CHEMICAL COMPOSITION AND NUTRITIVE VALUE OF

CANOLA-QUALITY SINAPIS ALBA MUSTARD

Bogdan A. Slominski ¹, Heather D. Kienzle ¹, Ping Jiang ¹,

Lloyd D. Campbell ¹, Mark Pickard ² and Gerhard Rakow ³

¹ Department of Animal Science, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2; e-mail address: B_Slominski@UManitoba.ca

² InfraReady Products Ltd, 850C 56^th St.E, Saskatoon, Saskatchewan,

Canada S7K 5Y8

³ Agriculture and Agri-Food Canada Research Station, Saskatoon,

Saskatchewan, Canada S7N 0X2

ABSTRACT

            The Sinapis alba (yellow mustard) species has potential as a high protein and high energy alternative to full fat soybean.  In comparison to full fat soybean, low-glucosinolate, low-erucic acid S. alba seed was found to contain more oil (26.4% vs 20.2%), less protein (37.5% vs 41.4%), more methionine and cystine (3.60 vs 3.33 g/16g N), less lysine (5.78 vs 6.49 g/16g N) and similar amount of carbohydrate (ie., sucrose, starch)(4.6% vs 5.2%).  Considering the remarkable difference in seed size, yellow mustard having a 20-fold smaller seed than soybean, only a small difference in total dietary fibre content was observed (22.2% vs 18.8%).  A lower content of oligosaccharides (3.4% vs 5.1%) and higher contents of calcium (0.66% vs 0.39%) and available (non-phytate) phosphorus (0.27% vs 0.12%) were among other positive characteristics associated with the S. alba crop. Micronization of the S. alba seed decreased glucosinolate content, inactivated myrosinase enzyme, and increased in vitro digestible protein content.  Micronized S alba seed showed a higher TME_n (true metabolizable energy) value than the raw sample (2460 vs 2270 kcal/kg). The highest TME_n value (3790 kcal/kg) was observed for micronized soybean seed. Broiler chickens fed rations containing 15% of raw and micronized S. alba seeds showed smaller weight gain and higher feed to gain ratios to those fed micronized soybean seed. Supplementation of S. alba diets with mucilage (soluble fiber) depolymerizing enzymes reduced digesta viscosity, increased energy utilization and improved broiler chicken performance. Further improvement in energy and amino acid utilization by poultry was observed when the seed was crushed to a small particle size (#0.6 mm).

KEYWORDS:  Yellow mustard, glucosinolates, mucilage, available energy, poultry

INTRODUCTION

            It has been recognized by our group that the canola-quality yellow mustard (Sinapis alba L.) has potential as a high protein and high energy alternative to full fat soybean.   S.  alba has superior heat and drought tolerance in comparison to conventional  Brassica napus and B. rapa canola, and is therefore well suited to production in dryland areas.  It has a high resistance to blackleg disease (Lestophaeria maculans) and flea beetle attacks.  Further advantages include a high shatter-resistant seed pod and a large bright-yellow seed.  Although it has been grown for sometime as a condiment crop, it is only recently that plant breeders have developed a “canola quality” cultivar with low contents of glucosinolates and erucic acid (Krzymanski et al., 1991; Raney et al., 1995).

            The objective of the present study was to gather detailed information on the chemical composition of S. alba seed, to investigate any beneficial effects of micronization on seed quality and to examine the effect of soluble fiber (mucilage) and fineness of seed grinding on energy utilization by poultry.

EXPERIMENTAL

            Chemical composition of S. alba is shown in Table 1.  In comparison to full fat soybean, S. alba contained more oil and less protein but the content of these two major nutrients was greater for the S. alba sample (63.9% vs 61.6).  S. alba seed contained slightly more methionine and cystine but less lysine than soybean. The sucrose content was lower in S. alba although this was offset by the presence of starch which was not detected in soybean.  Considering the remarkable difference in seed size, S. alba having a 20-fold smaller seed than soybean (i.e., 6.5 vs 138 g/1000 seeds), only a small difference in the total fiber content was observed.  A lower content of oligosaccharides and higher contents of calcium and available (non-phytate) phosphorus were among other positive characteristics associated with the canola-quality S. alba.

