Effect of acid detergent lignin concentration for diets formulated to be similar in NDF content on energy utilization in lactating Jersey cows

Lignin is a polyphenolic polymer that is an important factor in limiting fiber digestibility by ruminants. The objective of the current study was to evaluate lignin’s impacts on whole animal energy utilization in diets similar in NDF content. A low lignin (LoLig) diet was formulated to contain 32.5% NDF (DM basis) and 9.59% lignin (NDF basis) and the high lignin (HiLig) diet was formulated to contain 31.0% NDF (DM basis) and 13.3% lignin (NDF basis). These diets were randomly assigned and fed to 12 late-lactation (214 ± 14.9 DIM) multiparous Jersey cows (435 ± 13.9 kg) in a 2-period crossover design. Cows fed the LoLig treat-ment consumed more DM than cows on the HiLig diet (19.9 vs. 18.7 ± 0.645 kg/d) while the LoLig diet was concurrently of a greater gross energy concentration (4.27 vs. 4.23 ± 0.03 Mcal/kg). As expected, increasing the concentration of lignin resulted in a reduction in total-tract NDF digestibility (45.5 vs. 40.4 ± 0.742%). Increasing lignin also resulted in a reduction in the digestibility of starch (97.7 vs. 96.3 ± 0.420) and CP (65.0 vs. 60.0 ± 0.829). Lignin also decreased the concentration of digestible energy (2.83 vs. 2.63 ± 0.04 Mcal/kg) and metabolizable energy (2.52 vs. 2.36 ± 0.05 Mcal/kg) but the concentration of net energy of lactation was similar (1.81 vs. 1.75 ± 0.06 Mcal/kg. Increasing the concentration of lignin also reduced yields of energy-corrected milk (33.7 vs. 30.0 ± 0.838 kg/d), milk protein (1.00 vs. 0.843 ± 0.027 kg/d), and milk fat (1.30 vs. 1.19 ± 0.058 kg/d). Decreasing the dietary lignin concentration did not affect daily methane emissions, averaging 391 ± 29.6 L/d. Results of this study indicate feeding a diet greater in lignin decreases the digestibility of nutrients and provides less energy for production responses and that energy supplied from digestible NDF may be less than predicted by some nutrition models.


INTRODUCTION
Neutral detergent fiber usually accounts for 25-33% of the DM in dairy rations (NASEM, 2021).The digestibility of NDF is a major factor affecting energy available to the cow (Dado and Allen, 1996).The acid detergent lignin (ADL) concentration of the NDF fraction is a plant-based factor limiting ruminal digestion of NDF (Jung et al., 1997).Therefore, reducing the ADL content of dairy diets is of interest because this may increase digestibility, feed intake, and overall energy supply (Jung et al., 2011).Many studies have described the negative impact of ADL on NDF digestibility (Jung et al., 2011;Raffrenato et al., 2017;Van Soest et al., 2018).Although the current dairy NASEM (2021) model uses dietary ADL content to estimate NDF digestibility the interrelationships between ADL, NDF digestibility, and whole animal energy utilization are still not completely understood.
Lignin and some of the other bound polymers are assumed to be completely indigestible and therefore the energy contained within these nutrients are not available to the animal and also may negatively affect the digestibility of other organic fractions (Dong et al., 2011).The gross energy (GE) concentration of ADL is reported to be 6.0 Mcal/kg (Voitkevich et al., 2012).Additionally, we recently observed the GE concentration of the NDF fraction averaged 4.03 Mcal/kg (Stypinski et al., 2023) while the dairy NASEM assumes it to be 4.20 Mcal/kg (NASEM, 2021).Consequently, increasing the proportion of ADL within the NDF fraction should result in an increase in the GE concentration of the entire NDF fraction, while also limiting the digestibility of the NDF and digestible energy (DE).We are not aware of any studies that have analytically determined the GE concentration of feed and fecal NDF in a single controlled feeding study.Analytically determining these values could provide important insight for predicting the energetic contribution from fiber by our nutritional models.Should the GE concentration of NDF be lower than that currently estimated in nutritional models they may be overestimating their predictions of GE and DE provided by NDF.This in turn, could have impacts on the dietary concentrations of metabolizable energy (ME) and net energy (NEL).The objective of this study was to evaluate energy metabolism of lactating cows fed diets with different ADL concentrations and to investigate broader effects on milk production.We hypothesized cows consuming a diet higher in ADL concentration will convert GE to DE at lower efficiency, and this will result in negative impacts on the observed supply of ME and NEL.

