Characterization of physiochemical and nutrient profiles in canola feedstocks and co-products from bio-oil processing: impacted by source origin

Objective The objective of this study was to characterize physiochemical and nutrient profiles of feedstock and co-products from canola bio-oil processing that were impacted by source origin. The feedstocks and co-products (mash, pellet) were randomly collected from five different bio-oil processing plants with five different batches of samples in each bio-processing plant in Canada (CA) and China (CH). Methods The detailed chemical composition, energy profile, total digestible nutrient (TDN), protein and carbohydrate subfractions, and their degradation and digestion (CNCPS6.5) were determined. Results The results showed that TDN1x was similar in meals between CA and CH. CH meals and feedstock had higher, truly digestible crude protein (tdCP) and neutral detergent fiber (tdNDF) than CA while CA had higher truly digestible non-fiber carbohydrate (tdNFC). The metabolizable energy (ME3x), net energy (NELp3x, NEm3x, and NEg3x) were similar in meals between CA and CH. No differences were observed in energy profile of seeds between CA and CH. The protein and carbohydrate subfractions of seeds within CH were similar. The results also showed that pelleting of meals affected protein sub-fractionation of CA meals, except rapidly degradable fractions (PB1), rumen degradable (RDPB1) and undegrdable PB1 (RUPB1), and intestinal digestible PB1 (DIGPB1). Canola meals were different in the soluble (PA2) and slowly degradable fractions (PB2) between CA and CH. The carbohydrate fractions of intermediately degradable fraction (CB2), slowly degradable fraction (CB3), and undegradable fraction (CC) were different among CH meals. CH presented higher soluble carbohydrate (CA4) and lower CB2, and CC than CA meals. Conclusion The results indicated that although the seeds were similar within and between CA and CH, either oil-extraction process or meal pelleting seemed to have generated significantly different aspects in physiochemical and nutrient profiles in the meals. Nutritionists and producers need to regularly check nutritional value of meal mash and pellets for precision feeding.


INTRODUCTION
Canola has been produced in Western Canada since 1974, when it was developed as a low erucic acid and low glucosinolate rapeseed, to supply for the high demand of cooking oil [1]. When canola oil is extracted, it generates a co-product low in fat and rich in protein. This co-product, canola meal, is mainly used in dairy rations because its amino acid profile is ideal for milk synthesis [2].
Due to the high production of canola and the high global demand, besides being extensively used in Canada, it is also exported to many countries. China is one of the main SEM  from China; however, the samples from plants 3 and 4 from Canada that were pelleted presented EE of 1.46 and 1.06% DM respectively, which can be associated with the coating of the pellets with oil, but this higher EE was not observed on the pellets from plant 5 (0.63% DM). Soluble crude protein and neutral detergent indigestible crude protein (NDICP or NDIP) were different between Canada and China. While China presented higher CP (p = 0.003) and SCP (p<0.001), Canada presented higher NDICP (p<0.001). Acid detergent insoluble crude protein (ADICP) was not significantly different (p = 0.075, Table 1).
Mustafa et al [20] stated that the NDICP of regular canola meal was 105 g/kg CP which is lower when compared to meals from Canada that averaged 19.34% CP, but close to the samples from plants A and C from China (11.07% and 10.83% CP), however, still lower than China's average of 13.52% CP. They also reported ADICP as 45 g/kg CP was lower than this project's meals (Canada [5.53% CP] and China [5.80% CP]).
According to Newkirk [4], different cultivars, canola growth environments and harvest, and the processes the seeds and meals go through can all affect the final nutrient profile of the meal. Since five different companies were sampled in the production of different batches of meals both in Canada and in China, it is safe to assume that these results are representative of the companies and their quality is steady through different batches, and small variations are expected due to the variability of crop conditions, cultivars, and harvest.
The chemical profile of the canola seeds studied on this project is displayed in Tables 3 and 4. The DM of seeds from Canadian plants was higher than those from Chinese plants (p = 0.008). Crude protein content was similar (p = 0.100, Canada vs China). Soluble CP was higher for China plants (p = 0.002). And NDICP was higher for Canada plants (p<0.001). Neutral detergent fiber, ADF and cellulose were higher for Canada plants (p = 0.004, p = 0.003, and p<0.001, respectively), while ADL was higher for the China plants (p = 0.017).
The Canadian Grain Commission (CGC) [24] summarized the canola seed production of 2020 and observed an oil content of 44.1% DM and CP of 20.8%. On their report from the 2015 [25] production, they observed seeds with 44.2% DM EE and 20.7% DM of CP. These results give us a basis to safely assume that Canada produces canola with a high and stable along the years.
Burbulis and Kott [26] investigated the variation in color and oil content influenced by the environmental temperature on black-seeded spring rapeseed varieties Brassica napus L. 'Bolero' (owned by Raps GbR) and 'Star' (owned by Dansk Planteforaedling/DLF) and 11 lines originated from their crossing. They found that temperatures higher than 28°C during the day, resulted in offspring with lighter seeds (more yellow) and temperatures lower than 20°C resulted in darker seeds (more brown or black). They also observed differences in oil content on the seeds from different environments. The oil content of the darker seeds (colder climate) ranged from 31.2% to 51.6% DM, and lighter seeds (warmer climate) ranged from 31.4% to 49.4% DM. On average, lighter seeds presented lower oil content.
Tramontini [23] likely used canola seeds from a different climate, since her study was conducted in Brazil, a tropical country with higher temperatures, as Burbulis and Kott [26] study suggests, the higher temperatures in that country could have influenced the seeds she used, explaining the lower EE content. The seeds analyzed on our project, however, were in accordance with the standard quality of the Canadian canola seeds.
The higher cellulose content on the Canada plants (p< 0.001) could have been the cause for higher contents of NDF (p = 0.004), ADF (p = 0.003), and NDICP (p<0.001) on the samples from that country.

