Slide 1 Hello. My name is Neil Jaworski, and I'm going to present some research I have done as a part of my Master's thesis, working in the Stein Monogastric Nutrition Laboratory here at the University of Illinois at Urbana-Champaign. The title of my presentation is "Carbohydrate Composition and In Vitro Digestibility of Dry Matter and Non-starch Polysaccharides in Grains and Grain Coproducts." Slide 2 This is the outline I will follow today. Starting with a brief introduction, then I'm going into Experiment 1, which was ingredient composition, and we did starch analysis and non-starch polysaccharide (or NSP, as I'll refer to it the rest of the presentation) analysis. Then Experiment 2, which was an in vitro digestibility of dry matter and NSP, then I'll give you some conclusions and some take home messages. Slide 3 So why do we feed grains and grain coproducts to pigs? Well, being in the Corn Belt of the United States, they are readily available, especially corn. And in the past, these grains have been relatively inexpensive and very economical to feed to pigs. But currently, as we all know, the price of grains has increased due to ethanol production and this year's drought. So they're uneconomical to feed to pigs. And so we're replacing some of these grains with grain coproducts to bring the price of the diets down. And both grains and grain coproducts are high in carbohydrates, and these carbohydrates supply the majority of energy in the diet. Slide 4 So what are some of these carbohydrates in grains and grain coproducts? Well, we see that sugars and oligosaccharides are both low in grains and grain coproducts. Whereas starch is high in grains, but lower in grain coproducts. And non-starch polysaccharides and lignin are low to intermediate in grains, and high in grain co-products. For the remainder of this presentation, I'll be focused on starch and non-starch polysaccharides because of the difference between grains and grain coproducts. Slide 5 Starch is composed of amylose, which are linear chains of glucose monosaccharides joined by alpha 1,4 bonds, and amylopectins, which are linear chains of glucose joined by alpha 1,4 bonds but every 20 to 25 glucose monomers, there are alpha 1,6 linkages which are branch points. Which you can see in the picture in the bottom right. Because starch is composed of the alpha 1,4 and alpha 1,6 linkages, it is enzymatically digested by the pig. Because the pig secretes salivary amylase and pancreatic amylase which digests the starch molecule, producing glucose as the end product of digestion, which we all know is readily absorbed by the pig and utilized. Slide 6 Non-starch polysaccharides, or NSP, are cellulose, beta-glucan, arabinoxylan, pectins, and there's many others. And these are multiple monosaccharides joined by beta linkages. And this is the difference between starch and non-starch polysaccharides. These beta linkages are not digested by the pig's enzymes, and therefore, non-starch polysaccharides move from the small intestine to the large intestine, where they're fermented. And the fermentation varies on the amount and type of NSP and the age of the pig, among other things. And here's a picture of arabinoxylan, which you can see is a linear chain of xylose units joined by beta 1,4 linkages, and it's randomly substituted by arabinose units. And those arabinose units are joined by beta linkages, making arabinoxylans not digested by the pig's endogenous enzymes, and therefore it becomes fermented. Slide 7 So, to recap starch versus non-starch polysaccharides, starch is enzymatically digested by the pig's endogenous enzymes. So it is 100% oxidized to glucose. And this is a very efficient process. And then we all know that glucose is readily absorbed and can be utilized in the pig for ATP production, lipid synthesis, amino acid synthesis, and generating reducing equivalents through the pentose phosphate pathway. Whereas NSP, they are not enzymatically digested, but fermented. And the end products of fermentation are volatile fatty acids, which can be used for gluconeogenesis, lipid synthesis, and an energy source to colonocytes in the pig. But also, fermentation produces water, which is reabsorbed by the pig; carbon dioxide, which is exhaled through the lungs; and methane, which comes out the other end. And both carbon dioxide and methane result in losses of energy to the pig. Therefore, NSP are a less efficient means of