Animals and Treatments
The University of Nebraska-Lincoln Animal Care and Use Committee approved animal care and experimental procedures.Twelve multiparous Jersey cows 214 ± 14.9 DIM, weighing 424 ± 46.5 kg and 65.7 ± 80.0 d pregnant were housed in individual tie stalls in a climate-controlled environment (20°C) at the University of Nebraska-Lincoln Dairy Metabolism Facility in the Animal Science Complex.Stalls were surfaced with rubber mats and cows were milked at 0700 and 1800 h.The experimental design was a 2-period crossover with periods of 28 d each.This included 24 d of diet adaptation followed by 4 d of collection.In period 1, cows were randomly assigned to 1 of 2 treatment diets (6 cows per treatment per period): 1) a low-lignin diet (LoLig) or 2) high-lignin diet (HiLig).For period 2, the alternative diet was fed.Dietary treatments were formulated to be similar in CP and NDF but ingredients were manipulated so that they differed in the concentration of ADL as a percent of NDF, specifically 9.6 and 13.3 (% NDF for LoLig and HiLig respectively (Table 1).This was primarily achieved by including more alfalfa hay and cottonseed hulls in the HiLig treatment.Concentrate mixes were prepared and mixed at the University of Nebraska-Lincoln feed mill (Mead, NE).Corn silage, alfalfa hay, and concentrate were added to a Calan Data Ranger (American Calan, Inc., Northwood, NH), mixed, and fed as a TMR once daily at 0930 h with a target refusal rate of 5%.During data collection diets were fed at 100% of the prior week's intake.Because the HiLig diet was formulated primarily with dried alfalfa hay, water was added during the mixing process using a garden hose to increase the diet's moisture content to be similar to that of the LoLig diet.
Total fecal and urine output was collected and sampled from each individual cow during the collection period for 4 consecutive d as described by McLain et al. (2021).After collections, approximately 600 g feces were dried at 60°C for 48 h and ground to pass through a 1 mm screen (Wiley Mill; Arthur A. Thomas Co., Philadelphia, PA).The ground feces were analyzed for chemical composition, and urine for nitrogen using the same methods as described for feeds.Milk production was measured daily, and milk samples were collected during the morning and evening milking of collection periods as described by McLain et al. (2021).Milk from individual milking events was preserved with 2-bromo-2nitropropane-1,3 diol and sent to Heart of America DHIA (Kansas City, MO).Milk samples were analyzed for fat, true protein, lactose, SNF, MUN, and SCC using a Bentley FTS/FCM Infrared Analyzer (Bentley Instruments, Chaska, MN).Additionally, milk from each milking event was composited on a weight basis from each milking for nitrogen analysis.Cows were weighed before feeding on the first and last day of each collection period.Heat Production and Energy Utilization and Calculations.Heat production was determined indirectly through the headbox-type indirect calorimeters as described previously (Carroll et al., 2023).Total volume of gas flow through the headbox was measured using mass flow meters (MCW-1000SLPM-D Whisper, Alicat Scientific) and corrected to standard temperature and pressure (0°C, 101.3 kPa) with adjustment for moisture content of exhaust air (Nienaber and Maddy, 1985).All cows were trained and acclimatized to the headboxes before the start of the study.To do so, cows were first placed in the headboxes for 8 h to ensure they would eat and drink normally.Following their first successful 8-h session, cows were placed in the headboxes again for increasing lengths of time until they ate and drank typical amounts of feed and water over a 24-h period approximately 2 weeks before the first sample collection period.Once acclimatized cows were then placed in the headbox for one 23 h period during the experimental period and were provided their daily allotment of feed in the headbox for this time.System efficiency (head box and gas analyzer) was also determined by burning 100% ethyl alcohol and measuring gas recoveries.Recoveries of O 2 and CO 2 were (average ± SD) 100 ± 2.9 and 99 ± 2.5%, respectively.
The respiratory quotient (RQ) was calculated using the ratio of carbon dioxide produced to oxygen consumed (L/L).Methane energy was estimated by multiplying CH 4 production by its enthalpy (9.45 kcal/L).Calculations to estimate digestible energy (DE), ME and NEL were as follows: Unaccounted for energy was assumed to represent tissue energy retention or mobilization which was corrected to an NEL basis as follows (NRC, 1989): Where k T is the efficiency of utilizing body reserve energy for milk production, k G is the efficiency of utilizing ME intake for tissue gain, and k L is the efficiency of utilizing ME for milk synthesis (Moe et al., 1971).Values of 0.66 and 0.74, and 0.89 were used for k L , k G, and k T respectively (Moraes et al., 2015).Fetal energy was estimated using equations in the dairy NASEM (2021).
To isolate NDF residues from feed and fecal samples for GE analysis, feed and fecal samples were dried at 60°C for 48 h and ground through a 1-mm sieve (Wiley Mill; Arthur A. Thomas Co., Philadelphia, PA).Once ground, NDF residues were obtained from feed and fecal samples using the Ankom technique (Ankom200 Fiber Analyzer, Ankom Technology Corp., Fairport, NY) in quadruplicate with sodium sulfite and α amylase (Sigma A3306; Sigma-Aldrich, St. Louis, MO).Neutral detergent fiber residues were then ground further using a mortar and pestle and dried again at 60°C for 24 h.Once completely dried, 0.2 g of ground NDF residue and 0.4 g of mineral oil were weighed for bomb calorimetry (Parr 6400 Calorimeter, Moline, IL).Weighed samples were soaked in mineral oil for 24 h before being placed in the bomb calorimeter for determination of GE.

