Article pubs.acs.org/JAFC
Molecular Structures and Metabolic Characteristics of Protein in Brown and Yellow Flaxseed with Altered Nutrient Traits Nazir Ahmad Khan,†,§ Helen Booker,‡ and Peiqiang Yu*,†,⊥ †
Department of Animal and Poultry Science, and ‡Department of Plant Sciences, University of Saskatchewan, Saskatoon, S7N 5A8 Saskatchewan Canada ⊥ College of Animal Science and Animal Veterinary, Tianjin Agricultural University, 22 Jinjin Road, Tianjin 300384, China ABSTRACT: The objectives of this study were to investigate the chemical profiles; crude protein (CP) subfractions; ruminal CP degradation characteristics and intestinal digestibility of rumen undegraded protein (RUP); and protein molecular structures using molecular spectroscopy of newly developed yellow-seeded flax (Linum usitatissimum L.). Seeds from two yellow flaxseed breeding lines and two brown flaxseed varieties were evaluated. The yellow-seeded lines had higher (P < 0.001) contents of oil (44.54 vs 41.42% dry matter (DM)) and CP (24.94 vs 20.91% DM) compared to those of the brown-seeded varieties. The CP in yellow seeds contained lower (P < 0.01) contents of true protein subfraction (81.31 vs 92.71% CP) and more (P < 0.001) extensively degraded (70.8 vs 64.9% CP) in rumen resulting in lower (P < 0.001) content of RUP (29.2 vs 35.1% CP) than that in the brown-seeded varieties. However, the total supply of digestible RUP was not significantly different between the two seed types. Regression equations based on protein molecular structural features gave relatively good estimation for the contents of CP (R2 = 0.87), soluble CP (R2 = 0.92), RUP (R2 = 0.97), and intestinal digestibility of RUP (R2 = 0.71). In conclusion, molecular spectroscopy can be used to rapidly characterize feed protein molecular structures and predict their nutritive value. KEYWORDS: protein molecular structure, protein subfraction, flaxseed, yellow flaxseed
■
INTRODUCTION Flaxseed (Linum usitatissimum L.), also known as linseed, is one of the most important commercial oilseed crops of Western Canada, providing valuable inputs for the food, feed, manufacturing, and pharmaceutical industries. Since 1994, Canada has been the world largest flaxseed producer and exporter of raw seeds,1 with a total production of 48.9 million tons in 2012.2 In the last two decades, the use of full-fat flaxseed in dairy ration has been increased due to its known positive effects on animal fertility and reproductive health,3,4 fatty acid composition of milk,5 and as a consequence on long-term human health.6 Because of high contents of oil (>38%) and crude protein (CP; >20%), and adequate content of neutral detergent fiber (NDF; ∼18%), full-fat flaxseed can be used (up to 10% of dry matter (DM)) as an attractive energy and CP source in dairy ration.5,7 Moreover, the defatted coproduct (meal/presscake), produced after the extraction of oil from flaxseed for industrial applications, is predominantly used as a rich source of CP (>36% of DM)8,9 in dairy ration. Flaxseed meal contains a relatively high amount of NDF (∼28%)10 due to the high content of hull in the traditional brown-seeded cultivars. Decreasing the proportion of hull in the meal will enhance the overall quality of flax meal as a food and feed ingredient. Increasing the oil content of flaxseed, and the proportion of C18:3n-3 in the total oil, has been the focus of recent flax breeding in Canada. As a result, yellow flaxseed breeding lines with a high content of oil and C18:3n-3 have been recently developed by the Crop Development Centre (CDC), University of Saskatchewan (Saskatoon, SK, Canada). The yellow flaxseed has a larger seed size and thinner seed coat compared to those of the brown flaxseed varieties, resulting in © 2014 American Chemical Society
higher contents of oil and CP, and lower content of NDF in the seeds. A higher content of CP and lower content of NDF have been reported for yellow-seeded canola meal11 and carinata seeds12 compared to those of their brown-seeded counterparts. Moreover, the CP in flaxseed contains a high fraction of soluble protein (SP; 76% CP)13 compared to other oilseeds such as canola (22% CP) 11 and carinata (49% CP). 12 As a consequence, flaxseed-derived CP rapidly and extensively degrades in the rumen,6 causing lower fermentation efficiency and losses of gaseous N to the environment. Recently, it has been shown that yellow-seeded canola meal had a significantly lower content of SP, resulting in a lower rate and extent of ruminal CP degradation compared to those of brown-seeded canola meal,11 whereas no significant differences in the SP and degradation kinetics of yellow and brown seeded carinata seeds were observed.12 This background provides an impetus to investigate the nutritive value, digestive behavior, and utilization of CP in the newly developed yellow flaxseed in ruminant nutrition. Recent research has unequivocally shown a strong relationship between feed protein’s inherent molecular structural makeup and CP solubility, ruminal degradation characteristics, intestinal digestibility, and utilization.14−16 Therefore, quantifying the protein’s inherent molecular makeup may be vital for understanding the variation in CP subfraction profiles, solubility, and digestibility between the different varieties of flaxseed. The objectives of the present study were to investigate Received: Revised: Accepted: Published: 6556
