Feeding spent hemp biomass to lactating dairy cows: effects on performance, milk components and quality, blood parameters, and nitrogen metabolism.

The legalization of industrial hemp by the 2018 Farm Bill in the US has driven a sharp increase in its cultivation, including for cannabinoid extraction. Spent hemp biomass (SHB), produced from the extraction of cannabinoids, can potentially be used as feed for dairy cows; however, it is still illegal to do so in the US, according to FDA-CVM, due to the presence of can-nabinoids and the lack of data on the effect on animals. To assess the safety of this by-product as feed for dairy cows, late-lactation Jersey cows (245 ± 37 DIM; 483 ± 38 kg BW; 10 multiparous and 8 primiparous) received a basal TMR diet plus 13% alfalfa pellet (CON) or 13% pelleted SHB (SHB) for 4 wk (intervention period or IP) followed by 4 wk of withdrawal period (WP) where all cows received the basal TMR only during WP. The DMI, BW, BCS, milk yield, milk components and fatty acid profile, blood parameters, N metabolism, methane emission, and activity were measured. Results indicated that feeding SHB decreased DMI mainly due to the low palatability of the SHB pellet, as the cows only consumed 7.4% out of 13.0% of the SHB pellet offered in the ration. However, milk yield was not affected during the IP and was higher than CON during the WP, leading to higher milk yield/DMI. Milk components were not affected, except for a tendency in decreased fat %. Milk fat produced by cows fed SHB had a higher proportion of oleate and bacteria-derived FA than CON. The activity of the cows was not affected, except for a shorter overall lying time in SHB vs. CON cows during the IP. Blood parameters related to immune function were not affected. Compared with CON, SHB cows had a lower cholesterol concentration during the whole experiment and higher BHBA during the WP, while a likely low-grade inflammation during the IP was indicated by higher ceruloplasmin and reactive oxidative metabolites. Other parameters related to liver health and inflammatory response were unaffected, except for a tendency for higher activity of ALP during IP and a lower activity of gamma-glutamyl transferase during WP in SHB vs. CON. The bilirubin concentration was increased in cows fed SHB, suggesting a possible decrease in the clearance ability of the liver. Digestibility of the dry matter and protein and methane emission were not affected by feeding SHB. The urea, purine derivatives, and creatinine concentration in urine was unaffected, but cows fed SHB had higher N use efficiency and lower urine volume. Alto-gether, our data revealed a relatively low palatability of SHB affecting DMI with minimal biological effects, except for a likely low-grade inflammation, a higher N use efficiency, and a possible decrease in liver clearance. Overall, the data support the use of SHB as a safe feed ingredient for lactating dairy cows.


INTRODUCTION
Historically, industrial hemp was an important crop in the US, especially to produce fiber.However, its close association with marijuana led to a ban on its cultivation in 1970, with the passage of the Comprehensive Drug Abuse Prevention and Control Act (Cherney and Small, 2016).The 2018 Farm Bill removed hemp from the Controlled Substances Act, classifying it as an agricultural product in the US (McConnell, 2018).Different from marijuana (Cannabis sativa forma indica), hemp contains ≤ 0.3% of the psychoactive cannabinoid tetrahydrocannabinol (THC) but may contain a significant amount of cannabidiol (CBD), a nonpsychoactive cannabinoid.Particularly important in the US is the cultivation of hemp for the extraction of Feeding spent hemp biomass to lactating dairy cows: effects on performance, milk components and quality, blood parameters, and nitrogen metabolism.
CBD from the flower and leaves of feminized hemp (i.e., production without seed production) (Nichols, 2018); this type constituted > 60% of the hemp cultivated in 2020 (National Agricultural Statistics Service, 2022).Cannabinoids are typically extracted from the plant matter by a high-pressure and low-temperature procedure (Valizadehderakhshan et al., 2021) that generates a by-product: spent hemp biomass (SHB).
A recent study indicated that the nutritive value of SHB is quite similar to alfalfa and could be safely fed to finishing lambs at up to 20% in the diet and can increase the antioxidant capacity in the animals (Parker et al., 2022).The anti-inflammatory and antioxidant properties of CBD may have a potential beneficial effect on the immune system (Jensen et al., 2018;Vuolo et al., 2019;Atalay et al., 2020).Hemp is rich in polyphenols with antioxidant properties and other secondary metabolites (Flores-Sanchez and Verpoorte, 2008;Teh and Birch, 2014), that may be able to modulate rumen fermentation, increase nitrogen use efficiency, and reduce methane production (Semwogerere et al., 2020).These beneficial properties can help improve the performance and health of high-producing dairy cows, particularly during the transition from pregnancy to lactation where oxidative stress, immune dysfunction, and metabolic stress are more common (Lopreiato et al., 2020).
Alternative and cost-effective feedstuffs can be highly valuable for the dairy industry, considering the volatility of the feed market and generally low milk prices (Manchester & Blayney, 2021;IFCN, 2021).Given the nutritive value and availability of SHB in the US, it could be an economical feed ingredient.However, feeding hemp or its by-products to livestock is currently not permitted in the US by the Food and Drug Administration -Center for Veterinary Medicine (FDA-CVM) due to the potential for residuals of THC and CBD in animal products, including milk (FDA, 2018).Besides concerns about cannabinoid residuals, SHB would be a new feedstuff and its effects on animal performance and health need to be assessed, as urged by the Association of American Feed Control Officers (AAFCO, 2018).
Our long-term goals are to assess the safety of using hemp by-products as feed for livestock and to investigate their nutritional and potential medicinal effects on animal health and the quality of animal products.To our knowledge, only one study was carried out in feeding SHB to lactating dairy cows (Wang et al., 2023); however, the level of cannabinoids in SHB was 0.03% in that study, approximately 10-fold lower than the SHB we used previously that was 0.3% (Parker et al., 2022).The level of cannabinoid residuals between 0.5 and 3.0% appears somewhat typical after cannabinoid extraction with > 60% being CBD (Wilson et al., 2022).
Due to the expected high levels of cannabinoids in SHB that are bioactive, particularly CBD, and the need for research on hemp by-products before being legalized as a feed ingredient in the US as advised by AAFCO, it is imperative to determine the effects of SHB containing high residuals of cannabinoids on production and health aspects of dairy cows.
The objective of the present study was to assess in lactating dairy cows the effects of feeding SHB on feed intake; milk yield, components, and quality; nitrogen metabolism; methane emission; and blood profile, including parameters related to oxidative stress, immune status, metabolism, and liver function.We hypothesize that feeding SHB decreases feed intake, but it is safe to be fed to dairy cows and does not negatively affect lactation performance or the health of the animals.