Statistical Analysis
Data were analyzed using PROC GLIMMIX function of SAS.The statistical model considered effects of treatment and period as fixed and cow within period as random.All data are presented as least squares means ± standard error.Significance was declared as P-value ≤0.05 and tenancies were declared as 0.05 < P-value ≤0.10.

RESULTS
Twenty-two out of the 24 planned observations were obtained for gas related calculations.During the training and acclimatization period, one cow refused to drink water while in the headbox thus gas or energy related measures were not collected or used on this cow.However, all other measures were collected and used from this cow in both periods.Diet composition of the 2 treatment diets and chemical composition of forages and grain mixtures are listed in Tables 1 and 2. The ADL content of the HiLig diet was approximately 1 percentage unit greater than that of the LoLig diet, averaging 4.38 ± 0.129 and 3.29 ± 0.208% for the HiLig and LoLig diets, respectively.The NDF content of the LoLig diet (32.5 ± 0.655%) was similar to that of the HiLig diet (31.0 ± 0.404%), but ADL as a percent of NDF was lower for the LoLig treatment (9.59 ± 0.388%) than the HiLig treatment (13.3 ± 0.02%).

Energy and N utilization
Feeding the HiLig diet reduced GE intake by 5.  5).Urinary energy tended to decrease with increasing ADL content from 1.90 to 1.47 ± 0.16 Mcal/d (Table 5).Increasing the concentration of ADL in the diet did not have an effect on NEL concentration or heat production, averaging 1.67 ± 0.06 Mcal/kg and 21.7 ± 1.21 Mcal/d across treatments (Table 5).The GE concentration of NDF residues in either the feed or feces were not different averaging 4.12 ± 0.081 Mcal/ kg and 3.96 ± 0.028 Mcal/kg respectively (Table 6).The difference between feed and fecal NDF residues was greater for the HiLig diet compared with the LoLig diet (0.110 vs. 0.209 ± 0.0277 Mcal/kg) (Table 6).Nitrogen intake was greater for cows consuming the LoLig diet compared with cows consuming the HiLig diet (525 vs. 497 ± 16.9 g/d; Table 7).Feeding the HiLig diet decreased and tended to decrease milk and urinary N as a percent of N intake (35.6 vs. 33.0± 0.616% and 27.4 vs. 23.1 ± 1.73%, for LoLig and HiLig, respectively).