March 14, 2014 June 11, 2014 June 15, 2014 June 16, 2014 dx.doi.org/10.1021/jf501284a | J. Agric. Food Chem. 2014, 62, 6556−6564
Journal of Agricultural and Food Chemistry
Article
the (1) chemical profile; (2) CP subfractions according to Cornell Net Carbohydrate and Protein System (CNCPS); (3) ruminal CP degradation characteristics and intestinal digestibility of rumen nondegraded protein (RUP); and (4) the protein’s molecular structural features using Fourier transform/ infrared-attenuated total reflectance (FT/IR-ATR) molecular spectroscopy of newly developed CDC yellow flaxseed in comparison with that of brown flaxseed. The fifth objective was to quantify the relationship between the protein’s molecular structural features and their chemical profiles and digestive behavior in the rumen and intestine of dairy cattle.
■
MATERIALS AND METHODS
Flaxseeds, Samples Collection, and Heat Processing. Four varieties of flaxseed (Linum usitatissimum L.), including two brown oilseed flax varieties (CDC Bethune and CDC Sorrel) and two newly developed yellow oilseed flax breeding lines (coded as FY11140 and F08×425) were selected for this study. Seeds (∼2 kg) from each genotype were sampled from two replicate plots at two different locations, namely, from the crop grown on a clay-loam soil at Kernen Crop Research Farm (52°9′ N and 106°32′ W; University of Saskatchewan, Saskatoon, SK, Canada) and from the crop gown on a silty-clay soil at Melfort Research Station (52°8′ N and 104°59′ W; Melfort, SK, Canada) of Agriculture and Agri-Food Canada. The samples of each genotype were pooled by location and mixed, and subsamples of ∼1 kg were collected and stored at −20 °C for laboratory analysis. Subsequently, the seeds were ground through a 2 mm screen for in situ studies and through a 1 mm screen for wet chemical analysis using a laboratory mill (Retsch ZM-1, Brinkmann Instruments Ltd., Mississauga, ON, Canada). The seeds were ground through a 0.25 mm screen for molecular spectral analyses. Because of high oil content of the seeds, the samples were kept cool and gradually fed to the grinder to avoid sticking and clumping and to make sure that the seeds were cracked and not extruded. Molecular Spectroscopy. The samples were analyzed for the protein’s molecular spectral profiles using a JASCO FT/IR-ATR-4200 (Jasco Inc., Easton, MD) at the molecular spectroscopy laboratory of the Department of Animal and Poultry Science, University of Saskatchewan (Saskatoon, SK, Canada). The spectrometer was outfitted with a ceramic IR light source, a deuterated L-alanine doped triglycine sulfate detector, and a MIRacle ATR (Pike Technologies, Madison, WI, USA) sampling accessory equipped with a pressure clamp and ZnSe crystal plates. The molecular spectrum was generated in the mid-IR region (ca. 4,000−800 cm−1) at a spectral resolution of 4 cm−1. For each sample, five spectra were obtained, and each spectrum was generated with 128 co-added scans. Jasco Spectra Manager II software (Jasco Inc., Easton, MD) was used to collect, correct (against background and CO2 noise), and process the molecular spectral data. Ominic 7.3 software (Spectra-Tech Inc., Madison, WI, USA) was used to identify and quantify the absorbance bands associated with the protein’s primary and secondary structures. The protein’s primary structural characteristics were quantified from two bands at the amide I (1712−1580 cm−1; Figure 1) and amide II (1,580−1482 cm−1; Figure 1) regions using literature information as reported earlier.14 The protein’s secondary structural characteristics were quantified from the absorption peak height of the α-helix (centered at ca. 1,653 cm−1) and β-sheet (centered at ca. 1,630 cm−1) in the amide I region. The α-helix and β-sheet peaks were indentified using Fourier self-deconvolution spectra, obtained with the secondderivative function of OMNIC 7.2 as described by Yu.13 In Situ Rumen Incubation. Four rumen fistulated (10 cm internal diameter; Bar Diamond Inc., Parma, ID, USA) lactating Holstein dairy cows (body weight, 680 ± 10 kg) were used for in situ incubation. The cows were handled and cared for according to the recommendations of CCAC.17 A balanced TMR was individually fed to the dairy cows twice daily (8:00 and 16:00 h) in equal portions. The TMR contained 53% dairy concentrate, 31% barley silage, and 16% chopped alfalfa hay. Detailed information on the ingredients and chemical composition of
Figure 1. Protein molecular spectra of raw brown flaxseed varieties (CDC Bethune and CDC Sorrel) and newly developed yellow flaxseed breeding lines (FY11140 and FO8×425) in the amide I and amide II regions (1712−1482 cm−1).