Ethical Statement
The study was conducted at the Oregon State University Dairy Center, from May to July 2021.All animal procedures were approved by the Institutional Animal Care and Use Committee of Oregon State University (approval number: #2020-0130).

Experimental Design
In total, 18 late-lactation Jersey cows with (mean ± SD) 2.0 ± 1.1 in parity [10 multiparous with 2 -4 lactation cycles (2.8 ± 0.87) and 8 primiparous], 247 ± 37 DIM, 19.9 ± 3.5 kg/d milk yield, 483 ± 38 kg of BW, and 3.38 ± 0.42 BCS were enrolled in the experiment.Power analysis to justify the number of animals used is provided in Supplemental File (https: / / doi .org/ 10 .6084/m9 .figshare.21804048).The cows were recruited from the dairy herd at the OSU Dairy Center where the experiment took place.Two wk before the commencement of the experiment, complete blood count, leukocyte migration, AM milk component analysis, BW, and BCS measurements were completed to randomly allocate the animals after blocking for parity (5 multiparous and 4 primiparous in each group) (Figure 1).
The whole experiment lasted for 8 wk consisting of 4 wk intervention period (IP) and 4 wk withdrawal period (WP).One wk before starting the experiment, the cows were moved into a pen equipped with individual American Calan feeding gates (American Calan, Northwood, NH).During the first 4 d of the adaptation to the Calan gates, all cows were fed the same diet in the form of a TMR that was formulated for mid-lactation cows (basal diet).Afterward, a progressive inclusion of Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows SHB or pelletized alfalfa meal as the placebo control to the ration began.On d 4 of adaptation, cows were fed with the targeted level of supplementation, leading to the beginning of IP where the cows received the basal TMR diet with either 13% DM pelleted SHB (SHB group) or 13% DM alfalfa pellet (CHS, Oregon, USA; CON group) during IP.Then, 4 wk of withdrawal period (WP) followed the IP, where all cows received only the basal diet.This extended period was performed to assess any carry-over effect of the SHB.

Diets
The experimental diets were formulated using Spartan Dairy Ration Evaluator/Balancer (version 3.0.3;Michigan State University, USA) to meet or exceed the nutrient requirement of lactating Jersey cows producing 19.9 kg milk/d, 4.65% fat, and 4.0% protein according to NRC (NRC, 2001) and were offered ad libitum once daily to target 10% feed refusal.The use of 13% SHB in the ration of dairy cows was based on our prior study with lambs (Parker et al., 2022), where feeding 20% SHB affected feed intake and various blood parameters while feeding 10% SHB did not affect feed intake and had a limited effect on blood parameters.Thus, in the present experiment, we provided an amount of SHB that was slightly higher than 10% to ensure observation of an effect on blood parameters without excessively compromising the feed intake.Using the alfalfa pellet, the rations were isocaloric and isonitrogenous.To further balance the rations for energy content and Ca, a commercial fat (Energy Booster Merge for Jer-seys, Milk Specialties Global, MN, USA) and dicalcium phosphate (Simplot, Lathrop, CA, USA) were mixed with the alfalfa meal for the CON diet.The composition and nutritional characteristics of the diets are provided in Table 1.
The FA profiles of the TMR, alfalfa, and SHB and the 2 dietary treatments are reported in Table 2. Compared with alfalfa, the SHB contained a proportionally lower amount of fatty acid (FA) with 16 or fewer carbons and fewer C18:3 isomers but more C18:1 and C18: 2n6 isomers.Despite those differences the 2 diets had a similar FA composition when mixed into the TMR and commercial fat added to the CON diet to balance the higher amount of fat in SHB compared with alfalfa.The 2 diets were balanced for energy, protein, Ca, and FA composition.
The SHB was donated by the Columbia Hemp Trading Company (Corvallis, OR) and delivered in pellet form.The complete chemical composition of the SHB was reported in our previously published article (Parker et al., 2022).The alfalfa pellet for the CON group and the SHB pellet for the treatment group were added and mixed manually with the TMR in the individual feed bin for each cow.To maximize the consumption of the SHB in the diet, each cow was fed sequentially with 2 separate bins, one small and one large.At 0830 h, the small bin was filled with 10 kg TMR plus supplemental feed (i.e., SHB or alfalfa pellets) and placed in the Calan feeder.At around 1500 h, the leftovers from the small bin were combined with the rest of the day's TMR in the large bin and left on offer to the cows until the following morning.

Feed Analysis and Calculation of Nutrient Intake, Consumption of SHB, and Digestibility
Daily orts from individual cows were collected and weighed.After weighing, orts from each animal were all combined, and mixed, and a subsample was collected to determine the DM content using a microwave (5 min + mixing + 5 min at full power using approx.200 g of sample) to determine DMI.Furthermore, a daily sample of the TMR was collected to determine the DM as for the orts.The DM of the TMR and orts were used to adjust the amount of feed provided on the next day to have approx.10% orts.Weekly samples of TMR and biweekly samples of orts were collected and stored for chemical analysis.In addition, orts samples from members of the SHB group were collected daily and combined every 3 d in a small bin (one per individual cow) to determine the amount of SHB pellet consumed by each animal over the course of the IP.The same was done for the CON cows until it became clear that no alfalfa pellets ever remained.The amount of SHB pellet was determined by using the upper sieve of a Penn State Particle Separator (Kononoff et al., 2003;Heinrichs, 2013), followed by DM determination using a microwave as described above.The amount of SHB consumed (as DM) was calculated as kg DM per d and as a percentage of total DMI (Suppl.Figure 2; https: / / doi .org/ 10 .6084/m9 .figshare.21804048.v7).
Parts of the dried samples of TMR and orts were finely ground to pass a 1-mm screen and then subjected to chemical composition analysis of DM, ash, ether extract (EE), fiber, and crude protein (CP) (AOAC, 2005).ADF and NDF were analyzed by using an Ankom 200 Fiber Analyzer (Ankom Technology, Macedon, NY); CP content was determined using a LECO FP828 (Leco Corporation, St. Joseph, MI, USA) for the N For digestibility, fecal samples were collected from individual cows during the IP by pooling 3 samples collected at 0730 h and 1400 h on d 26 and at 0200 h on d 27 (n = 27 samples per group), freeze-dried, and finely ground at 1-mm screen.ADL was used as an internal indigestible marker to estimate DM digestibility (DMD) and CP digestibility.Briefly, the ADL of the fecal samples was assayed by incubating ADF residues for 3 h using 72% H 2 SO 4 in a Daisy incubator (AN-KOM Technology Corp.).The ADL content was used to estimate the fecal output (FO) where FO (g/d) = [DMI (g/d) × ADL in feed (g/g)]/ADL in feces (g/g).Then, the DMD was determined using the formula: DMD (%) = [DMI (g/d) -FO (g/d)]/DMI (g/d)] × 100 (Cochran and Galyean, 2015).The CP digestibility was calculated as (%) = [CP intake (g/d) -CP output (g/d)]/CP intake (g/d)] × 100.