Table 1. Chemical composition of Sinapis alba yellow seed as compared to full fat soybean and Brassica napus canola (% DM unless otherwise specified)  

Component	S. alba	Soybean	Canola
Protein:        - Lysine (g/16 g N)        - Sulphur amino acids (g/16 g N) Oil Sucrose Oligosaccharides^1 Starch Total fiber:      - Non-starch polysaccharides      - Lignin and others^2 Ash Calcium Phytate phosphorus Non-phytate phosphorus	37.5   5.8   3.6 26.4   3.3   3.4   1.3 22.2 15.3   6.9   5.4     0.66     0.62     0.27	41.4   6.5   3.3 20.2   5.2   5.1 nd^3 18.8 14.7   4.1   5.4     0.39     0.51     0.12	25.7   6.2   4.5 41.5   4.1   2.0   0.4 20.4 11.8   8.6   5.0     0.46     0.55     0.34
^1Includes raffinose and stachyose; ^2  Includes lignin with associated polyphenols, cell wall protein and minerals associated with the fiber fraction; ^3 Not detected

            Although still relatively high and potentialy a limiting factor, the glucosinolate content of the new S. alba cultivar was shown to be 35.3 Fmol/g, a level substantially lower than that present in the current condiment varieties (ie., ~200 Fmol/g; Hemingway, 1995) (Table 2). The glucosinolate content of S. alba may be influenced by the internal standard used in the analysis and as shown in Table 2 the glucosinolate content averaged 7.7 Fmol/g when benzyl glucosinolate (glucotropaeolin) was used as the internal standard.  Similar low values were reported by Raney et al. (1995) (13.5 Fmol/g) and Uppstrom (1995) (6.0 Fmol/g).  In the results from our laboratory, careful examination of the gas-liquid chromatograms indicated that the peak of benzyl glucosinolate, the internal standard commonly used in glucosinolate analyses, was larger than that normally seen in typical analyses performed on canola meal.  When the internal standard was changed to allyl glucosinolate (sinigrin) it was evident that the sample of S. alba contained a significant amount of benzyl glucosinolate and as a consequence the use of sinigrin as an internal standard resulted in much higher glucosinolate content than that originally determined (Table 2).  This finding was later confirmed by the laboratories of Agriculture Canada in Saskatoon and Plant Breeding Institute in Poland (Rakow and Krzymanski, personal communications).

Table 2.  Glucosinolate content of S. alba as influenced by internal standard (IS) used in the anaysis (µmol/g oil free meal)

Glucosinolate	Benzyl glucosinolate as IS	Allyl glucosinolate as IS
3-butenyl 4-pentenyl 2-OH-3-butenyl 2-OH-4-pentenyl benzyl 4-OH-benzyl 3-indolylmethyl 4-OH-3-indolylmethyl	0.1 nd^1 4.1 nd - 0.9 1.0 1.7	0.4  nd 12.1  nd 12.0   2.7   3.4   4.9
Total	7.7	35.3
^1 Not detected

            In subsequent studies, the S. alba seed was subjected to micronization (140EC) and autoclaving (108EC) to investigate the effect of heat treatment on quality parameters including available energy content and nutritive value of the seed.  Heat treatment of the seed decreased glucosinolate content, effectively inactivated the myrosinase enzyme, and increased digestible protein content in vitro (data not shown).  However, regardless of heat treatment employed, the true metabolizable energy (TME_n) content of S. alba seed was lower than that of micronized soybean seed (Table 3).

Table 3.  True metabolizable energy content (TME_n) of variously treated S. alba seeds in comparison to micronized soybean seed

Sample	TMEn (kcal/kg DM)
Raw S. alba seed Micronized S. alba seed Autoclaved S. alba seed Micronized soybean seed	2270^d 2460^c 3031^b 3790^a

^abcd Values within columns with no common superscripts differ significantly (P#0.05)