the concentrates, barley silage, and alfalfa hay has been reported earlier.18 All cows had 24 h/d access to fresh drinking water. The standard in situ nylon bag protocol described by Peng et al.14 was used to determine the rumen degradation kinetics of DM and CP. Briefly, 7 g of ground (2 mm) sample of each feed was randomly incubated for 48, 24, 12, 8, 4, 2, and 0 h in rumen of the four cows, using preweighed and coded nylon bags (10 × 20 cm) with a pore size of 41 μm (Nitex 03-41/31 monofilament open mesh fabric, Screentec Corp., Mississauga, ON, Canada). All samples were incubated in two replicated runs, and multiple bags were used for each incubation time to obtain more representative data and enough residues for chemical analysis. For 48, 24, 12, 8, 4, 2, and 0 h incubation periods, the replicate bags were 4, 4, 4, 2, 2, 2, and 2, respectively. The gradual addition, all out method was used. A maximum of 20 bags were incubated in the rumen at any given time. All bags were removed from the rumen at the end of 48 h of incubation and immediately washed (6 times) in a bucket with cold tap water to remove adhering ruminal contents and stop enzymatic and microbial degradation. The bags were subsequently dried (55 °C for 48 h), cooled to room temperature (20−22 °C), and reweighed. Residues were pooled together as per incubation time and run, ground to 1 mm particle size, and stored at 4 °C for chemical analysis. Rumen Degradation Kinetics. Rumen degradation kinetics of DM and CP was computed according to the first-order kinetics equation of Ørskov and McDonald19 with a slight modification on lag time.20,21 Y (t ) = U + (100 − S − U) × e−kd(t − T0) where Y(t) was the residues at t h of incubation; S was the soluble fraction; U was the rumen nondegradable fraction; kd was the ruminal degradation rate (%/h) of the potentially degradable fraction, and T0 was the lag time (h). The kd, U fraction, and T0 were estimated from the in situ data using the nonlinear (PROC NLIN) procedure of SAS22 with iterative least-squares regression (Gauss−Newton method). The rumen nondegraded DM (RUDM) and RUP were computed according to the NRC-200123 as follows:
⎡ ⎤ Kp ⎥ RUDM/RUP = U + D × ⎢ ⎢⎣ (Kd + K p) ⎥⎦ where D is the potentially rumen degradable fraction and calculated as D = 100 − S − U, and KP is the passage rate of digesta from the rumen, which was supposed to be 0.06/h.21 The rumen degradable DM (RDDM) and CP (RDP) were calculated as 6557
dx.doi.org/10.1021/jf501284a | J. Agric. Food Chem. 2014, 62, 6556−6564
Journal of Agricultural and Food Chemistry
Article
Table 1. Chemical Profiles of Flaxseed: Comparison of the Brown Seed Type vs Yellow Seed Typea brown seed type
contrastc
yellow seed type b
Item
CDC Bethune
CDC Sorrel
FY11 140
F08×425
SEM
brown vs yellow
P- valuec
Basic Chemical Profile DM (%) ether extract (EE) (% of DM) ash (% of DM) crude protein (CP) (% of DM) NDIPd (% of CP) ADIPe (% of CP)
94.28 ab 42.88 b 3.70 a 22.04 ab 15.89 ab 2.56 b
94.25 b 39.96 c 3.69 a 19.79 b 17.02 a 4.20 b
95.46 a 42.60 b 3.82 a 25.70 a 11.39 b 5.05 b
95.47 a 46.49 a 3.38 b 24.18 ab 11.49b 10.54 a
0.206 0.351 0.041 0.960 0.793 0.811
** *** † *** ** **
* *** ** ** * **
a−c: means within a row with different letters differ at the P < 0.05 level. bSEM = standard error of the mean. c†, P < 0.1; *, P < 0.05; **, P < 0.01; ***, P < 0.001. dNDIP = neutral detergent-insoluble protein. eADIP = acid detergent-insoluble protein. a
⎡ ⎤ Kd ⎥ RDDM/RDP = S + D × ⎢ ⎢⎣ (Kd + K p) ⎥⎦