Total digestible nutrients and digestible (DE), metabolizable (ME), and net energy (NE) values of feedstocks and co-products: comparison among bio-oil processing plants and between two countries
The energetic profile of canola meals and pellets are represented in Tables 5 and 6. Total digestible NDF (tdNDF), total digestible CP (tdCP), and total digestible nutrients (TDN 1x ) were different among Canadian plants (p<0.001, p = 0.001, and p = 0.001, respectively). The contrast indicated that the Table 5. Energy profile of co-products from different oil processing plants (canola meals and pellets): comparison among bio-oil processing plants and between Canada and China

Items
Digestible nutrients profile (% DM)       For each plant, sample size n = 5. DM, dry matter; ME 3x , metabolizable energy for gain at three times the maintenance level; NE Lp3x , net energy for lactation at a productive level of intake three times the maintenance level; NE m3x , net energy for maintenance; NE g3x , net energy for gain; M, meal; P, pellet; CA, Canada; CH, China; SEM, standard error of the mean. a-c Means within a column without a common superscript letter differ (p < 0.05). meals pelleted (Plants 3, 4, and 5) resulted in higher tdNDF and TDN 1x (p<0.001, and p = 0.002) than the mash. When pelleting, it is common practice to add back to the process fines collected during the screening step and that might have contributed to a lightly higher tdNDF in this study. Also, as a final step of pelleting, there is the spraying of oil to increase the durability of the pellet, which might have been the cause for a slightly higher TDN 1x on Plant 3. tdNDF and tdCP were also variable among the meals from Chinese plants (p<0.001 and p = 0.002). When analyzing the overall meals from Canada and China, it was observed that tdNDF, tdNFC, and tdCP were different (p<0.001, p = 0.006, and p<0.001), of these, Canada had higher tdNFC, while China presented higher tdNDF and tdCP. Metabolizable energy at three times maintenance (ME 3x ), net energy for lactation (NE Lp3x ), maintenance (NE M3x ), and gain (NE g3x ) were all observed to be different among the meals from the Canadian plants (p<0.001, for all of them). Differences between mash and pelleted meals were also observed of these parameters (p<0.05) with the Plant 3 showing the higher results. While these differences were present on the Canadian samples, no differences were observed on the Chinese samples. Moreover, the overall comparison of canola meals from Canada and China showed that they are similar. Damiran [28] analyzed canola meals from yellow and brown canola seeds and showed some differences in their energy profiles. Therefore, the higher TDN (71.5%) on Damiran et al [27] might be explained by that canola meal being from a yellow seeded cultivar or as a consequence of the higher fat and protein content of that meal, since the TDN value is based on the values of digestible carbohydrates, protein and fat of a feedstuff [29].
The energy profile of canola seeds is displayed in Tables 7  and 8. As expected, the seeds presented less variations. No differences were observed on the digestible nutrients profile from Canadian plants. Only the tdCP of canola seeds from the Chinese companies were different in this study (p = 0.006). This might be due to the varieties difference. The overall comparison of the energetic parameters of canola seeds from Canada and China only the tdNDF from Canadian plants were higher (p = 0.023), while all the other parameters were Table 7. Energy profile of canola seeds from different oil processing plants: comparison among bio-oil processing plants and between Canada (CA) and China (CH)
While the rumen degradable fractions of PA2, PEP, and total RDP, and the rumen undegradable fraction of PA2 were      higher in the Chinese meals (p<0.001), the rumen degradable PB2, rumen undegradable PB2, and intestinal digestible PB2 and FP fractions were higher for the Canadian meals. Higher availability of protein in the rumen (degradable fractions) guarantees enough amino acid supply for the rumen microbiota, however higher availability of protein for intestinal digestion and absorption (intestinal digestible fractions) means that a higher variability of amino acids will be available for the animal to use for muscle deposition and milk production. The protein fractions of the canola seeds analyzed in this study are represented in Table 10. The Canadian seeds presented some variation on the contents of PB2, PC, and TP fractions (p<0.001, for all). The Canadian Plant 2 had the highest content of PB2, while Plant 5 presented the lowest. Plant 4 showed higher content of PC and lower content of TP. The opposite was observed on Plant 3 that showed the lowest PC and the highest TP. All the seeds from the five different Chinese companies were similar for all protein fractions presented. Only the slowly degradable fraction (PB2) was different between Canada and China (p<0.001), where Canadian seeds presented higher amounts of this fraction.
The rumen and intestinal fractions are presented in Table  16, where we see a similar ruminal degradation and intestinal digestion profile. The RDPB2, RUPB2, RUPC, and DIGPB2 are different among Canadian plants (all p<0.001). No difference is observed among the seeds from the various Chinese plants, and RDPB2, RUPB2, and DIGPB2 are higher in the seeds from Canada (p<0.001).
The carbohydrate fractions of canola meals and pellets are given in Table 11. Canadian canola meals different among the five plants for digestible (CB3) (p = 0.002) and indigestible fiber (CC) (p<0.001). Plant 4 showed the lowest amount of digestible fiber (CB3) and the highest of indigestible fiber (CC). Plant 5 displayed the highest content of CB3 and Plant 3 the lowest amount of CC. Only the CC fraction showed a difference between the mash and pelleted meals (p<0.001). The Chinese meals presented variability among companies on the CB2, CB3, and CC fractions (p = 0.012, p = 0.013,      Plant D resulted in the highest contents of RDCB3, RUCB3, and DIGCB3 (p = 0.007, for all). The rumen degradable, undegradable, and intestinal digestible CB2, the RUCC, and total RUC fractions of canola meals were higher in the Canadian companies (p = 0.009, p = 0.008, p = 0.008, p<0.001, and p = 0.009, respectively). Table 12 presents the carbohydrate fractions of canola seeds from Canadian and Chinese companies. Only the CB2 and CC fractions seemed to be different among companies (p = 0.002 and p<0.001, respectively), where Plant 3 showed the lowest values for both. All the samples analyzed from the five Chinese samples were similar. Only the amounts of watersoluble CHO (CA4) and digestible fiber (CB3) differed between countries (p = 0.022 and p = 0.006). Table 14 shows the predicted amounts of rumen degradable and undegradable and intestinal digestible carbohydrate fractions of canola seeds. This table shows that while Plant 5 exhibited the highest values of rumen degradable, undegradable, and intestinal digestible CB2, and total RDC, the Plant 3 exhibited the lowest values for those variables (p = 0.003, p = 0.003, p = 0.003, and p = 0.020, respectively). Apart from DIGFC (p = 0.043), all other variables analyzed on the Chinese canola seeds were similar. And excluding the CB3                   fractions (RDCB3, RUCB3, and DIGCB3; p = 0.006 for these three), all other fractions are similar between the canola seeds analyzed from Canadian and Chinese companies. Huang [31] reported a study on different temperatures and conditioning time during the pelleting of canola meals and showed that neither the carbohydrate fractions nor the predicted rumen degradable and undegradable carbohydrate fractions were affected by the different treatments. This finding is in accordance with our results because only the indigestible fiber fractions (CC, RUCC, and total RUC) expressed a difference between mash and pellets (p<0.001, for the three fractions).