            The TME_n results were confirmed in a broiler chicken trial in which micronized and autoclaved S. alba seeds showed lower available energy content (AME_n)  than micronized soybean seed (2574, 2713 and 2979 kcal/kg, respectively).  It was surmised  that the relatively high soluble fiber (mucilage) content of S. alba seed (3.4%) along with insufficient disruption of the seed structure on grinding may have contributed to the low energy availability.  A viscosity-reducing microbial enzyme was employed to investigate the effect of S. alba mucilage on broiler chicken performance and fat and energy utilization (Table 4).  As compared to a raw S.alba seed diet, two enzyme supplemented diets showed significantly better feed to gain ratios indicating that viscosity reduction may have had a positive effect on nutrient utilization.  This was substantiated to some degree by the intestinal viscosity values although  the difference was not statistically significant.   The two enzyme supplemented diets had higher AME_n contents and showed a statistically significant improvement in fat digestibility. 

Table 4.  The effect of enzyme supplementation on body weight gain (BWG), feed conversion ratio (FCR), intestinal viscosity, fat digestibility and diet energy content (AME_n) for broiler chickens (5-19 days-of-age) fed raw and micronized S. alba seeds

Treatment	BWG g/bird	FCR	Viscosity Cps	Fat dig. %	AMEn kcal/kg
Control Micronized soybean seed Raw S.alba seed Raw S.alba seed + Enzyme^1 Micronized S.alba seed + Enzyme^1	496^a 456^b 425^c  448^bc  446^bc	1.44^d 1.50^c 1.58^a  1.52^cb 1.54^b	4.9^b 5.2^b 8.2^a  6.6^ab  6.8^ab	84.7^a 78.5^b 53.8^e 59.1^d  69.5^c	3047^a 3027^a 2728^c  2778^bc 2882^b

¹ Carbohydrase enzyme added at 0.01%; ^abcd Values within columns with no common superscripts differ

   significantly (P#0.05)

            The difference in fat and energy utilization by broiler chickens fed raw and micronized seeds supplemented with the same enzyme indicated that the low available energy content of S. alba seed may have resulted from an ineffective crushing of the seed and insufficient disruption of the oil containing cells in the gizzard with part of the oil subsequently escaping digestion in the small intestine.  To investigate this hypothesis further, both raw and micronized S alba seeds were subjected to two grinding procedures that produced seed meals with different fineness of grind.  A distinct increase in TME_n content was observed when the particle size of the seed meal decreased from #2.0 mm to #0.6 mm (Table 5).  In addition, the overall higher TME_n values for micronized seed, regardless of grinding procedure,  indicated that heat treatment may also have facilitated the effectiveness of grinding.

Table 5.  Effect of particle size on true metabolizable energy content (TME_n) of Sinapis alba seed

Sample	Particle size	TMEn (kcal / kg)
Raw seed (fine) Raw seed (coarse) Micronized seed (fine) Micronized seed (coarse)	# 0.6 mm # 2.0 mm # 0.6 mm # 2.0 mm	4024^b 2965^d 4242^a 3464 ^c

¹ Mean ± SD;   ^a,b,c,d Values within a column with different superscripts differ significantly (P#0.05).

CONCLUSIONS

A potential for use of canola-quality yellow mustard (Sinapis alba L.) as a valuable alternative to soybean in poultry nutrition is indicated by the results of the current study. Seed micronization and fine grinding appear essential for optimal nutrient utilization.

REFERENCES

Hemingway, J.S., 1995. The mustard species: condiment and food ingredient use and potential as oilseed crops. In: Brassica Oilseeds: production and utilization. Komber, D.S and D.I. McGregor(editors). CAB International. Wallingford, UK, pp. 373-383

Krzymanski, J., T. Pietka, T. Ratajska, I. Byczynska, and K. Krotka, 1991. Development of low glucosinolate white mustard (Sinapis alba syn Brassica hirta).  Proceedings of the 8^th International Rapeseed Congress, Saskatoon, Canada, pp. 1545-1548.

Raney, P., G. Rakow and T. Olson, 1991. Development of low erucic acid, low glucosinolate Sinapis alba.  Proceedings of the 9^th International Rapeseed Congress. Cambridge, UK, pp. 416-418

Uppstrom, B, 1995. Seed Chemistry. In: Brassica oilseeds: production and utilization. Kimber, D.S. and D.I. McGregor (editors). CAB International. Wallingford, UK, pp. 217-242[i1] .

---

[i1]