The PROC CORR procedure of SAS22 was used to compute the strength of relationships between the protein’s molecular structural characteristics and CP chemical profile, subfractions, in situ degradation parameters, and intestinal digestibility of RUP. The PROC REG procedure of SAS22 with stepwise selection at P < 0.05 was used to obtain multiple regression equations for the prediction of the chemical profile, CP subfractions, in situ degradation parameters, and intestinal digestibility of RUP. The independent variables tested in the model were amide I area, amide II area, and their ratio; amide I height, amide II height, and their ratios; and α-helix and β-sheet height and their ratios. Only equations contributing significantly (P < 0·05) to the estimation of dependent variables are reported. Multivariate analysis, the agglomerative hierarchical cluster analysis (CLA), and principal component analysis (PCA) were performed on the entire spectral data (n = 249 spectral bands), obtained from the amide fingerprint region (ca. 1732−1483 cm−1), to classify and visualize varietal differences in intrinsic protein molecular structures. The concepts and procedure of using CLA and PCA for FT/IR-ATR spectral data are described earlier.31
Chemical Analysis. The standard methods of the AOAC24 were used to analyze the contents of DM (method 930.15), CP (method 984.13; adopted for Kjeltec 2400 autoanalyzer; Foss Analytical A/S, Hillerød, Denmark), ash (method 942.05), and ether extract (EE, method 920.39). The content of SP was analyzed according to the method of Roe et al.25 with slight modifications as described by Peng et al.14 The contents of acid detergent-insoluble protein (ADIP), neutral detergent-insoluble protein (NDIP), and nonprotein N (NPN) were analyzed according to the methods described by Licitra et al.26 The 3-step in vitro method of Calsamiglia and Stern27 was adopted to analyze the digestibility of RUP in air-dried rumen residues of 16 h rumen incubation as described by Peng et al.18 Fractionation of Crude Protein. The CP subfractions were computed according to the Cornell Net Carbohydrate and Protein System (CNCPS).28,29 The CP was partitioned into NPN (PA), true protein (PB), and ADIP (PC). The PA fraction contains peptides, free amino acids, ammonia, amides, amines, ureides, nucleotides, and nitrates, and is considered to degrade in the rumen at the rate of 200%/h.30 The PB fraction was further fractionated into a rapidly (10 to 40%/h) degradable fraction (PB1); a moderately (3 to 20%/h) degradable fraction (PB2); and slowly (0.06 to 0.55%/h) degradable fraction (PB3).30 PB1 was computed as SP − PA; PB2 as CP − (PA + PB1 + PB3 + PC); and PB3 as NDIP − ADIP. These CP fractions were computed from the data obtained with wet-chemical analysis using the following approach, the trichloroacetic acid precipitated CP = PB1 + PB2 + PB3 + PC; borate-phosphate buffer insoluble CP = PB2 + PB3 + PC; NDIP = PB3 + PC; and ADIP = PC.26 Statistical Analysis. The data on chemical composition, CP subfractions, and the protein’s molecular spectral characteristics were analyzed using the PROC MIXED procedure of SAS.22 The model used for analysis was
■
RESULTS AND DISCUSSION Chemical Profiles. The results of chemical profiles of the brown flaxseed varieties and yellow flaxseed breeding lines are shown in Table 1. Except for ash, the contents of all nutrients varied (P < 0.05) among the four genotypes. On average, the yellow-seeded lines had higher (P < 0.001) contents of EE (44.54 vs 41.42% DM) and CP (24.94 vs 20.91% DM) compared to those of the brown-seeded varieties. The higher contents of CP and EE in the yellow-seeded lines could be related to their larger seed size32 and thinner and translucent seed coat32,33 that results in a relatively greater proportion