BW, BCS, Activity, and Milk Yield
BW and BCS were measured once during the adaptation period and every 2 wk during both the IP and WP.
Individual BW was measured using a digital scale (Afimilk, Kibbutz Afikim, Israel).BCS was determined by 5 independent researchers plus the BCS Cowdition app (Elanco, US).Daily activity [i.e., number of steps/d, laying time in min/d and min/bout, and restlessness index, an Afimilk proprietary calculation where lower values indicate better rest (Silper et al., 2017) and milk yield were automatically recorded by the Afimilk system (AfiMilk, SAE Afikim, Afikim, Israel).Milking occurred twice a day, at 0600 and 1700 h.

Milk Component and Fatty Acids Profile
Milk samples were collected at 5 d intervals throughout the experiment (both AM and PM milking).Two milk samples were collected: one for milk component analysis in 60 mL cupped dairy vials (cat# CPP03EDM-CL, Capitol Plastic Products, NY) containing 1 tablet of Broad-Spectrum Microtabs II preservative (Weber Scientific, and the other in 15-mL centrifuge vials for milk FA profiling which was stored at −20°C until analysis. Milk components were measured using a LactoScope FT-A for milk fat, protein, lactose, C16:0, C18:0, C18:1, total solids, solid nonfat, other solids, de novo FA, mixed FA, preformed FA, estimated plasma nonesteri- fied fatty acid (NEFA), BHB, acetone, and unsaturated FA.A SomaScope (PerkinElmer, USA) was used for SCC.Both instruments were calibrated monthly using 14 calibration samples from Cornell University (Portnoy et al., 2021).The milk FA profile was quantified for each sample using instrument, protocol, and column as described by Oeffner et al. (2013) using C22:0 as the internal standard (cat# 11909, Sigma-Aldrich, St. Louis, MO, USA).The samples were analyzed using an Agilent 6890 GC series equipped with a flame ionization detector, CP-8400 autosampler, and ChromPerfect software (NJ, USA).For the GC instrument, the injector temperature was set at 250°C and the detector temperature was 260°C.The oven temperature was initially held at 140°C for 5 min and increased to 190°C at 15°C/min, held for 25 min, then ramped at 5°C/min to 220°C and held for 5 min.Forty FA were annotated using a TraceCERT(R) 37 Component FAME Mix certified reference material (cat# CRM47885, Supelco, Bellefonte, PA, USA) plus reference standards for rumen biohydrogenation intermediates (mix of methyl 9,11 and 10,12 conjugated linoleate, cat# UC-59M, and methyl transvaccenate, cat# U-49M, Nu-Check, Elysian, MN, USA).The reference standards were also used to build a 5-point standard curve in triplicate to calculate recovery.The FA data were presented as g FA/100 g of all annotated FA.

Blood Parameters
Blood samples were collected immediately after morning milking at baseline (d −12 before the experiment) and d 10, 25, 38, and 52 of the experimental periods (i.e., 2 collections each during the IP and the WP).Approximately 10 mL of blood samples were collected from the jugular vein into evacuated tubes (Becton Dickinson and Company, USA) containing lithium heparin (Cat# 366480) or EDTA (Cat# 366643).Heparinized tubes to obtain plasma were immediately put in ice after collection and centrifuged within 1h after collection.Non-heparinized tubes (Cat# 366430) to obtain serum were kept at body temperature for at least 30 min before centrifugation.EDTA-collected blood was used for complete blood count analysis using a VetScan HM5 (Abaxis, USA).Plasma and serum from each cow were transferred to 3 1.7 mL microfuge tubes (Olympus, USA) and immediately stored at −20°C.

Phagocytosis
The phagocytic activity of polymorphonucleated cells (PMN) was quantified once at baseline, once during the IP (d 26) and twice during the WP (d 39 and 53) using PHrodo BioParticles Phagocytosis kit (Cat#P35382, PHrodo Green S. aureus BioParticles, Life Technologies) following the manufacturer instructions until the last step where, after the supernatant was discarded, the cells were resuspended in 200 µL of primary antibody solution containing the anti-PMN antibody (MM12A, IgG anti-CD11b, Monoclonal Antibody Center, Washington State University, Pullman, WA 98164) in PBS, incubated on ice for 30 min, and centrifuged at 1000 × g for 5 min at 20°C.After centrifugation, the pellet was resuspended in 200 µL of secondary antibody solution in PBS (Allophycocyanin Goat anti-Mouse IgG, Cat# 31981, Thermo Fisher Scientific, Rockford, IL 61101) and DAPI and incubated for 30 min on ice.The samples were centrifuged for 5 min at 1000 × g at 20°C, the supernatant was discarded, and the pellet was resuspended with 1 mL of PBS before flow cytometer analysis using a Beckman Coulter CytoFLEX Cytometer.Data were analyzed by gating granulocytes using the MM12A antibody to determine the phagocytosis of PMN using the software associated with the instrument.

Neutrophil Migration Assay
Neutrophil migration analysis was performed once at baseline and twice during the IP period (d 11 and d 26, respectively).Cellular migration was assayed using a commercially available kit (Cat# CBA-104, Cell Biolabs Inc., San Diego, CA).Briefly, 100 µL of whole blood collected in EDTA vacutainer tubes was added to a migration plate containing 150 µL of endotoxinactivated sheep plasma as the chemoattractant.The latter was obtained according to Jahan et al. (2015) by treating each mL of whole blood collected in heparinized vacutainer tubes with 5 µg LPS from E. coli Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows EH100 (cat#IAX100010M001, Adipogen, USA) for 4 h at 37°C.The plasma was immediately obtained after centrifugation at 8500 × g for 10 min at 4°C and stored at −80°C in 1 mL aliquots.Following 24 h incubation, cells were detached from the migration plate using the provided cell detachment solution.Following the detachment, the solution was mixed with the provided buffers and dye solution, incubated, and added to a black 96-well plate for fluorescence measurement.As blank, RPMI media was used instead of the LPSactivated plasma.Fluorescence was measured using a Biotek Synergy H1 Plate Reader at 480nm/520nm and data were corrected for the blank and by the known number of neutrophils as determined by the complete blood count.This correction was performed by taking the fluorescence data for each sample and dividing it by the percentage of neutrophils for the corresponding sample.