obtaining energy because they're fermented rather than being enzymatically digested like starch. Slide 8 So, because of the energetic differences between starch and non-starch polysaccharides, the objective of Experiment 1 was to quantify the amount of starch and NSP in grains and grain co-products used as swine feed ingredients. And to do this, I had the opportunity to go to Dr. Stein's home country of Denmark, and work with Dr. Bach Knudsen and Dr. Lærke at Aarhus University and do some carbohydrate analysis with them. Slide 9 We chose to collaborate with Dr. Bach Knudsen because of the type of fiber analysis he does. And on this slide, you can see there are many different methods of fiber analysis. And we see on the left, fiber can be composed of lignin, cellulose, insoluble hemicellulose, and soluble hemicellulose. And you can see that crude fiber is an incomplete measurement of lignin, cellulose, and insoluble hemicellulose. ADF measures cellulose and lignin. Neutral detergent fiber measures the insoluble hemicellulose, cellulose, and lignin. And the total dietary fiber procedure measures the insoluble dietary fiber – which is the insoluble hemicellulose, cellulose, and lignin – and the soluble dietary fiber, which is the soluble hemicellulose. And finally, the procedure on the far right is the procedure that Dr. Bach Knudsen does, and he splits the fiber up into all its individual components. So Klason lignin measures the lignin, cellulose is measured as cellulose, insoluble hemicellulose is measured as insoluble non-cellulosic polysaccharides, and soluble hemicellulose is measured as soluble NSP. So we chose to do this because you get the most complete analysis of the fiber. Slide 10 So the materials of Experiment 1 were 12 ingredient samples from the United States. We used corn, four corn co-products from the wet milling industry: corn gluten feed, corn germ meal, corn bran, and corn gluten meal; and one co-product from the dry grind industry, which was distillers dried grains with solubles, or DDGS. Slide 11 Then we had a sorghum sample, and two sorghum DDGS samples. One sample from Kansas, and one from an ethanol plant in Texas. And the sorghum DDGS from Kansas had greater solubles added to it. Slide 12 And we had wheat and two wheat coproducts from the flour milling industry: wheat middlings and wheat bran. Slide 13 And with these 12 swine feed ingredients, we analyzed them for starch using an enzymatic colorimetric procedure, and then analyzed them for non-starch polysaccharides. And the neutral NSP were determined by gas chromatography, and the uronic acids were measured colorimetrically. Slide 14 Going into starch analysis, you begin with a 150 mg sample, and you boil it at 100 °C to gelatinize the starch and begin to leech some of the amylose from the starch molecules. And you incubate this with an alpha-amylase enzyme to begin degrading the starch, at a pH of 5 for one hour. Then, you're left with glucooligosaccharides. And these are further broken down into glucose by incubating with amyloglucosidase at a pH of 5 at 60 °C for two hours. Then, the glucose – you centrifuge, dilute, and measure the glucose with glucose oxidase, or GOD. And GOD has the power to turn your samples a certain shade of purple depending on the amount of glucose in your sample. And then you take your sample and measure the absorbance with a spectrophotometer. And in the bottom left-hand picture, you can see the standard curve of absorbance and known glucose values. And using this, you get the absorbance from your sample, and using this standard curve, you can calculate the amount of glucose in your sample. And then, to convert your glucose to starch, you multiply by 0.9 to compensate for the amount of water that's inside a starch polysaccharide. Slide 15 Now, moving into NSP analysis, which is a three-day procedure, and it has three parallel runs. A run on the left, for total NSP, in the middle, for non-cellulosic polysaccharides, and on the right, for insoluble non-starch polysaccharides. And the first step for all three procedures is starch gelatinization and hydrolysis. Which is the exact same as the starch procedure that I had described previously on the last slide. Then, after the starch is gelatinized and hydrolyzed, the total NSP and the non-cellulosic polysaccharide procedure, you precipitate the soluble NSP with 80% ethanol, accounting for the soluble NSP. Whereas in the