of oil and protein containing mass (cotyledons and endosperm) and smaller proportion of fiber containing mass (seed coat) compared to those of the brown seeded varieties.11 In agreement with our findings, higher contents of CP and EE have been reported for yellow-seeded canola meal compared to that of its brown-seeded counterpart.11,34 Within the yellowseeded lines, FO8×425 had a higher (P < 0.05) content of EE; and within the brown-seeded varieties, CDC Bethune had a higher (P < 0.05) content of EE and CP. The latter findings suggest some genotype differences in the chemical profiles of flaxseed within the yellow and brown seed types. The CP chemical profile revealed that on average the CP in yellow flaxseed lines had a higher (P < 0.01) content of ADIP (7.80 vs 3.38% CP). The ADIP is believed to be nondegradable in rumen and nondigestible in the postruminal tract. In contrast to our findings, previous studies have reported lower contents of ADIP in yellow-seeded carinata strains and canola meal compared to their brown-seeded counterparts.11,12 This discrepancy may be related to the inherent genetic variation
Yijk = μ + Vi + Lj + e ijk where Yijk is an observation on the dependent variable ijk; μ is the population mean; Vi was the fixed effect of flaxseed variety (CDC Bethune, CDC Sorrel, FY11140 and F08×425), Lj is the random effect of crop location (Kernen and Melfort) and eijk is the random error associated with the observation ijk. The data on in situ rumen degradation parameters and intestinal digestibility were analyzed with the following model:
Yijk = μ + Vi + R k + e ijk where Yijk is an observation on the dependent variable ijk; μ is the population mean; Vi is the fixed effect of flaxseed variety, Rk is the random effect of in situ experimental run, and eijk is the random error associated with the observation ijk. When significant (P < 0.05) effects were observed, posthoc analyses were conducted on the least-squares means to compute pairwise differences among the means, using the Tukey-Kramer test adjusted for multiple comparisons. 6558
dx.doi.org/10.1021/jf501284a | J. Agric. Food Chem. 2014, 62, 6556−6564
Journal of Agricultural and Food Chemistry
Article
Table 2. Cornell Net Carbohydrate and Protein System (CNCPS) CP Subfraction Profiles of Flaxseed: Comparison of the Brown Seed Type vs Yellow Seed Typea brown seed type item
CDC Bethune
contrastc
yellow seed type
brown vs yellow
P valuec
1.537 0.700 1.208 0.978 0.811 1.456
** * ns *** ** **
* ** * ** ** **
1.519 1.299 1.083
* * ***
* * **
CDC Sorrel
FY11 140
F08×425
SEM
3.12 b 49.00 b 30.90 a 12.83 a 4.20 b 92.69 a
7.98 ab 51.57 b 29.05 a 6.34 b 5.05 b 86.96 a
13.77 a 49.60 b 25.13 b 0.95 b 10.55 a 75.69 b
52.83 b 33.34 a 13.84 a
59.34 ab 33.41 a 7.26 b
65.59 a 33.19 a 1.22 b
b
d
Crude Protein Subfraction (% of CP) PA 4.71 b PB1 56.18 a PB2 23.23 b PB3 13.33 a PC 2.56 b total true protein 92.74 a True Proteine (% of True Protein) PB1 60.59 ab PB2 25.05 b PB3 14.35 a
a a−b: means within a row with different letters differ at the P < 0.05 level. bSEM = Standard error of the mean. cns, not significant; *, P < 0.05, **; P < 0.01; ***, P < 0.001. dPA = NPN, fraction of CP that is instantaneously solubilized at time zero; PB1 = fraction of CP that is rapidly degraded in the rumen and calculated as soluble protein − NPN; PB2, intermediately degradable fraction and calculated as CP − (PA + PB1+ PB2 + PB3 + PC); PB3, slowly degradable crude protein fraction and calculated as (NDIP − ADIP); and PC, fraction of CP recovered with ADF and is considered to be nondegradable. eTrue protein = PB1 (% of CP) + PB2 (% of CP) + PB3 (% of CP).