Nitrogen Metabolism
An aliquot of the freeze-dried fecal samples was subjected to total N content analysis by using an automatic N analyzer (LECO FP828, MI, USA).At d 26 of the IP, an indwelling 26fr foley catheter (cat#J0447Q, Jorgensen Labs, distributed by Covetrus, USA) was placed in 16 cows (9 in CON and 7 in SHB; due to technical issues, catheters were not placed in 2 cows in the SHB group), and urine was collected in collection bags (Ugo single use T-Tap 2L bags or 2 L Pevor Urinary Drainage Bags).Urine from individual cows was collected every 2 h for 36 h and the urine volume was measured using a digital scale.Ten mL of urine sample was collected in 15 mL tubes at each collection point, and 1 mL of 0.1 N sulfuric acid solution was added to prevent NH 3 -N volatilization.The tubes were immediately stored at −20°C until analysis.
Total purine derivatives (PD) (g/d) were calculated as the summation of allantoin and uric acid concentrations multiplied by the urine volume (L/d).Nitrogen intake (g/d) was calculated as N % × DMI (g/d) and N output (g/d) was calculated as urine N (g/d) + fecal N (g/d) + milk N (g/d).Whole N balance (g/d) was calculated as N intake (g/d) -N output (g/d).The ratio of PD to creatinine was calculated as an indirect index of microbial protein production (MPP).Nitrogen use efficiency was calculated as N excreted in milk (g/d) divided by the N intake (g/d).

Methane Emissions
Methane (CH 4 ) gas production was collected from d 19 to d 25 of the IP using the SF 6 tracer method with an evacuated PVC canister fitted to the individual animal plus 2 canisters were located in the barn close to the Caln gate pen to correct for methane background, and samples were analyzed essentially as previously described (Wilson et al., 2020).The only modification was the new design of the equipment for the collection of the gas (Suppl.Figure 1; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).Due to sampling errors, only data collected on d 24 and 25 were used.

Statistical Analysis
Potential outliers were examined by using the PROC REG procedure of SAS.Data with studentized t residuals ≥ 2.8 were removed before analyses.Final data were analyzed using a repeated measure of a generalized linear mixed model (PROC GLIMMIX, SAS 9.4) with treatment (TRT), time (T), and their interactions (TRT × T) as fixed effects and cows as a random effect with time as a repeated measure.The statistical analysis was performed for each period separately to examine the TRT, Time, and TRT × Time interaction effects and in a whole period to examine the TRT, Period, and TRT × Period interaction effects.The baseline data of each animal were used as a covariate for all the parameters except for milk fatty acid profile, methane emission, neutrophil migration and phagocytosis, and nitrogen metabolism-related parameters, including urine volume.Depending on the variable measured, baseline data were measured between d −14 and d −3 before the cows in the SHB group received the treatment diets.For DMI and milk yield, the mean of 3 d before starting to feed SHB was used as a covariate.Several covariance structures including ARH(1), AR(1), and SP(POW) were evaluated by Akaike Information Criterion value and model, and the lowest value was used as the covariance structure statement in the model.Methane production, milk FA profile, migration of neutrophils, phagocytosis of leukocytes, and N metabolism were analyzed using a GLM procedure because these data were recorded only during the IP without a baseline.Significance was declared with P ≤ 0.05 and tendency was discussed at 0.05 < P < 0.10.Results are presented as least squares means (LSM) and the highest standard error of the means.
One cow in the SHB group was removed from the final data set because the cow became able to access the Calan gate of a control cow.To keep the cow in the experiment, it was moved into an individual pen.However, likely because of the move, the cow had high SCC and several other parameters were affected by the move; thus, the cow was removed from the final data set, leaving 9 cows in the CON group and 8 cows in the SHB group.

Performance and Activity of the Cows
The intake of pelleted SHB was (mean ± SD) 7.5 ± 3.3% of DMI or 1.22 ± 0.55 kg DM/d.The daily intake vs. offered SHB is available in Suppl.Figure 2 (https: / / doi .org/ 10 .6084/m9 .figshare.21804048).Table 3 summarizes the main effect of feeding SHB during the 4 wk of IP and the 4 wk of WP.During the IP, feeding SHB did not affect milk yield but decreased (P = 0.001) DMI by 6.7%.The DMI was similar between groups during the WP.Cows fed SHB produced more milk (P = 0.03) during WP than cows in the CON group, resulting in higher (P < 0.001) kg milk yield/kg DMI for SHB vs CON over the 2 periods (Suppl.Figure 3; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).However, ECM and FCM were comparable between the groups in each period, but a TRT × Period interaction (P = 0.04) was observed for both calculated parameters due to the increase of their values in cows fed SHB vs. CON in WP than IP (Suppl.Figure 4; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).The lack of TRT effect for ECM and FCM was due to the tendency (P = 0.06) for lower fat percentage and yield in SHB vs CON group during the IP (Table 4 and Suppl.Figure 4; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).Consequently, dairy efficiency calculated as ECM/DMI or FCM/DMI was similar between groups during the IP but tended to be higher for the SHB group during the WP (P = 0.097 and P = 0.084.respectively; Table 3).Feeding SHB did not affect BW, BCS, ΔBCS, or ADG (Table 3).Using the data collected by the Afimilk system, we followed up the subsequent lactation performance of 8 cows of CON and 6 cows of the SHB group and detected only a numerical higher milk yield for SHB vs. CON (2.5 kg/d difference; P = 0.12) (Suppl.Table 1 and Suppl.Figure 5; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).
We did not observe much effect of feeding SHB on the activity of dairy cows throughout the 2 periods (Table 3 and Suppl.Figure 6; https: / / doi .org/ 10 .6084/m9 .figshare.21804048)or in the following lactation (Suppl.Table 1 and Suppl.Figure 5; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).However, during the first 2 wk of IP, cows fed SHB had less (P = 0.009) average total laying time (min/d) than CON but had similar lengths of lying bouts (min/bout) and restlessness ratios.

Milk Components and Quality
A tendency for a lower percentage of milk fat (P = 0.06), particularly due to a tendency for a lower amount of preformed fatty acids (P = 0.09) and acetone (P = 0.06) and a significantly lower percentage of C18:0 and percentage of de novo FA in SHB vs. CON was detected (Table 4).None of the other milk components measured was affected by TRT during the IP (Table 4).