insoluble non-starch polysaccharide procedure, you remove the soluble NSP with a neutral phosphate buffer. Next, you centrifuge, and then you remove the supernatant. And the supernatant contains all your starch and low molecular weight carbohydrates, such as sucrose and lactose, and also oligosaccharides like raffinose and stachyose. And so you're left with the residue at the end of Day 1, which is composed of fiber. Slide 16 Now for Day 2, the total NSP procedure on the left, and the insoluble NSP procedure on the right follow the same steps. Where you first add 12 M sulfuric acid to swell the cellulose and begin to break down all the cellulose and non-starch polysaccharides to their monosaccharides. Whereas the non-cellulosic polysaccharide procedure, you just hydrolyze with 2 M sulfuric acid, and in this procedure, you are not swelling the cellulose because you do not want to account for the cellulose in this procedure. Then, the total NSP and insoluble NSP are filtered through Gooch crucibles, and the residue that's remaining is known as Klason lignin, also known as "the residue that's insoluble to 12 M sulfuric acid." Whereas the NCP procedure, it's filtered through filter paper. And then all three procedures, you collect the acid hydrolysates, which contain all the individual sugars that compose the non-starch polysaccharides. Slide 17 Then, in Day 3, it's the synthesis of alditol acetate derivatives for GLC determination. The first step is the addition of an internal standard, allose, which is added to the acid hydrolysates. Next, potassium borohydride reduces the sugars to alcohols, then the alcohols are acetylated to alditol acetate derivatives. And finally, they're quantified by gas liquid chromatography. Slide 18 And so, the analysis gives the amount of sugars bound by beta linkages, which are rhamnose, fucose, arabinose, xylose, galactose, glucose, mannose, and uronic acids are determined separately but they're included in this. So it's sort of a fallacy of the name NSP analysis, because it doesn't give you the actual non-starch polysaccharides. It gives you the sugars that make up those non-starch polysaccharides. So, the NSP may be calculated by adding some of those sugars together and estimating which sugars make up which non-starch polysaccharides. Slide 19 So some calculations you can make: total NSP is equal to the sum of all the sugars. Cellulose is equal to the total NSP glucose that's been hydrolyzed with 12 M sulfuric acid, minus the glucose from the NCP procedure that's been hydrolyzed with 2 M sulfuric acid. Soluble NSP is equal to total NSP minus insoluble NSP, and dietary fiber is equal to total NSP plus Klason lignin. Slide 20 So now, we'll look at the total ingredient composition of corn and corn coproducts, which are on the x-axis. And we have moisture in blue, ash in gray, acid hydrolyzed ether extract in orange, crude protein in green, dietary fiber in yellow, and starch in red. And you see that the number for each ingredient represents the majority of that ingredient's composition. So for corn, 62% of corn is starch. Whereas dietary fiber, crude protein, fat, ash, and moisture make up lesser amounts. Moving from corn to corn DDGS, we see that the starch drastically decreases. And this is because, in the fermentation of corn to make ethanol, they ferment the starch, turning it into ethanol and removing it from the corn. So that's why you only see about 8% starch in the corn DDGS. And, because starch makes up approximately 2/3 of the corn, when you remove that, all the other nutrients are concentrated approximate three times in the corn DDGS. So you see that the dietary fiber has increased up to 28.9% in DDGS, and also crude protein, acid hydrolyzed ether extract, have also increased approximately three times from corn. Moving from corn DDGS to corn bran, we see that the starch concentrations increase slightly, but the most important increase is in dietary fiber, from 28.9% in corn DDGS to 45.7% in corn bran. And finally, corn gluten meal is the oddball ingredient, because it's a high protein ingredient. and as you can see, it contains approximately 60% crude protein and lesser amounts of dietary fiber, starch, and fat. Also, it is interesting to note that none of the ingredients, with maybe corn gluten meal being the exception, that none of these ingredients add up to 100%. So what's missing? Well, one thing that is for sure missing