily to a higher leaf proportion in the low lignin alfalfa line, although the ADL concentration of stem material did show some response (Kephart et al., 1989; Kephart et al., 1990). While a higher leaf-to-stem ratio of the low lignin lines did result in reduced NDF concentration, the increase in IVDMD was minimal and only correlated with NDF concentration (Kephart et al., 1990). The lack of a correlation of IVDMD with ADL concentration suggests that cell wall digestibility of the alfalfa had not been altered. Buxton et al. (1987) found that genetic variation exists in forage quality of alfalfa stems, so direct selection for quality in this crucial plant part should be possible.
In addition to greater leaf proportion in alfalfa selected for higher quality, changes in other morphological traits associated with forage quality have been investigated. Length of the second internode from the stem base was weakly correlated with stem crude fiber concentration and IVDMD (r = 0.14 and –0.22, respectively, P < 0.05) (Heinrichs et al., 1969). Kephart et al. (1989) found a lower Kalu and Fick (1981) maturity index and shorter stem length, but no difference in internode number, for low lignin alfalfa lines. In contrast, a comparison of two high-quality alfalfa cultivars with two check cultivars found no difference in maturity index related to forage quality (Hall et al., 2000). Recently internode length was shown to be correlated with length of individual cells of stem tissues in a single alfalfa genotype, where longer internodes exhibited greater degradation of tissues in large particles (Engels and Jung, 2005). Clearly understanding of how stem morphology impacts cell wall digestibility of alfalfa is still incomplete.
The detergent fiber analysis system is used regularly for forage quality analysis (Van Soest et al., 1991). However; NDF, acid detergent fiber (ADF), ADL, and cellulose and hemicellulose calculated from detergent fiber data are not accurate measurements of cell wall concentration or composition (Theander and Westerlund, 1993). Therefore, understanding the biological basis for differences in NDF digestibility among forages requires detailed knowledge of cell wall concentration and composition. In an earlier report, Jung and Lamb (2003) examined the correlations of cell wall concentration and composition with both rapidly digestible and potentially digestible stem NDF within a large population of alfalfa plants. It was concluded that reduced stem cell wall concentration should improve both rate and extent of stem NDF digestion, and that reduced stem cell wall lignin and increased stem pectin concentrations would increase the potential extent of stem NDF digestion and the rapidly digestible stem NDF fraction, respectively. Our objectives were to conduct a multi-environment field study with groups of alfalfa clones identified as either low or high for either rapid (16 h) or potential (96 h) stem in vitro NDF digestibility (IVNDFD) to test these predictions; and to examine the roles of stem cell wall concentration and composition, and stem morphology traits in stem NDF digestibility.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Four groups, of five alfalfa clones each, were identified that exhibited divergent phenotypes: low rapid (16 h) IVNDFD, high rapid (16 h) IVNDFD, low potential (96 h) IVNDFD, and high potential (96 h) IVNDFD. These 20 clones were vegetatively propagated and transplanted in August of 2001 to replicated field trials at two University of Minnesota sites; Sand Plains Research Farm, Becker, MN (Hubbard loamy sand) and UMore Park, Rosemount, MN (Tallula silt loam; coarse-silty, mixed, superactive, mesic Typic Hapludolls). The Rosemount site was rain fed, while the Becker site was irrigated to meet plant moisture needs using the checkbook method (Wright and Bergsrud, 1991). At each site the clones were planted in a randomized complete block design with three replications. Field plots consisted of double rows spaced 15-cm apart with 12 plants of each clone per plot, at 15-cm intervals. Two similarly spaced rows of ‘Agate’ alfalfa were planted as borders on all sides of the plots. At both sites, soil test levels of P and K were maintained at recommended levels for high yields of alfalfa (Rehm et al., 2001). All plots were monitored for weeds and insects, and herbicides and insecticides were applied as needed.