Milk Fatty Acids Profile
Individual milk FA profiles were calculated as proportions of total FA (Suppl.Table 2 in https: / / doi .org/ 10 .6084/m9 .figshare.21804048).During the IP, the only FA that were proportionally different in milk between the 2 groups were C15:0 (P = 0.03) and C18: 1cis9 (P = 0.04), which were in higher concentrations in milk fat of SHB vs. CON.Thus, as the C15:0 FA is synthesized by bacteria, the proportion of bacterial FA tended (P = Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows 0.06) to be higher in SHB vs CON.In addition, the estimated overall delta-9 desaturase activity tended (P = 0.08) to be higher in SHB vs. CON.The FA in the milk fat from SHB cows had a tendency for a higher (P = 0.07) hypocholesterolemic/hypercholesterolemic and a lower (P = 0.03) thrombogenic index than CON cows.During the WP, the only observed difference in FA was a higher proportion of C8:0 (P = 0.03) and C18: 2t9 ,c12 (P = 0.05) in the butterfat of SHB vs. CON.

Blood Parameters and Leukocyte Phagocytosis
The complete blood count, neutrophil migration, and phagocytosis data are presented in Suppl.Table 3 (https: / / doi .org/ 10 .6084/m9 .figshare.21804048).None of those data were affected by SHB treatment, except for an overall larger (P = 0.04) number of monocytes in SHB vs. CON during the whole experiment.
Among metabolic parameters (Table 5 and Suppl.Figure 8; https: / / doi .org/ 10 .6084/m9 .figshare.21804048), the concentration of cholesterol was lower (P < 0.001) during the IP and tended (P = 0.07) to be lower during the WP in SHB vs CON, while the concentration of BHBA was higher (P = 0.036) in SHB vs. CON during the WP, and a TRT × Time interaction was observed for protein (P = 0.015), driven by the same effect on globulin (P = 0.033), and for P (P = 0.002) and Mg (P = 0.007) during the IP.Those interactions were due to a temporal change in the CON rather than the SHB group.
The concentration of ceruloplasmin (P = 0.004) and ROM (P = 0.032) were higher in SHB vs. CON during the IP with a concomitant decrease through time of paraoxonase and Zn, with the latter having higher values in SHB vs. CON at 10 d of the IP (Table 5 and Suppl.Figure 9; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).

Nutrient Intake, Nitrogen Metabolism, and Methane Production
During the IP, intake of NDF, ADF, ME, NE L , and N in cows fed SHB was lower (P < 0.01; Table 6) than CON.The CON group ate between 3% and 10% more than the requirement for all the above parameters while the SHB group ate between 3% and 16% lower than the requirement, with CP being 104% of the requirement for the CON and 89% for the SHB group (Suppl.Table 4 in https: / / doi .org/ 10 .6084/m9 .figshare.21804048).
Urinary N content (%) was not different between the dietary treatments (Table 6) but was 19.5% lower (P = 0.05) for SHB vs. CON group when expressed as daily excretion due to 23% lower urine output (P = 0.05) (Suppl.Figure 11; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).The daily urinary N excretion was lower in SHB vs. CON (P = 0.05) due to a lower daily excretion of NH 3 (g/d) and urea (g/d) (P = 0.06 and P = 0.03, respectively), while the concentration of those parameters was almost identical between the 2 groups (Table 6).Creatinine in urine was not affected nor was the concentration of purine derivatives allantoin and uric acid, but the daily excretion was lower in SHB vs. CON for uric acid (P = 0.04) and tending toward lower PD (P = 0.10) (Table 6).The estimated MPP tended (P = 0.10) to be lower in SHB vs. CON cows, but the ratio of nitrogen in MPP/N intake was not different.
N content of feces tended to be lower in SHB vs. CON (P = 0.086), but the fecal output (kg DM/d) and the N output in feces (g/d) were not different (Table 6).Milk N concentration and output did not differ, but the efficiency of N utilization was higher in SHB vs. CON (P = 0.01).
The digestibility of the DM and CP as well as the N balance were not affected by feeding SHB (Table 6 and Suppl.Table 5; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).SHB in the diet had no effect on CH 4 or CO 2 production when expressed as daily emission (g/d) or relative to milk yield or DMI (Suppl.Table 5; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).