are the low molecular weight carbohydrates such as sucrose and lactose, and also the galactooligosaccharides. And these are approximately 4% in corn. So possibly, these low molecular weight carbohydrates could bring the corn total up to 100%. But it's unlikely that the coproducts contain upwards of 10% of these low molecular weight carbohydrates. And so the only other conclusion to make is that the proximate analysis that we're using to determine these could be a little inaccurate. For instance, crude protein, you measure the amount of nitrogen and multiply by 6.25, and so this is an estimation for all proteins. And we know that not all proteins contain the same amount of nitrogen. And so because of this estimation in the proximate analysis, these may account for the reason why not all the ingredients add up to 100%. Slide 21 Now, moving on to sorghum and wheat, and sorghum and wheat coproducts, we see that the grains sorghum and wheat contain a majority of starch with 69% in sorghum and 61.8% in wheat. Now if we move from sorghum to the two sorghum DDGS samples, we see that the starch is almost completely removed. And again, this is because it's a DDGS sample. Also, you can see that the starch in these sorghum DDGS samples are much lower than the corn DDGS sample. And that shows the wide variability in ethanol production, and also shows that ethanol production is a live biological process using live strains of yeast to ferment the starch, and therefore not all DDGS samples are the same. Also, you can see that there's 32.9% dietary fiber in the sorghum DDGS sample from Kansas, and 41.3% dietary fiber in the sorghum DDGS sample from Texas. And this is because the Kansas sample had greater solubles added to it. And again, from sorghum to sorghum DDGS samples, you can see that the dietary fiber, crude protein, and acid hydrolyzed ether extract have all increased approximate 3x. And now, if we look at the two wheat coproducts, we see the majority of them are made up of dietary fiber with 38.1% in wheat middlings and 41.4% in wheat bran. Slide 22 Now, we'll look at non-starch polysaccharide percentage in all 12 ingredients. In the green portion of the bar, we have the insoluble NSP. And the number within the green bar is the percentage of insoluble NSP within each ingredient. In the yellow bars, we have the soluble NSP. But the numbers that are above each bar are the total amount of NSP within each ingredient. And from this graph, we can see that going from corn to corn bran, we get an increase in total NSP. Also, when we go from sorghum to the two sorghum DDGS samples, we get an increase in NSP. And finally, you see that same pattern from wheat to the two wheat coproducts. And so what this is telling us is that as you go from grain to grain coproduct, you get an increase in total NSP. And most of that increase occurs in the insoluble NSP. But also, you do increase soluble NSP. However, if you look at the soluble NSP as a percentage of total NSP, corn, corn germ meal, corn gluten meal, and wheat all have about 30% soluble NSP as a percentage of total NSP, whereas all the other ingredients have around 10 to 15%. And this difference is important because the soluble NSP are rapidly fermented and provide a majority of the VFAs that the pig can use for energy, whereas only some of the insoluble NSP may be fermented and so you get less VFAs and subsequently, less energy from the insoluble NSP. Also, you can see that, again, corn gluten meal is the oddball and it has the least amount of total NSP with 3.6%. And again, this is because it is a high protein feed ingredient. Slide 23 Now, if we look at the percent sugar as a percentage of the total NSP in corn on the left and corn DDGS on the right, we see that corn contains 8.1% total NSP, and of that 8.1%, 22.2% is arabinose, 27.2% is xylose, and 27.2% is glucose. And the other sugars make up the rest of the pie. And here we see that it's about equal parts arabinose, xylose, and glucose. Now, if we move on to corn DDGS, with 25% total NSP, we see that approximately 20.4% of that total NSP is arabinose, 28.4% xylose, and 31.6% glucose. And the other sugars, again, are all minimal. And we see that going from corn to corn DDGS, whereas we increase total NSP, we see that the sugars making up that total NSP do not change much. So we see equal parts of glucose, arabinose, and xylose as a percentage of total NSP in both corn and corn