Table 3. In Situ Rumen Degradation Kinetics and Intestinal Digestibility of Flaxseeds: Comparison of the Brown Seed Type vs Yellow Seed Typea brown seed type item
CDC Bethune
contrastc
yellow seed type
CDC- Sorrel
FY11 140
F08×425
SEMb
brown vs yellow
P valuec
8.470 ab 0.08 25.52 a 67.17 b 7.31 bc 64.79 a 35.20 c
9.66 a 1.05 15.74 c 80.18 a 4.09 c 65.18 a 34.82 c
0.286 0.245 0.628 1.097 0.782 0.372 0.372
** ns *** ns *** *** ***
** * *** *** *** *** ***
10.62 a 0.04 b 30.78 a 64.63 c 4.59 b 72.04 a 27.96 c
11.28 a 1.19 a 19.93 c 76.07 b 4.00 b 69.57 b 30.43 b
0.462 0.230 0.881 1.101 0.551 0.462 0.707
** ns *** ** *** *** ***
** * *** *** *** *** ***
68.75 a
3.391
***
d
In Situ Rumen Degradation Kinetics of Dry Matter (DM) Kd, %/h 7.90 b 8.10 b lag time (T0, h) 0.40 0.33 S fraction, % DM 20.72 b 7.56 d D fraction, % DM 68.15 b 82.39 a U fraction, % DM 11.13 a 10.07 ab RDDM, % DM 59.40 b 54.77 c RUDM, % DM 40.60 b 45.16 a In Situ Rumen Degradation Kinetics of Crude Protein (CP)d Kd, %/h 10.67 a 9.38 b lag time (T0, h) 0.70 ab 0.94 ab S fraction, % CP 26.71 b 10.96 d D fraction, % CP 65.92 c 81.92 a U fraction, % CP 7.38 a 7.13 a RDP, % CP 68.85 b 60.90 c RUP, % CP 31.14 b 39.12 a Intestinal Digestibility of RUPe IDCP, % RUP 56.31 ab 51.05 b
67.38 a
**
a−d: means within a row with different letters differ at the P < 0.05 level. ns, not significant; *, P < 0.05, **; P < 0.01; ***, P < 0.001. SEM = standard error of the mean. dKd = the degradation rate of the D fraction (%/h); S = soluble fraction; D = rumen degradable fraction; U = rumen nondegradable fraction; RDDM, effective degradability of DM in the rumen; RUDM, rumen nondegraded DM; and RUP, rumen nondegraded protein. eIDCP, intestinal digestibility of CP. a
b
c
Crude Protein Subfractions. Results of the CNCPS CP subfractions of the brown-seeded and yellow-seeded flax are presented in Table 2. The contents of all CP subfractions varied (P < 0.05) among the four genotypes. The yellow-seeded lines had significantly (P < 0.01) higher contents of the nonprotein fractions, namely, the instantaneously solublilized (PA) fraction (10.88 vs 3.91% CP). In contrast, Theodoridou and Yu,11 found lower content of the PA fraction (0.24 vs 3.1%) in yellow-seeded canola meal compared to that in brown-seeded canola meal. Whereas, Xin et al.12 found no differences in the PA content of yellow- and brown-seeded carinata strains. Moreover, the yellow-seeded lines contained a high content of nonavailable (PC) fraction (7.80 vs 3.38% CP) compared to
in these seeds. Compared to brown-seeded varieties, the content of NDIP was lower (P < 0.01; 16.45 vs 11.44% CP) in the yellow-seeded lines, which is in line with literature findings.11,12 Among the yellow-seeded lines, the content of ADIP was higher (P < 0.05) in F08×425, whereas no significant differences were observed in the CP chemical profile among the brown-seeded varieties. Overall, the results of chemical profiles showed that the yellow-seeded lines had higher contents of EE and CP; however, the CP in the yellow-seeded lines may not be optimally utilized by dairy cows due to the relatively high proportion of rapidly degradable SP (data not shown) and nondigestible ADIP fractions. 6559
dx.doi.org/10.1021/jf501284a | J. Agric. Food Chem. 2014, 62, 6556−6564
Journal of Agricultural and Food Chemistry
Article
Table 4. Protein’s Inherent Molecular Structural Characteristics Revealed by FT/IR-ATR Molecular Spectroscopy: Comparison of Brown Seed Type vs Yellow Seed Typea brown seed type item
CDC Bethune b
contrastd
yellow seed type
CDC Sorrel
brown vs yellow
P valued
0.0004 0.0246 0.0393 0.0571 0.0133 0.0239
ns * * ** ** ***
*** *** ** *** *** ***
0.0004 0.0004 0.0023
ns ns ***
*** ** ***
FY11 140
F08×425
SEM
0.064 0.033 1.956 3.940 1.558 2.558
0.058 0.029 2.018 3.547 1.360 2.655
b c b bc b b
0.0565 b 0.048 b 1.184 b
c
−1 e
Protein Amide (ca. 1715−1482 cm ) amide I height 0.059 b amide II height 0.030 b amide I/II height ratio 1.975 b amide I area 3.639 b amide II area 1.365 b amide I/II area ratio 2.667 b Protein Structureb α-Helix and β-Sheetf α-helix (height) 0.0580 b β-sheet (height) 0.0480 b α-helix/β-sheet ratio 1.193 ab
0.054 0.024 2.296 3.447 1.074 3.240
c d a c c a
0.0515 c 0.0455 c 1.135 c
a a b a a c
0.0625 a 0.0515 a 1.120 a
a a−d: means within a row with different letters differ at the P < 0.05 level. bThe measurements are expressed in IR absorbance units. cSEM = standard error of the mean. dns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001. eProtein amide bands were measure at ca. 1712−1482 cm−1. fα-Helix and β-sheet bands were measured in the amide I region second derivative and Fourier self deconvolution spectra. The α-helix and βsheet were centered at 1,653 and 1,630 cm−1, respectively.