SHB has low palatability affecting feed intake
SHB is a high-quality feed ingredient for ruminants, especially given its high CP, energy, and mineral content (Zn, Cu, and Mn), and low NDF content, as discussed previously (Parker et al., 2022).Nutritive analysis of the SHB used in our experiment is very similar to one study conducted in China (Wang et al., 2023).Despite the high nutritive quality, our experiment revealed a low palatability of SHB for dairy cows, as demonstrated by the amount of SHB eaten vs SHB offered, resulting in about 7% lower DMI in cows fed SHB vs CON in the IP.In our previous experiment with lambs (Parker et al., 2022), a similar detrimental effect on DMI caused by low palatability of SHB was also observed, where the animals tended to reject the SHB at the commencement of the experiment.In that study, a 14.6% decrease in DMI during the first month of the experiment was    Mean number of unsaturation per FA (Wojciechowski and Barbano, 2016) 3 Mean chain length as carbon per FA (Wojciechowski and Barbano, 2016) 4 TRT = P -value of treatment effect; TRT × T = P -value of interaction effects between treatment and time; TRT × P = P -value of interaction effects between treatment and period (IP vs WP) Means within a row with different superscripts differ (P < 0.05).
2 TRT = P -value of treatment effect; TRT × T = P -value of interaction effects between treatment and time; TRT × P = P -value of interaction effects between treatment and period (IP vs WP).
observed when lambs were fed SHB at a 20% level, with no effect when fed at 10%.This low palatability was probably due to the sensory characteristics of SHB.
Studies on specific metabolites contributing to the scent of Cannabis sativa (Oswald et al., 2021) have identified 3-methyl-2-butene-1-thiol among > 200 metabolites.That compound has a strong smell and is part of the "skunk odor" commonly ascribed to cannabis (Wood, 1990).The drying and extraction of cannabinoids from the hemp could have concentrated the compound thereby affecting the odor that was likely retained by pelletizing the SHB.Additional compounds that could have affected the intake are flavonoids, as those are known to decrease feed intake in dairy cows (Zhan et al., 2017).We did not measure any of those compounds in the SHB used in our study, and it is not known if cows can detect them.
In a recent study conducted in Germany, Wagner et al. ( 2022) reported that feeding a high dose of hemp silage (1.68 kg/d or 11.1% diet) containing 1% cannabinoids (0.82% CBD) to lactating Holstein cows for 6 d decreased DMI by around 16%, but there was no dif-  a-b Means within a row with different superscripts differ (P < 0.05) 1 Estimated from excretion of urea (g/d) to creatinine (g/d) 2 Milk N efficiency was calculated as milk N/N intake × 100 3 Urinary excretion as mg /BW 0.75 according to Balcells et al. (1991) 4 MPP (Microbial protein production) = ({[PD production, mmol/d -(0.385 × BW 0.75 )]/ 0.85} × 70 × 6.25)/ (0.13 × 0.83 × 1,000) (Janicek et al., 2008). 5Nitrogen MPP/Nitrogen intake ference in DMI when feeding 0.84 kg/d (5.6% diet) of hemp silage.The authors suggested as a possible cause of the reduced feed intake either the higher amount of fat in the silage or the presence of cannabinoids, especially Δ 9 -THC and CBD.This idea cannot be discarded.
In our previous study with lambs (Parker et al., 2022) where the animals were fed 10% or 20% SHB in their diets, ingesting 1.18 ± 0.18 and 2.12 ± 0.20 mg Δ 9 -THC /kg BW and 82.9 ± 12.6 and 148.5 ± 13.9 mg total CBD/kg BW, respectively, a clear decrease in feed intake was observed only in lambs fed 20% SHB and only during the first month.Wagner et al. (2022) reported an intake of 1.6 mg Δ 9 -THC /kg BW and 10.7 mg CBD/kg BW for the cows receiving 0.84 kg/d of hemp silage and 3.1 mg Δ 9 -THC /kg BW and 20.4 mg CBD/kg BW for the cows receiving 1.68 kg/d of hemp silage.The cows in the present study ingested < 2-fold lower amount of Δ 9 -THC (0.79 ± 0.17 mg /kg BW) but more than twice the amount of CBD (55.2 ± 12.2 mg /kg BW) than the cows in the study of Wagner et al. (2022) receiving 0.84 kg/d of hemp silage showing no statistical effect on feed intake.Furthermore, our cows had a smaller decrease in DMI despite receiving a higher amount of CBD than the cows in the study of Wagner et al. (2022) that received 1.68 kg/d of hemp silage.Thus, the above observations would indicate that neither Δ 9 -THC nor CBD is likely the cause of the reduction of DMI in dairy cows.However, in a recent Chinese study, lactating Holstein cows fed between 6 or 11% SHB containing 0.03% CBD (i.e., approx.10-fold lower than our study) reported no statistically significant effect on DMI (Wang et al., 2023).Although there was no effect on DMI, lower CP intake was observed in the Chinese study.As the level of CP was the highest in SHB than all the other feed ingredients, the data might support a low palatability also for the SHB in the Chinese study.The nutritional composition of the SHB in the Chinese study was very similar to our study, except for a higher NDF.Overall, the combined data from the 3 studies do not fully support the role of CBD on DMI.Furthermore, our previous study in lambs indicated that the long-term voluntary intake of 20% SHB was similar to the control diet (Parker et al., 2022) suggesting the possibility that cows could also adapt to eating SHB.

Feeding SHB does not affect performance with minimum effect on milk components
Lowering DMI generally results in the depression of milk production.However, cows fed SHB had similar milk production as well as FCM and ECM throughout the IP and surprisingly a higher milk yield than CON during the WP.The lack of an effect on performance, despite lower DMI in animals fed SHB, was also observed in the study with lambs (Parker et al., 2022), while no effect on performance was observed in lactating cows SHB in a Chinese study (Wang et al., 2023).The reason for the lack of any negative effect on performance is unclear.Contrary to our and the trial of Wang et al. (Wang et al., 2023), Wagner et al. al. (2022) found a decrease in milk yield of cows receiving hemp silage.The major differences between our study and the study of Wagner et al. (2022), besides the different hemp material (SHB vs hemp silage), were the use of a different breed of cows (Jersey vs. Holstein), the number of cows used in each group (9 vs. 4 cows), and the length of feeding (4 wk vs. 6 d).However, in our study we did not observe any decrease in milk yield in the short time after initiating feeding SHB, i.e., during the first wk of the IP (Suppl.Figure 3; https: / / doi .org/ 10 .6084/m9 .figshare.21804048).The stage of lactation was not reported in the study of Wagner et al. (2022).
The lack of an effect on most of the milk components, as also observed by Wang et al. (2023), supports SHB as a potential substitute for alfalfa or other forages in dairy animal diets.The cannabinoids in SHB, especially CBD, could improve milk fat synthesis by activating PPARγ (Peng et al., 2022), a potential key transcription factor in controlling milk fat (Osorio et al., 2016).However, our study, the study of Wagner et al., 2022), and -although using SHB with low concentration of cannabinoids -the study of Wang et al. (2023), do not support this effect; rather, we observed a tendency for a decrease in milk fat, especially synthesis of de novo FA during the IP.No difference in the level of conjugated linoleic acids or other trans-intermediates was observed in our study, indicating that the effect was not due to classical milk fat depression (Palmquist, 2009;Osorio et al., 2016).
The data indicated a higher proportion of bacteriaderived FA in the milk of cows fed SHB, especially C15:0, which is associated with an increased abundance of bacteria that degrade pectin and ferment sugar (Fievez et al., 2012).This might indicate some effects of SHB on the rumen microbiota.The increase in the proportion of C15:0 might indicate a change in rumen fermentation that could have also affected the production of acetate and propionate during the IP.A shift of the ratio toward higher propionic vs. acetic acid production could have a role in the decrease of milk fat (Osorio et al., 2016) and the maintenance of milk production despite the lower DMI (Seymour et al., 2005;McKay et al., 2019;Totakul et al., 2022).The above speculation needs to be verified by measuring VFA in the rumen.In the study with lambs, we observed a higher digestibility of the diet containing 20% SHB but not the 10% SHB Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows diet (Parker et al., 2022).Similarly, we did not observe any effect on digestibility when providing 13% SHB to lactating cows, although only approximately half that amount was ingested in the present study.

SHB does not affect the activity of the cows
In our study, SHB did not affect the activity of the cows as shown by similar steps per d, laying times as measured by min/bout, and restlessness ratios between groups.In a recent study, a longer lying time was observed in dairy heifers when administered by 5.5 mg/kg BW of cannabidiolic acid (CBDA) without any effect on activity (steps/d) (Kleinhenz et al., 2022).Wagner et al. (2022) argued that hemp silage affected the behavior of the animals due to the observed decrease in respiration and heart rate.The SHB used in our experiment contained 1.88% CBDA (Parker et al., 2022) corresponding to an intake of 46.3 ± 10.2 mg CBDA/kg BW in the cows in our experiment during the IP, a > 8-fold higher ingestion of CBDA than the heifers in the study of Kleinhenz et al. (2022) and > 2-fold higher ingestion of CBD than the cows in the study of Wagner et al. (2022).Despite the higher intake of CBDA and total CBD, we confirmed that activity was not particularly affected.We did not notice any obvious change in the cows' behavior, but for SHB vs. CON cows during the IP, we did observe a 53 min/d lower total laying time in the SHB group.The lower daily time but a similar number of steps/d in SHB vs. CON may be due to the time the SHB cows spent at the feeder sorting out the SHB from the TMR instead of laying.