Samples were collected from primary spring growth and the first summer regrowth period in 2002 and 2003. All plots at a site were harvested when visual observation suggested that approximately 50% of the plants had open flowers. Harvest dates were 14 June and 12 July 2002 and 17 June and 16 July 2003 at Becker, and 17 June and 15 July 2002 and 16 June and 11 July 2003 at Rosemount. At each harvest, all plants were cut at a 2-cm stubble height and 10 representative stem shoots were collected from each plot's harvested herbage for physical measurements. Total stem length, number of internodes, and presence of open flowers was determined for each of the 10 stem shoots collected from each field plot. The number of elongating internodes per stem was estimated by pliability and coloration of internodes. Data for these physical measurements were averaged by plot. The remaining whole herbage from each plot was bulked and placed in paper bags. After drying bulk herbage samples at 60°C, stems were hand separated from the leaf material and ground to pass a 1-mm screen in cyclone-type mill. The leaf/stem separation was done by separating individual stems from the mass of dried herbage, and then plucking leaves and petioles from the stems. Branches and stipules were not removed from the stem shoots. Small pieces of stem tips and branches that had been dislodged from the stems during sample bag handling were sorted out from the mass of loose leaves and included in the stem fraction.
Chemical and Digestibility Analyses
Data for cell wall traits and IVNDFD were predicted using NIRS. Spectra for analysis were collected for dried and ground alfalfa stem samples with a Foss (Foss North America Inc., Eden Prairie, MN) Model 6500 scanning monochrometer with a range of 1100 to 2500 nm. The prediction equations used for analysis of the alfalfa stem samples from this study were based on 221 calibration samples collected from several other studies of alfalfa stem quality conducted in Minnesota from 1998 through 2003, and updated with an additional 23 samples from the current study that were identified by the Intrasoft International (ISI, Port Matilda, PA) NIRS 3 ver. 4.0 software program "Select" as being outside the spectral characteristics of the previous calibration population. Final NIRS prediction equations were developed using the ISI NIRS 3 ver. 5.0 software program "Calibrate" with the modified partial least squares regression option and two passes to eliminate outliers (Shenk and Westerhaus, 1991). Outlier samples were identified as those samples with a residual between laboratory chemistry and NIR prediction results that had a t test of >1.8 times the standard error of calibration. The 1, 4, 4, 1 math treatment (first derivative, gap over which derivative was calculated, number of data points used in first smoothing, and second smoothing where one equals no smoothing; respectively) was used for all equation development. The calibration statistics for NIRS predicted traits are shown in Table 1. Reference methods used in developing the calibration data were as follows. All analyses of calibration samples were done in duplicate.
|
In vitro ruminal digestibility of NDF was determined as described by Jung and Lamb (2003). The alfalfa stem samples were incubated with a 20:80 (v/v) mixture of rumen fluid and McDougall's buffer (McDougall, 1948) at 39°C for 16 and 96 h. The rumen fluid was collected approximately 3 h post-feeding from a fistulated, lactating Holstein cow (Bos taurus L.) fed a total mixed ration including alfalfa hay, maize (Zea mays L.) silage, and a concentrate mixture. The samples were placed in Ankom filter bags and incubated in an Ankom Daisy oven. After the incubations were complete, residues in the filter bags were extracted with neutral detergent solution, as described above, to calculate IVNDFD. All calibration samples were tested, in duplicate, for IVNDFD at both incubation times. Filter bags with added glass fiber were used as blanks to correct for indigestible NDF entering the bags from the rumen fluid inoculum.
Statistical Analysis
Because of extensive winter kill between the 2002 and 2003 sampling years, there were insufficient surviving plants to include the Rosemount site in the data set for 2003. Therefore, the 2002 and 2003 harvests from Becker and the 2002 harvests from Rosemount were treated as three separate growth environments. Several clones did not survive the winter of 2002/2003 at both sites, therefore the number of alfalfa clones included in each stem IVNDFD clonal group was reduced. Both the low and high rapid stem IVNDFD clonal groups were reduced from five to four alfalfa clones each. The low potential stem IVNDFD group was reduced from five to two clones and the high potential stem IVNDFD group was reduced to three clones. Means were calculated across these reduced clone numbers within stem IVNDFD clonal groups for use in the analysis. Because of the loss of some clones and all the 2003 samples from the Rosemount site, the number of stem shoots collected for measurement of stem morphology was not equal among clonal groups. Total number of stem shoots examined ranged from 360 stem shoots for the low potential stem IVNDFD clonal group to 540 stem shoots for the high potential stem IVNDFD clonal group to 720 stem shoots for each of the rapid stem IVNDFD clonal groups.