Milk production and composition
Observed milk production and composition is listed in Table 3, Increasing the concentration of ADL in the diet decreased milk production and ECM (28.2 vs. 25.1 ± 0.799 kg/d, and 33.7 vs. 30.0± 1.08 kg/d, for LoLig and HiLig respectively; Table 3).Milk protein and fat concentrations were not affected by treatment averaging 3.47 ± 0.155% and 4.74 ± 0.264% for LoLig and HiLig across treatments.In contrast yields of milk protein and fat were lower for the HiLig diet (1.00 vs. 0.843 ± 0.027 kg/d and 1.30 vs. 1.19 ± 0.058, for LoLig and HiLig, respectively).When the ADL content of the diets was increased, energy-corrected milk produced per kg of DMI was also reduced from 1.69 to 1.61 ± 0.027.

DISCUSSION
The objective of this study was to manipulate the ADL content of the diet fed to lactating dairy cattle and then characterize the effects on feed intake, digestibility, and whole animal energy utilization.The HiLig diet was formulated to contain a higher concentration ADL by substituting corn silage for alfalfa hay as the primary forage and also adding cottonseed hulls.This could be considered a weakness of the current study because associated effects and interactions of different feeds may exist but the concentration of lignin within the same feed types could not be manipulated.Lignin has been reported to possess a GE concentration of 6.0 Mcal/kg (Voitkevich et al., 2012) which compared with other nutrients such as NDF and starch (4.20 Mcal/ kg), is high (NASEM, 2021).However, ADL is also considered to be indigestible and therefore should not supply useable energy to the cow.It is for this reason, we hypothesized that diets containing greater proportions of ADL will decrease the conversion of GE to DE and have resulting negative effects on ME and NEL.

Feed intake and nutrient digestibility
Feed intake is affected by an array of complex factors that are both physical and metabolic in nature (NASEM, 2021).Feeding the HiLig diet reduced DMI by 1.2 kg and such a response is consistent with that of Benchaar et al. (2014) and a trend observed by Colombini et al. (2015) evaluating the influence of ADL on feed intake.Reductions in DMI have been observed when forage NDF alone is increased but in the current experiment the 2 treatments only differed slightly in forage NDF (21.4 vs. 22.2% DM for LoLig and HiLig, respectively).We suggest that a major factor in the reduction in DMI was digestibility of NDF which was reduced from 45.5 to 40.4% for LoLig and HiLig, respectively.Diets tested in our study differed in the proportion of ADL within the NDF fraction (9.59 to 13.3% NDF in LoLig and HiLig respectively) which is known to negatively affect fiber digestibility and in turn, DMI (Chow et al., 2008;Brown et al., 2018;Lyons et al., 2019).We observed that this 39% increase in ADL (% NDF) reduced NDF digestibility by 11% and aNDFom digestibility by 9%.Both diets were evaluated using the dairy NASEM (2021) model.Using the lignin-based equation to predict total-tract NDF digestibility (see Equation 3.3a; page 24), these simulations predicted  an 8% reduction in digestibility of NDF.We also input values generated from the 48 h in vitro NDF digestibility assay conducted on forages (see Equation 3-3b; page 24) and the model only predicted a 1% reduction in digestbility.The NASEM estimates also adjust for negative effects of starch concentration and DMI.
Results of the current study support the notion that ADL content of the NDF stands as important factor in affecting NDF digestibility.Although it has long been recommended, expressing NDF on an organic matter basis (Van Soest et al., 1991 a) this representation is only now commonly practiced in the field.Our results indicate that although this correction yields in a lower fiber content, the digestibility of this fiber is also greater thus the difference in the calculated amount of digestible fiber consumed is small (2.8 and 4.6% less for the LoLig and HiLig treatments respectively).In addition to NDF, the apparent digestibility of CP and starch were also lower in cows consuming the HiLig treatment.We did not anticipate this response and since   nutrient digestibility of individual feedstuffs cannot be measured.the difference in digestibility in CP and starch may be a result of the source of these nutrients.
In the case of starch, more originated from corn grain and less from corn silage in the HiLig treatment and in the case of CP, more originated from alfalfa hay and less from feeds such as non-enzymatically browned soybean meal, DDGS, and soybean meal.Results should be interpreted with the knowledge that experimental diet formulations differed by inclusion of ingredients and the LoLig formulation is likely indicative to common diets.The HiLig diet included ingredients such as cottonseed hulls that are not frequently fed to lactating cows and likely not well represented in diets used to develop NASEM predictions.