SHB has minimal effect on metabolism
In the experiment with lambs (Parker et al., 2022), dietary SHB had a significant effect on the blood metabolic profile of the animals, especially when fed 20% SHB in the diet.SHB inclusion at 20% had a limited benefit to the immune system but apparently a better antioxidant response, with minimal effect on liver health.Feeding 10% SHB did not affect feed intake and had less effect on metabolism and liver function but improved the antioxidant status when fed for 2 mo in lambs.No effect on metabolic-related parameters was observed by feeding 6 or 11% SHB to Holstein dairy cows (Wang et al., 2023).
In the present experiment, the decrease in feed intake was insufficient to induce a significant increase in NEFA, as would be expected (Kuhla et al., 2016;Leduc et al., 2021).Activation of cannabinoid receptors appears to inhibit lipolysis in the adipose tissue, as recently reviewed (Myers et al., 2021).Cannabinoids, especially THC, can also induce adipogenesis and lipogenesis in the adipose tissue (Myers et al., 2021), competing with the mammary gland for the triglycerides present in the very low-density lipoproteins (Bionaz et al., 2020).This could have explained the tendency for lower milk fat in SHB vs. CON.However, the activation of lipogenesis in the adipose tissue of cows fed SHB would have resulted in a higher BCS, but in our experiment, it was not statistically affected.
The negative effect on the concentration of cholesterol in blood seen in cows fed SHB could be partly explained by the lower DMI.However, the lower concentration of cholesterol persisted for a period after withdrawal of SHB, where there was no difference in DMI between groups.Furthermore, in our prior study with lambs, feeding 10% SHB decreased circulating cholesterol occurred without a change in DMI (Parker et al., 2022).Thus, as previously argued, the observed data may be explained by the effect of cannabinoids on the expression of hepatic Apolipoprotein A-I, which has a functional role in HDL production (Haas et al., 2012;Bionaz et al., 2020).As previously observed (Parker et al., 2022) the proposed effect on HDL is partly supported by the detected decrease in paraoxonase during the IP in the present study.
Consistent with feeding 20% SHB in our prior study with lambs, we detected an increase in BHBA once SHB was withdrawn from the diet (Parker et al., 2022).As for the study with lambs, the higher plasma BHBA was likely the consequence of a change in the production of butyrate in the rumen, as butyrate is the main precursor of BHBA (DeFrain et al., 2004).Greater BHBA production might explain the higher milk production during the WP, as, speculatively, higher butyrate (and, possibly, propionate production) in the rumen might have increased the synthesis of lactose (Gessner et al., 2015).Although it was in very low concentration in the SHB used in our experiments (Parker et al., 2022), THC has been found to affect bacteria composition in human lungs and gut with an increase in propionic acid production mediating an anti-inflammatory response (Mohammed et al., 2020).Measurement of the volatile fatty acids in the rumen would be necessary to confirm the effect of SHB on rumen microbes.However, if true, the increased difference in BHB between the 2 groups during the withdrawal period might indicate a positive long-term effect of SHB on the rumen microbiota and fermentation.Inconsistent with our study and contrary to our suggestion, a decreased ruminal concentration of butyrate was observed in Holstein cows fed 11% SHB with no changes in the composition of the ruminal microbiota (Wang et al., 2023).As indicated above, the level of cannabinoids in the SHB of the latter study was substantially lower than in the SHB we used.

SHB had a minimal effect on inflammation and oxidative stress and no effect on health
Data on immune and oxidative stress parameters indicated that the animals consuming SHB experienced minor inflammation and oxidative stress.Those data are contrary to prior studies.Feeding SHB to lambs improved the level of FRAP but did not affect inflammatory markers (Parker et al., 2022).A lack of effect on inflammatory markers, specifically the positive acute phase protein haptoglobin, was also seen in dairy heifers fed hemp flowers (Kleinhenz et al., 2022).A decrease in the level of the pro-inflammatory interleukin 1β was observed in Holstein cows fed 6 or 11% SHB (Wang et al., 2023).In the present study, acute inflammation was not present, as haptoglobin, one of the major markers of inflammation (Bionaz et al., 2007), remained unchanged.However, the negative effect on paraoxonase, Zn, and (with a tendency) albumin, associated with a higher concentration of the positive acute phase protein ceruloplasmin (Mohiuddin and Manjrekar, 2018) in plasma of cows fed SHB vs CON during the IP could be indicative of low-grade inflammation.The level of ceruloplasmin may be also associated with a higher amount of Cu in the diet (Twomey et al., 2005), as this element was > 2-fold more abundant in SHB than alfalfa (Parker et al., 2022).The higher ROM also supports the possible low-grade inflammation (Mezzetti et al., 2020), which may be a consequence of CBD metabolism, as this produces a substantial amount of reactive oxygen species (Usami et al., 2008).In our prior study, however, no difference was found in the concentration of ROM in lambs fed 20% SHB which might be due to a longer feeding period (60 d) (Parker et al., 2022).Similarly, no effect on oxidative status-related blood parameters was observed in Holstein cows fed up to 11% SHB (Wang et al., 2023).

SHB negatively affects liver clearance
Consistent with our prior study in lambs (Parker et al., 2022), we observed a decrease in the ability of the liver to clear bilirubin without any detrimental effect on the liver, as indicated by the lack of an effect on the various liver-related parameters measured, such as GGT and GOT.The activity of ALP, another marker of liver health, was overall increased by feeding SHB; however, the change of this parameter might be due to an effect of CBD on bone metabolism rather than as a consequence of liver issues, as previously argued (Parker et al., 2022).The levels of plasma bilirubin that we observed were in the normal ranges for dairy cows (Bayat et al., 2022).As previously discussed, the increase in bilirubinemia is likely a consequence of the inhibition of CBD on bilirubin transporters, which could also negatively affect drug clearance (Anderson et al., 2022;Parker et al., 2022).The latter can have important repercussions on the use of drugs in animals; thus, a direct assessment of the effect of SHB on the drug clearance system of the liver is warranted.