DDGS, even though corn DDGS has approximately three times as many total NSP. Slide 24 Moving on to sorghum and sorghum DDGS, on the left, sorghum has 6.6% total NSP, and of that, 25.8% is arabinose, 19.7% is xylose, and 37.9% is glucose. And the other sugars are minimal. And as we increase the total NSP from sorghum to sorghum DDGS to 24.7%, of that 24.7%, 19% is arabinose, 22.3% is xylose, and 43.3% is glucose. Again, we see the same trend as we go from sorghum to sorghum DDGS, increasing total NSP, we still see the same ratios of the sugars that make up that total NSP in sorghum and sorghum DDGS. So we can conclude that glucose makes up the majority, or cellulose, and then arabinose and xylose, or arabinoxylans, make up a smaller amount, and then the other sugars make up the rest of the total NSP. Slide 25 Finally, if we look at the percent sugar as a percentage of total NSP in wheat and wheat bran, we see that in wheat, there's 9.5% total NSP, and of that 9.5%, 25.3% is arabinose, 37.9% is xylose, and 25.3% is glucose. Again, the other sugars are minimal. And again, we see the same trend. As we increase the total NSP, so moving from wheat to wheat bran with 34.5%, we see that arabinose makes up 22.6%, xylose 41.7%, and glucose 28.4%. So here we can conclude that arabinoxylans, or arabinose and xylose within wheat and wheat bran, make up the majority of the total NSP, and then glucose or cellulose to a smaller extent. And then the other sugars make up the rest. Slide 26 So in summary of Experiment 1, we saw that corn, sorghum, and wheat were high in starch, whereas coproducts were high in non-starch polysaccharides. Then, in coproducts, we saw that their NSP composition was similar to the parent grains. So, for corn and corn coproducts, we saw equal arabinose, xylose, and glucose. Whereas in sorghum and sorghum coproducts, we saw less arabinose and xylose, and greater glucose. And finally, in wheat and wheat coproducts, we saw greater arabinose and xylose, and less glucose. Slide 27 Moving on to Experiment 2, the objective was to determine in vitro ileal and total tract digestibility of dry matter and NSP in grains and grain coproducts. Slide 28 Here's the summary of the in vitro procedure that we modified from Boisen and Fernández, 1997. And on the left side, you can see the ileal procedure, which is a two-step procedure where you add pepsin at a pH of 2 for 75 minutes, and then pancreatin at a pH of 6.8 for 18 hours. And then before filtration, you add ethanol to precipitate the soluble NSP in order to account for the soluble NSP. Then you filter it, and you're left with the indigestible residue. On the right, we have the total tract, or three-step procedure, where we added pepsin and pancreatin, just like the ileal procedure, but then we introduced this third step, which you add a Viscozyme enzyme, which is a concoction of many different NSP-degrading enzymes, for instance cellulase and xylanase, and you incubate it at a pH of 4.8 for 18 hours. And then you filter it, and you're left with the non-fermentable residue. Slide 29 Then, we took the residues and performed NSP analysis on them by acid hydrolyzing the samples. So, moving to the assumptions, we assumed that all starch was digested in vitro, and that is why we acid hydrolyzed the samples to begin with, and we did not remove the starch. Also, we assumed that all the soluble fiber was fermented in vitro, or we did not account for the soluble fiber in the three-step total tract procedure, because roughly all the soluble fiber will be fermented in the pig by the end of the total tract, and therefore we did not feel it was necessary to account for it in the total tract in vitro procedure. Slide 30 Looking at the in vitro dry matter digestibility of corn and corn coproducts, we see that the green bars represent the ileal in vitro dry matter digestibility, and the yellow bars represent the total tract in vitro dry matter digestibility. And the numbers above each bar are the percentage of that digestibility. So if we look at corn, which had 89.4% ileal in vitro dry matter digestibility, and move on to corn bran, we see that both the ileal and total tract dry matter digestibilities decrease as we move from corn to corn bran. Again, we move on to corn gluten meal, and we see that it has a very high in vitro ileal and total tract dry matter digestibility. And again, this is because it is a high protein feed ingredient