Energy metabolism and supply of energy from NDF
Feeding the HiLig diet reduced the conversion of observed GE to DE from 66.2 to 62.6% but this response is likely not due to NDF alone because digestibility of other nutrients was also reduced.A major contributing factor to this decrease was that the HiLig contained a greater proportion of ADL, which has GE concentration of 6.0 Mcal/kg (Voitkevich et al., 2012) but supplies no DE (Morris, 2020).Using observed animal inputs and diet composition the dairy NASEM (2021) model predicted that increasing the ADL content of diet would reduce concentration of DE by 2.7% (2.98 vs. 2.90 Mcal/kg).Although this response was observed, the measured difference was much larger (7%).The under-prediction was likely because differences in digestibility were not isolated to NDF but also a result of the effects of other nutrients.In support of this, NASEM only predicted a small decrease on OM digestibility (69.6 to 68.8%) with increasing diet ADL content, yet we observed this to be much greater (69.0 to 65.9%).Similar to DE, the concentration of ME was lower when cows consumed the HiLig treatment.This difference was driven by changes in DE and not differences in methane or urinary energy.The NASEM (2021) predicts urinary energy loss using an equation based upon estimated urinary N excretion (Morris et al., 2021).The observed urinary energy losses by treatment  2021), the observe TE would be expected to gain 0.57 kg/d.This estimate should be interpreted with caution as it is calculated by difference and consequently includes measurement error associated with all other energy losses.
As previously noted, the NASEM (2021) assumes that the GE concentration of NDF to be 4.20 Mcal/ kg and this is used in the model's predictions of both GE and DE.In the current study, we directly measured the GE concentration of diet NDF residue which was not affected by treatment and averaged 4.12 Mcal/kg.We have previously reported that the GE of NDF in many feeds is less than the NASEM assumed value 4.20 Mcal/kg (Stypinski et al., 2023).When observed DMI and formulations of our study were entered into the NASEM (2021) model solutions resulted in an overprediction of DE supply for both diets (3.1 and 4.5 Mcal/d, or by 5 and 8% for LoLig and HiLig respectively).If the measured digestibility and 4.12 Mcal/kg NDF GE coefficient were used in place of the calculated NDF digestibility and the assumed NDF GE coefficient, the model would have only over-predicted DE by 3 and 6% for LoLig and HiLig, respectively.We also measured the GE content of fecal NDF residues which similar to feed, was not affected by treatment but was lower than the GE of feed NDF residues.This finding was not expected as ADL should be found in greater proportion in fecal NDF residue and this should in turn, increase the overall GE concentration of fecal NDF residues.The lower concentrations of GE observed in fecal NDF is likely because ash content of samples was higher for animals on the HiLig treatment (5.96%) than the LoLig treatment (4.06%).Because this is inorganic in nature, ash provides no GE, yet because it accumulated more in fecal NDF residues in the HiLig treatment it resulted in a greater difference in GE between feed and fecal NDF (0.209 vs. 0.110 Mcal/kg for HiLig and LoLig respectively).We measured the ash content of NDF residue in feed and feces and also the GE content of these fractions, thus we calculated the GE content of aNDFom residue by assuming the heat of combustion of the ash was zero.In the case of feed samples, the GE of aNDFom was computed to be to be 4.11 and 4.26 Mcal/kg for the LoLig and HiLig treatments.Additionally, the GE of aNDFom of fecal samples was computed to be 4.18 and 4.32 Mcal/kg.The greater GE in feces and feed and feces in cows consuming the HiLig treatment was expected and likely because of the presence of more lignin.Results from this study suggest the nature of digestion of energy from fiber is not consistent across diets and we suggest that in the future GE of NDF be evaluated on an organic matter basis and perhaps CP free basis as well.