seeded varieties, the higher RUDM of CDC Sorrel was mainly caused by its markedly lower S-fraction (7.56 vs 20.72% CP) compared to that of CDC Bethune. Except for lag time, all ruminal CP degradation parameters varied (P < 0.05) between the brown-seeded and yellow-seeded genotypes. The brown-seeded varieties had a higher (P < 0.001) content of RUP compared to that of the yellow-seeded genotypes. The higher RUP content of brown-seeded varieties could be related to their lower S fraction and lower kd of the D fraction compared to those of the yellow-seeded genotypes. However, there were significant differences in the in situ CP degradation parameters within the yellow- and brown-seeded genotypes. Among the brown-seeded flax, CDC Sorrel had a higher (P < 0.05) content of RUP. The higher RUP content of CDC Sorrel could be related to its markedly lower S fraction and lower kd of the D fraction compared to those of CDC Bethune (Table 3). In the yellow-seeded lines, FY08×425 had a significantly higher content of RUP. As the kd was not different between the two genotypes, the higher RUP content of FY08×425 was mainly caused by its markedly lower S fraction and higher lag time. The RUP in the yellow flaxseed lines were more (P < 0.001) digestible in the small intestine compared to that in the brown flaxseed varieties. These results are consistent with earlier findings.11 The high intestinal digestibility of RUP in yellow-seeded genotypes was, however, not supported by its significantly higher content of PC subfraction. Previous studies have shown a strong negative correlation (r = −0.81) between the PC (ADIP) subfraction and the total tract digestibility of CP,35 and a strong positive correlation (r = 0.84) between the content of PC and the content of N excreted in the feces.36 Nevertheless, the (in vitro)-intestinal digestibility of RUP in the brown flaxseed varieties was close to the reported values of intestinal digestibility for the brown flaxseed measured with the mobile nylon bag technique.7 The lower digestibility of RUP in the brown flaxseed could be partly explained by the presence of condensed tannins in the brown seed coat.11,37,38 Condensed tannins can make bonds with protein as well as with digestive enzymes and can decrease protein digestibility.39,40 Overall, these results suggest that the CP in yellow-seeded lines was more extensively degraded in the rumen causing a lower supply of RUP to the small intestine compared to that in brown-
that in brown-seeded varieties. As a consequence of high PA and PC fractions, the true protein fraction of the yellow-seeded lines was markedly lower (P < 0.01; 81.31 vs 92.71% CP) than that in the brown-seeded varieties. The PA fraction very rapidly (200%/h)31 degrades to NH3-N in the rumen and often contributes to NH3-N losses in the environment. Whereas, the PC fraction neither degrades in the rumen nor is digested in the postruminal tract and excreted in feces to the environment.28 These results suggest that CP in the yellow flaxseed genotypes is not optimally digested and utilized in the gastrointestinal tract of dairy cows compared brown flaxseed varieties. However, the brown flaxseed varieties had a markedly higher (13.13 vs 3.65% CP) PB3 subfraction. The PB3 slowly (0.06 to 0.55%/h)29,31 degrades in the rumen, and a large fraction of PB3 bypasses the rumen and is digested in the small intestine. Indeed, the in situ results showed higher RUP (35.1 vs 29.2% CP) contents for the brown-seeded varieties. Among the brown-seeded varieties, CDC Sorrel had lower (P < 0.05) PB1 and higher (P < 0.05) PB2 and PB3 fractions (Table 2), suggesting that CP in CDC Sorrel will be more gradually broken down to NH3-N in the rumen and that a higher fraction of CP will bypass the rumen. This was confirmed by the in situ results, which showed high RUP values for CDC Sorrel (Table 3). Rumen Degradation and Intestinal Digestibility. Results on the ruminal DM and CP degradation kinetics and intestinal digestibility of RUP in the yellow-seeded lines and brown-seeded varieties are shown in Table 3. The yellowseeded lines had lower (P < 