that is very digestible. Also, if we look at the green or the ileal dry matter digestibility versus the yellow, which is the total tract dry matter digestibility, within each ingredient, we see that moving from the ileal to the total tract, we get an increase in digestibility. And this increase shows that there is an effect on the Viscozyme enzyme on degrading dry matter within all of these ingredients. However, the increase from ileal to total tract varied drastically from as little as less than a percent in corn gluten meal to as great as almost 20% in corn germ meal. And so this shows that the Viscozyme enzyme used in the total tract in vitro dry matter digestibility procedure digests dry matter, but the amount that it digests varies among all ingredients. Slide 31 Now, if we look at sorghum and wheat, and sorghum and wheat coproducts, we see the same trend. Going from sorghum and wheat – the grains – down to their coproducts, we get a decrease in both the ileal and total tract in vitro dry matter digestibilities. And again, looking at the ileal versus the total tract in vitro dry matter digestibility within each ingredient, we see that the total tract digestibility increases compares with the ileal in vitro dry matter digestibility. And again, this increase varied widely amongst all the ingredients. Slide 32 If we plot the in vitro ileal dry matter digestibility, which is on the y-axis, versus the NSP content of the ingredients, which is on the x-axis, where on the x-axis we have the NSP content ranging from 3.6% in corn gluten meal up to 41% in corn bran, and on the y-axis we have the in vitro ileal dry matter digestibility, ranging from as high as about 94% in corn gluten meal to as low as 49% in corn bran, we get a very strong correlation between the two, with an R-squared of 0.97. And this relationship was expected because, in the in vitro ileal dry matter digestibility procedure, we're digesting away all the starch, lipid, and proteins. And so what's left is the majority of the NSP. Therefore, we expected this relationship. And you can see that for every 1% increase in non-starch polysaccharides in your ingredient, you get a reduction of 1.2% in the in vitro ileal dry matter digestibility. Slide 33 Finally, looking at the in vitro total tract digestibility of non-starch polysaccharides of all 12 ingredients, we see that there's a wide range of digestibilities, ranging as low as 6.5% in corn bran to as high as 57.3% in corn gluten meal. Also, we see that corn germ meal and wheat had relatively high total tract digestibilities of non-starch polysaccharides. And then, to lesser extents, corn and corn DDGS and wheat bran and the sorghum DDGS from Kansas. And so what this can tell us is that the non-starch polysaccharide composition of each ingredient influences the energy value of these ingredients, because this is simulating the fermentation of non-starch polysaccharides in the large intestine of the pig, and the more fermentation that occurs or the more in vitro total tract digestibility of NSP, the more VFAs that will be produced and subsequently supply the pig with energy. So in conclusion, the NSP composition of each ingredient plays an important role in determining the extent of fermentation of non-starch polysaccharides. Slide 34 So the conclusions of Experiment 2: we saw that the in vitro ileal NSP digestibility is zero. Now for the sake of time, I did not present this, but this is important because we determined it was zero, and that shows that pepsin and pancreatin enzymes do not degrade any NSP. And therefore, the only degradation of NSP in the total tract procedure is due to the Viscozyme enzyme. Also, we saw that the in vitro total tract NSP digestibility varied among ingredients. And finally, in vitro ileal dry matter digestibility is strongly correlated to NSP content, with an R-squared of 0.97. Slide 35 So the take home message of today's presentation is that the majority of grains is composed of starch, whereas the majority of grain coproducts is non-starch polysaccharides. And the NSP composition of each ingredient plays an important role in determining the extent of fermentation of NSP. So therefore, the NSP composition influences the energy value of the ingredients. Slide 36 With that, I would like to thank you for joining me today, and I'd like to refer you to the Hans H. Stein Monogastric Nutrition Laboratory web site for more information.