Data on nitrogen metabolism provide support for an improved N efficiency that is not due to a change in ruminal metabolism
Our data do not support an effect on enteric methane production by feeding SHB but we observed an improvement of N use efficiency.Other parameters related to N metabolism, including MPP, were not greatly affected as also previously observed (Wang et al., 2023); however, a few observations can be made from the data.The 23% decreased urine volume in cows fed SHB might indicate an antidiuretic effect of the SHB, similar to what was suggested in cows fed hemp silage (Wagner et al., 2022).The diuretic effect is indirectly supported by a recent observation of a 29% lower water intake and a 43% lower urine output in lambs fed green hemp than a control diet, although the latter was not statistically different (Stevens et al., 2022).Bioactive compounds present in plants can affect urine volume in dairy cows (Mangwe et al., 2019).Somewhat surprisingly, the lower urine volume detected in our study did not result in a higher concentration of N in the urine.This is partly explained by the 15% lower N intake in SHB vs. CON.The 22% decrease in urea excretion in cows fed SHB after a 7% decrease in DMI is similar to earlier studies where a decrease in N intake of 13-14% due to restricted or lower CP diets resulted in 22-30% lower urea excretion in urine (Chibisa and Mutsvangwa, 2013;Stevens et al., 2021).
The detected increased N use efficiency in cows fed SHB is likely due to a greater uptake of N by the mammary gland rather than a change in rumen N utilization as indicated by the tendency for lower MPP with similar proportion of N utilization for MPP between cows fed SHB and CON (Kaswari et al., 2007;Zhu et al., 2013;Lu et al., 2019), indicating that the higher efficiency of N utilization was not due to a change in the microbiota of the rumen.A recent SHB feeding study from China might support this statement where they found a decrease in total VFA production and very minor changes of rumen microbiota by feeding 6 or 11% SHB replacing alfalfa to lactating Holstein cows (Wang et al., 2023).

Limitations
The lack of data on rumen VFA and microbiota has precluded a clear conclusion on the effect of SHB on the rumen.Such data could have helped explain the effect observed on BHB, FA in milk, and milk production and composition.The use of late-lactation cows is also not typical and could have hidden some effects that could be observed in other stages of lactation.The reason for using late-lactation cows in our study was prompted by the Food Use Authorization obtained from the FDA that required drying off the cows at the end of the study.Several milk parameters measured by the LactoScope that used the Fourier Transformed Infrared System to estimate amounts of each milk component, although reported, should be taken with caution as BHBA, acetone, and the various fatty acids can be useful parameters at herd level, but the accuracy is not adequate for analysis at the cow level (van der Drift et al., 2012).The digestibility data are limited by the fact that samples were collected from only 2 d, which could be considered adequate if collected 12 and 24 h post-feeding (Cavallini et al., 2023) whereas typically they should be collected for 4 d (Weiss et al., 2009).Another limitation in the digestibility data was the use of ADL as an internal marker for the estimation of DMD Sunvold and Cochran, 1991); although it cannot provide accurate absolute values it suffices to detect differences between groups (McAvoy et al., 2020).

CONCLUSIONS
Our study confirmed a low palatability of SHB for ruminants, which likely caused the detected lower DMI in lactating cows.Although the lower DMI did not compromise the milk yield in these late-lactation cows, this may present a major challenge for including SHB in the diet of high-yielding dairy cows.The increase in milk production after the withdrawal of SHB along with the possibility of a long-term adaptation of the animals to eat SHB are of interest, as these observations might indicate that feeding SHB could prime the animals for higher efficiency.Interestingly, feeding SHB improved the efficiency of N utilization.Contrary to prior studies, our investigation did not reveal any effect of SHB on the activity of the animals.Our extensive analyses indicated that SHB may affect the metabolic status of dairy cows but did not affect the immune system or any health-related parameters measured, except for a possible low-grade inflammation as indicated by an increased plasmatic level of ceruloplasmin and ROM.Similar to our prior study with lambs, the increase in bilirubin concentration in the plasma of dairy cows after feeding SHB may be indicative of compromised liver clearance capacity, which could also affect drug clearance.
Overall and in line with our initial hypothesis, feeding SHB with a relatively high level of cannabinoids decreases feed intake but does not affect lactation performance or the health of the animals, at the least based on the measured health-related parameters, providing the first evidence that SHB might be feasible for inclusion in the diet of lactating dairy cows.
Figure 1.Experimental design of the study, where spent hemp biomass was fed to dairy cows for 4 wk (intervention period, IP), followed by 4 wk of withdrawal period (WP).Reported in the figure is the timeline for the collection of various samples.CON = diet supplemented with 13% DM of alfalfa meal; SHB = diet supplemented with 13% DM of spent hemp biomass (SHB).
Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows Irawan et al.:  Spent hemp biomass as a feed for lactating dairy cows Table3.Feed intake, milk production, BW, BCS, and activity of lactating Jersey cows fed spent hemp biomass (SHB) for 28 d (Intervention) followed by 28 d post-SHB removal from the diet ( Means within a row with different superscripts differ (P < 0.05) 1 ECM = [(0.327× kg milk yield/d) + (12.95 × kg of fat) + (7.20 × kg of protein)]; FCM = [(0.4324× kg of milk yield/d) + (16.216 × kg of fat)]; ADG = average daily gain (kg/d) calculated bi-weekly; ΔBCS = difference of BCS measured compared to BCS of baseline 2 TRT = P -value of treatment effect; TRT×T = P -value of interaction effects between treatment and time; TRT×P = P -value of interaction effects between treatment and period (IP vs WP) Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows

Table 1 .
Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows Ingredients and chemical composition of the diets method.Intake of CP, NDF, ADF (kg/d), and ME and NE L (Mcal/d) were determined by the amount of the nutrients offered subtracted by the amount left in the orts. combustion

Table 2 .
Irawan et al.:Spent hemp biomass as a feed for lactating dairy cows Fatty acid profiles of ingredients and diets

Table 4 .
Milk components of lactating Jersey cows fed spent hemp biomass (SHB) for 28 d (Intervention) followed by 28 d post-SHB removal from the diet (Withdrawal) a-b Means within a row with different superscripts differ (P < 0.05). 1 de novo FA are from C4 to C15; Mixed FA are C16:0 and C16:1, and preformed FA are C18 and longer.2

Table 5 .
Blood parameters related to metabolism, minerals, inflammation, and oxidative stress in lactating Jersey cows fed spent hemp biomass (SHB) for 28 d (Intervention) followed by 28 d post-SHB removal from the diet (Withdrawal) a-b

Table 6 .
Irawan et al.:Spent hemp biomass as a feed for lactating dairy cows Nutrient intake and nitrogen metabolism of lactating Jersey cows fed spent hemp biomass (SHB) Irawan et al.: Spent hemp biomass as a feed for lactating dairy cows