0.001) content of RUDM compared to that of the brown-seeded varieties. The D fraction of DM did not significantly change among the two seed coat types. Therefore, the lower content of RUDM of the yellowseeded lines could be related to its higher (P < 0.05) S fraction and kd of the D fraction. The lower proportion of fiber (hull), higher proportion of CP and EE,11,32 and higher kd of CP (Table 3) can all contribute to the lower RUDM of the yellow seed coat genotypes. Within the yellow seed coat genotypes, FY11140 had higher (P < 0.05) S fraction and lower (P < 0.05) D fraction; however, no significant differences were observed in the RUDM among the two yellow-seeded lines. In the brown flaxseed varieties, CDC Sorrel had higher (P < 0.05) RUDM. As the kd was not significantly different among the two brown6560
dx.doi.org/10.1021/jf501284a | J. Agric. Food Chem. 2014, 62, 6556−6564
Journal of Agricultural and Food Chemistry
Article
Table 5. Coefficient of Correlation (Pearson)a and Its Level of Significance between Protein Molecular Structural Characteristics and Nutrients Profile, Crude Protein Subfraction Profile, in Situ Degradation Kinetics, and Intestinal Availability in Yellow- and Brown-Seeded Flaxseed
seeded varieties. However, the RUP in the former seeds were more digestible in the small intestine. These findings also demonstrate that it is important to determine the intestinal digestibility of RUP in order to more accurately estimate the metabolizable protein supply to dairy cattle. Protein’s Molecular Structures. Recently, advance molecular spectroscopy methods have been developed to quantitatively evaluate the primary and secondary molecular makeups of feed protein.31,41 In general, the amide I and amide II spectral intensities in terms of peak area and height reveal information about the concentration of protein in feed.18,42 Table 4 shows that the yellow seed coat FY11140 line had greater (P < 0.05) amide I and II height and amide I and II peak area among the four genotypes, which correspond to the higher concentration of CP in this variety. In agreement with our findings, Dariman and Yu42 found that greater amide I and II height and area in zero-amylose waxy cultivar of barley were associated with a higher content of CP compared to those of waxy, high amylose and normal cultivars. However, the ratios of amide I/II and α-helix/β-sheet reveal information about the protein’s molecular makeup.42,43 The molecular makeup of feed protein strongly influences CP degradation in the rumen and the digestibility of RUP by affecting their solubility and access to microbes and proteolytic enzymes.16,31 Among the four genotypes, CDC Sorrel had the highest amide I/II area ratio and lowest α-helix/β-sheet ratio, which were associated with the lowest contents of PA and PB1, and the highest content of RUP. Our results are consistent with the findings of Samadi and Yu15 and Peng et al.18 The correlation analysis (Table 5) showed that the amide I/II area ratio had a significant (P < 0.05) negative correlation to SP (r = −0.77) and the in situ S fraction (r = −0.94) and significant positive correlations to the D fraction (r = 0.88), RUP (r = 0.96), and intestinal digestibility of RUP (r = 0.63). The α-helix/β-sheet ratio had significant (P < 0.05) positive correlation to the content of SP (r = 0.85), kd (r = −0.82), S fraction (r = 0.91), and RDP (r = −0.92), and had significant negative correlations to the D fraction (r = −0.88), RUP (r = −0.92), and intestinal digestibility of RUP (−0.71). These results demonstrate that there is a strong relationship between the protein’s molecular structures and CP chemical profiles, ruminal degradation characteristics, and intestinal digestibility of RUP. Multivariate Analysis. Figure 2 shows the results of multivariate CLA and PCA analyses that classify and visualize the overall differences in the protein’s inherent molecular makeup of brown-seeded and yellow-seeded genotypes. The CLA dendrograms (Figure 2a) show that the protein’s molecular spectral data of CDC Sorrel can be distinguished from the other genotypes at a linkage distance of