Milk production and composition
Feeding the diet containing more ADL decreased energy-corrected milk yield from 33.7 to 30.3 kg, and this was likely a function of reduced intake, nutrient digestibility, and energy intake.Similarly, Uddin et al. (2020) increased the proportion of lignin through the replacement of corn silage with alfalfa and reported a reduction in NDF digestibility, DE intake, and consequently energy-corrected milk yield.Additionally, in our study yields, but not concentrations, of milk fat and protein were lower for the HiLig diet compared with the LoLig.The reduction in milk fat and protein secretion was likely caused by reduced energy intake.Erdman et al. (2011) andOelker et al. (2009) also observed increase in milk fat and protein yields, respectively, when replacing alfalfa hay and silage with corn silage.Overall, this study indicates that the composition of the NDF fraction has pronounced effects on the digestibility of multiple nutrients, and therefore whole animal energy supply and utilization.

CONCLUSIONS
Increasing the concentration of ADL in diets fed to lactating Jersey cows reduced feed intake and digestibly of fiber along with other nutrients, culminating into a reduction in energy supply.The effects of decreased energy intake associated with the higher lignin diet had decreasing effects on milk yield and composition.When using the NASEM (2021) model to predict energy supply and nutrient digestibility, these predictions were closer to in vivo observations in a diet containing less lignin.Lastly, the current study suggests that the dairy NASEM (2021) model more accurately predicts DE when utilizing an adjusted GE concentration coefficient for NDF of less than 4.20 Mcal/kg coefficient currently assumed in the model.
(1.90 to 1.47 Mcal/d) were slightly lower to what was predicted (2.15 to 2.05 Mcal/d) by this equation.Treatment differences were observed in both DE and ME.In contrast, no differences were observed on NEL (1.81 and 1.75 Mcal/kg for LoLig and HiLig respectively) and this may have been at least in part been due to larger variation associated with this measurement.Although no difference in TE was observed the mean estimate observed was high (3.18 ± 1.52 Mcal/d).Assuming 5.6 Mcal are needed for every kg of gain (NASEM (

Table 2 .
Stypinski et al.: LIGNIN AND DAIRY DIETS Chemical composition of corn silage, alfalfa hay, and concentrate mixes used to formulate the low-lignin and high-lignin diets fed to lactating Jersey cattle. 1

Table 3 .
Dry matter intake, milk production and components, water intake, body weight (BW), and body condition score (BCS) for lowlignin and high-lignin diets fed to lactating Jersey cattle 4Feed efficiency calculated according to the dairy NASEM (2021), equation 3-21 on page 36; (milk energy + tissue energy)/GE.

Table 4 .
Total-tract digestibility of major nutrients in lactating Jersey cows fed of low-lignin and high-lignin diets

Table 5 .
Gas and energy measures of low-lignin and high-lignin diets fed to lactating Jersey cattle Stypinski et al.: LIGNIN AND DAIRY DIETS 1 LoLig = low-lignin diet; HiLig = high-lignin diet. 2 Least squares means; largest standard error of treatment mean is shown.

Table 6 .
Gross energy concentrations (Mcal/kg) of feed, feces and the difference between the two NDF residues when lactating Jersey cows are fed either low-lignin and high-lignin dietary treatment

Table 7 .
Stypinski et al.:LIGNIN AND DAIRY DIETS Fecal and urinary output and nitrogen excretions for low-lignin and high-lignin diets fed to lactating