Our objective was to evaluate differences between the high and low selections; therefore, the analyses of variance were conducted separately for the rapid and potential IVNDFD clonal group pairs. Both sets of data were analyzed by analysis of variance using a split-plot-in-time model combined over the three environments (Steel et al., 1996). All model parameters, other than replicate, were considered fixed effects. Clonal groups were considered the whole plot factor and harvests (spring and summer) were considered split-plots in time. Error terms used to test model parameters are shown in Tables 2 and 3. For those parameters having a significant F-test (P < 0.05), treatment means were compared using the least significant difference method. Pearson correlations were calculated for morphological, detergent fiber, and cell wall traits with IVNDFD results. Statistical calculations were done using the SAS version 9.1 PROC GLM and PROC CORR software packages (SAS Institute Inc., Cary, NC, USA).
|
|
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Rapid Stem IVNDFD Clonal Groups
The high rapid stem IVNDFD clonal group was greater for both 16- and 96-h IVNDFD than the low rapid stem IVNDFD clonal group (Table 4). Compared to the low rapid stem IVNDFD group, the high rapid stem IVNDFD group had lower stem NDF, ADF, and total cell wall concentrations. Stem cell walls of the high rapid stem IVNDFD clonal group had less hemicellulose and more pectin than did the corresponding low group. Rapid clonal group differences in lignin concentration were inconsistent between methods of analysis. Concentration of ADL as a component of NDF was slightly greater for the high rapid clonal group compared to the low clonal group, whereas the opposite effect was seen for cell wall Klason lignin. The high rapid stem IVNDFD clonal group had a small reduction in the xylose concentration of the stem hemicellulose fraction (85.9 vs. 84.9 ± 0.05 mol xylose mol–1 hemicellulose for low and high groups, respectively, P < 0.05). Mannose and fucose concentrations of the hemicellulose showed corresponding small increases in the high rapid stem IVNDFD clonal group (data not shown). The only shift in pectin composition between the low and high rapid stem IVNDFD clonal groups was a reduction in the galactose content of the high group (13.3 vs. 13.1 ± 0.03 mol galactose mol–1 pectin for low and high groups, respectively, P < 0.05); however, this response was not environmentally stable. The interaction of environment x clonal group was significant because galactose content of pectin was greater for the low rapid clonal group in two environments, but similar in the third environment compared to the high rapid clonal group (data not shown).
|
For the rapid stem IVNDFD clonal groups, stem 16-h IVNDFD was negatively correlated with stem NDF and total stem cell wall concentration, and positively correlated with pectin concentration of the stem cell wall (Table 5). Concentrations of stem ADF, ADL, and cell wall Klason lignin, cellulose, and hemicellulose were not correlated with 16-h IVNDFD; however, the molar proportions of fucose in hemicellulose and galactose in pectin were correlated with 16-h IVNDFD (r = 0.87 and –0.84 for fucose and galactose, respectively, P < 0.05). Concentrations of stem NDF and cell wall were not related to stem 96-h IVNDFD for the rapid stem IVNDFD clonal groups. Stem ADL and cell wall Klason lignin concentrations were negatively correlated with stem 96-h IVNDFD, and pectin concentration of the stem cell wall was positively correlated with stem 96-h IVNDFD (Table 5). Molar proportions of fucose in hemicellulose and galactose in pectin were correlated with stem 96-h IVNDFD (r = 0.69 and –0.59 for fucose and galactose, respectively, P < 0.05). Stem morphological traits were not correlated with either 16- or 96-h IVNDFD for the rapid stem IVNDFD clonal groups (data not shown).
|
The low potential stem IVNDFD clonal group had fewer flowers, and shorter stem and mean internode length than did the corresponding high group (Table 4). The number of internodes per stem did not differ between the potential stem IVNDFD clonal groups. The significant interaction of environment x clonal group for number of elongating internodes (Table 3) was because the high potential clonal group had more elongating internodes than the low clonal group in one environment, but the two clonal groups did not differ in number of elongating internodes in the other two environments (data not shown).