Summary and conclusion
Summary: The chemical profile of canola meals from Canada and China presented significant differences on DM, ash, CP, SCP, and NDICP. Whereas the chemical profile of canola seeds from Canada and China presented differences on DM, SCP, NDICP, NDF, AF, ADL, and cellulose. Because variations can be caused by crop environment, cultivar, and processing, these differences do not seem relevant.
The pelleting of canola meals by the Canadian companies seemed to have influenced tdNDF and TDN 1x . On the other hand, the meals from China were not pelleted and differences were observed on tdNDF and tdCP. On the overall comparison of the mash meals, China presented higher tdNDF, and tdCP, and lower tdNFC than Canada.
The energy values of canola seeds were very similar among companies on Canada and China except for tdCP on the Chinese samples that showed some variations among plants. Between countries, only tdNDF was higher in Canada. No differences were observed on the energy values (NE Lp3x , NE m3x , and NE g3x ) of canola seeds from China or Canada.
The protein fractions of the canola meals from Canada and China were similar, except for PA2 and PB2, where PA2 was higher in China and PB2 in Canada. The content of PB2 was also higher for the Canadian seeds. RDPA2, RUPA2, RDPEP, and total RDP were higher on the Chinese meals, whereas RDPB2, RUPB2, DIGPB2, and DIGPF were higher on the Canadian meals. While the Chinese seeds presented higher amounts of RDPB2, RUPB2, and DIGPB2.
The Chinese meals and seeds showed higher content of water-soluble carbohydrates (CA4). Canadian meals presented higher soluble (CB2) and indigestible (CC) fiber contents, and consequently higher RDCB2, RUCB2, RUCC, and DIGCB2 than the ones from China. The meals from Canada were also higher in RUCC and Total RUC. While the rumen degradable, undegradable and intestinal digestible