Stem 96-h IVNDFD was negatively correlated with stem ADL and cell wall Klason lignin concentrations, and positively correlated with stem cellulose concentration of the cell wall for the potential stem IVNDFD clonal groups (Table 4). Components of the hemicellulose and pectin fractions was not correlated with 96-h IVNDFD for these clonal groups. The only cell wall trait which was correlated with 16-h IVNDFD was molar proportion of uronic acids in pectin for the potential stem IVNDFD clonal groups (r = 0.62, P < 0.05). As observed with the rapid stem IVNDFD clonal groups, no stem morphological traits were correlated with either 16- or 96-h IVNDFD for the potential stem IVNDFD clonal groups (data not shown).
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The stem tissues of alfalfa which never lignify during development (epidermis, collenchyma, chlorenchyma, secondary phloem, cambium, and protoxylem parenchyma) are rapidly and completely degraded by rumen microorganisms (Jung and Engels, 2001; Jung et al., 2004). In the current study, those alfalfa plants with rapidly digestible NDF (high rapid stem IVNDFD clonal group) may have had relatively more of their stem mass in nonlignified tissues than the low rapid stem IVNDFD clonal group, thereby resulting in the observed digestibility phenotype. Nonlignified alfalfa stem tissues are also believed to be lower in total cell wall concentration, and richer in pectin and branched hemicellulose than lignified, thick-walled phloem fiber and xylem tissues (Jung and Engels, 2002). The lower cell wall concentration, greater pectin and reduced Klason lignin concentration of the cell wall material, and reduced xylose content of the hemicellulose fraction of the high rapid stem IVNDFD clonal group all suggest that a larger proportion of stem mass was present in these nonlignified tissues of the alfalfa plants with rapidly digestible NDF. Such a greater proportional mass of nonlignified tissues compared to lignified tissues could result from either greater numbers of cells for the nonlignified tissues or a reduced rate of development for the lignified tissues. Another mechanism by which alfalfa stems may have more nonlignified stem tissues would be if the stems had more elongating internodes because such internodes contain only a single lignified tissue (protoxylem vessels) (Jung and Engels, 2002). However, this was not the case in the current study where all clonal groups had similar numbers of elongating and total internodes per stem. Unfortunately, we did not measure the mass of the individual internodes and could not determine if the high rapid stem IVNDFD clonal group had proportionately more elongating internode mass than the low group.
Because the high clonal group for potential stem IVNDFD did not differ in total stem cell wall concentration from the corresponding low group, a major shift in tissue proportions between lignified and nonlignified tissues is an unlikely explanation for the observed digestibility phenotype. This conclusion is based on the greater amount of cell wall material present in lignified than nonlignified tissues. The fact that the potential stem IVNDFD clonal groups did not differ for 16-h IVNDFD also supports this conclusion because nonlignified tissues are more rapidly degraded than lignified alfalfa stem tissues (Jung and Engels, 2001). A more likely explanation for the high potential stem IVNDFD phenotype is that development of lignified tissues was altered.
Phloem fiber, xylem fiber and vessels, and pith parenchyma all develop lignified cell walls, but the patterns of development differ. Phloem fiber cells develop a thick secondary wall which is believed to be rich in cellulose; however, lignification is limited to just the cell lumen side of an irregularly thickened primary wall (Engels and Jung, 1998). If phloem fiber cells had reduced lignin deposition, then lignin concentration of the stems would be reduced with little change in cell wall concentration. Whether alfalfa stems contain sufficient phloem fiber tissue to measurably impact total stem lignin concentration is unknown. Parenchyma cells in the pith region of alfalfa stems undergo minor secondary wall thickening in late post-elongation development, but the cell walls become lignified and poorly degraded (Engels and Jung, 1998; Jung and Engels, 2001). As for phloem fiber, if the pith parenchyma were to remain nonlignified, then lignin concentration would be reduced but cell wall concentration would be unchanged. The likelihood of such a developmental pattern is limited because the pith parenchyma of alfalfa stems often senesces and is degraded, leaving a hollow stem (Jung and Engels, 2002). At this time we do not know if the alfalfa clones in the current study developed a hollow stem pith region.
The majority of the stem cell wall material and lignin in alfalfa is located in the xylem tissues, particularly xylem fiber cells (Jung and Engels, 2002). Observations on developing internodes indicated that lignification of xylem fiber cell walls occurred immediately after addition of this tissue by post-elongation cambial activity (Engels and Jung, 1998). Secondary wall deposition and lignification were essentially simultaneous events in xylem tissues. If the amount of lignin deposited in xylem fiber cell walls were reduced or delayed while secondary walls continued to develop, then this would lead to the lower Klason lignin concentration observed in the high 96-h IVNDFD clonal group. Whether these highly synchronized events in secondary wall development can be decoupled is unknown. An alternative mechanism with regard to xylem fiber would be an increase in the amount of nonlignified additional secondary wall material deposited in some xylem fiber cells (Engels and Jung, 1998). This unique wall layer has been shown to be slowly, but highly degraded (Engels and Jung, 2005); therefore, increased deposition of this nonlignified wall layer would confer the observed high 96-h IVNDFD phenotype and reduce Klason lignin concentration. However, an increase in cell wall concentration would also be expected from such a developmental pattern, a result not observed in the present study.
Recently it was reported that length of a particular internode of alfalfa, from a single clone, was positively correlated with length of cells for several stem tissues (Engels and Jung, 2005). It was also noted that large particles (2–3 cm in length) excised from longer stem internodes from this alfalfa clone were degraded to a greater depth from a cut end than observed for large particles excised from shorter internodes. Engels and Jung (2005) hypothesized that, if other alfalfa genotypes have positive correlations of internode length and length of cells, maximum cell wall degradation of rapidly degradable nonlignified tissues will occur in genotypes with short internodes that allow rapid movement of rumen microorganisms from one degradable cell wall surface to the next. Additionally, because lignified walls have been shown to act as impenetrable barriers to microbial access through degradation (Engels, 1989; Wilson and Mertens, 1995); maximum extent of cell wall degradation, which is primarily a function of the partial degradation of thick-walled lignified tissue, should occur in genotypes with long internodes such that ruptured cells expose more degradable cell wall surface area per cell, and long cells are more likely to be ruptured than short cells (Engels and Jung, 2005). This hypothetical model for impact of internode length on cell wall degradability could be partially tested in the current study.
Internode number per stem was virtually identical among clonal groups in the current study, but stem length was greater for the low rapid and high potential stem IVNDFD clonal groups than their respective counterpart groups. As a result, alfalfa clones with a high rapid stem IVNDFD phenotype had shorter internodes than the low rapid stem IVNDFD clonal group, and the high potential stem IVNDFD clonal group had longer internodes than the low potential stem IVNDFD clonal group. These differences for the low and high IVNDFD clonal groups support the model of Engels and Jung (2005). It remains to be determined if individual cell lengths of stem tissues are correlated with internode length for clones from the current experiment. Other stem morphology traits such as flowering and internode number did not appear to impact alfalfa NDF digestibility which is in general agreement with the conclusion of Hall et al. (2000) that alfalfa stem morphology is not related to forage quality.
|
CONCLUSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Received for publication December 12, 2005.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. D. Casler, H. G. Jung, and W. K. Coblentz Clonal Selection for Lignin and Etherified Ferulates in Three Perennial Grasses Crop Sci., March 19, 2008; 48(2): 424 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Schnurr, H.-J. G. Jung, and D. A. Samac A Comparative Study of Alfalfa and Medicago truncatula Stem Traits: Morphology, Chemical Composition, and Ruminal Digestibility Crop Sci., July 30, 2007; 47(4): 1672 - 1680. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
