Utilizing Plant Leaves to Create Novel Feed Pellets to Encourage and Improve Kalasin Province’s Beef Cattle Production

Abstract

The effects of substituting pelleted diets manufactured from cassava, chaya, and mulberry leaves for concentrate on growth performance, feed intake, rumen fermentation, and microbial protein synthesis in beef cattle were evaluated. Four beef cattle (initial BW: 250 ± 50 kg) were assigned to four treatments: a control diet (T1) and diets in which 50% of the concentrate was replaced with cassava leaf pellets (T2), chaya leaf pellets (T3), or mulberry leaf pellets (T4). The data were analyzed using a 4 × 4 Latin square with animal as a period effect as appropriate. Rumen volatile fatty acids were determined by means of HPLC, and microbial protein synthesis was assessed using urinary purine derivatives. Cattle fed cassava leaf pellets (T2) showed the greatest average daily gain (0.79 kg/d) compared with the control (0.50 kg/d; p < 0.05). Compared with T1, T4 exhibited a higher ruminal propionate proportion and total VFA concentration, which was associated with a lower acetate-to-propionate pattern, suggesting reduced methanogenic potential. No adverse health effects were observed, as indicated by hematocrit and blood urea nitrogen values within normal ranges. Microbial protein production increased in the leaf-pellet treatments, with T4 showing the highest efficiency. Overall, cassava, chaya, and mulberry leaf pellets can partially replace concentrate while maintaining growth performance and improving rumen fermentation efficiency in beef cattle.

1. Introduction

Sustainable livestock production is now a major global concern due to growing concerns about the effects on the environment, rising feed costs, and the need to improve rural farmers’ incomes. In tropical countries like Thailand, especially in Kalasin Province, the production of beef cattle is a major source of household income for smallholder farmers; yet, its productivity is limited by seasonal feed shortages and reliance on costly commercial concentrates [1,2]. It is essential to look into alternative feed options that are affordable, nutritionally balanced, and environmentally sustainable to improve cow productivity in smallholder settings.

Plant leaves from regionally accessible species such as mulberry (Morus alba), cassava (Manihot esculenta), and chaya (Cnidoscolus aconitifolius) are potential feed ingredients because of their high crude protein content, beneficial fiber qualities, and rich phytonutrient composition [3,4]. Direct feeding of fresh leaves, however, usually leads to erratic intake and storage issues. Pelletizing these leaves into compound feed offers a novel way to reduce reliance on costly concentrate feeds and imported soybean meal while also facilitating handling, conserving nutrients, and improving feed palatability [5]. In beef cattle production, feed cost and the reliance on commercial concentrate remain major constraints, particularly under volatile ingredient prices and increasing pressure to reduce environmental impacts. At the same time, tropical plant leaves such as cassava, chaya, and mulberry are widely available and nutritionally promising, yet they are often underutilized as practical feed resources due to limitations in handling, preservation, and consistent intake by animals. Converting these plant leaves into innovative feed pellets offers a feasible strategy to improve on-farm feed self-reliance by partially replacing concentrate while maintaining diet quality and feeding convenience.

This study is therefore necessary to evaluate plant-leaf-based pellets as a significant alternative for sustainable beef cattle feeding. The approach is expected to (i) lower production costs by reducing concentrate dependence, (ii) enhance nutrient utilization through improved digestibility and rumen function, and (iii) mitigate greenhouse gas emissions by shifting rumen fermentation toward more efficient pathways. By addressing both economic and environmental challenges simultaneously, plant-leaf pellet production represents a practical option aligned with sustainable livestock production goals.

2. Materials and Methods

All research was carried out in accordance with the principles and regulations approved by Kalasin University’s animal husbandry and use ethical committee (KSU-AE-051). Sources of cassava chips and rain tree pods were obtained from Kalasin Province.

2.1. Feed and Preparation of Experimental Feeds

Cassava, chaya, and mulberry leaves were oven-dried at 60 °C for 72 h, ground using a hammer mill, and passed through a 0.1 mm screen. The leaf meal was mixed with rice bran, crushed cassava, molasses, sulfur premix, lime, and urea (1% of dry matter). The urea level was included as a non-protein nitrogen source to enhance rumen microbial protein synthesis and to balance the fermentable carbohydrate provided by cassava-based ingredients. Pellets were produced using a pellet mill (capacity 100 kg/h) with a 0.1 mm die diameter, yielding pellets of 0.1 mm diameter. Pellets were cooled for 120 min, and the final moisture content was 5%.

2.2. Experimental Design, Animals, and Treatment

Four native Thai beef cattle of similar age (approximately 24–30 months), all male and in the growing–fattening stage, with no prior health disorders, were used in this experiment. The animals had an initial body weight (BW) averaging 250 ± 50 kg. The relatively wide range of body weight reflected normal variation among locally sourced native cattle; however, this variation was controlled statistically through the 4 × 4 Latin square design, in which each animal received all dietary treatments across experimental periods. Each animal received a concentrate diet (16% CP) and was housed individually in separate pens. The total concentrate allowance was provided at 1.5% of body weight (dry matter basis), which was composed of 50% commercial concentrate and 50% experimental pellet feed. Rice straw was offered ad libitum as a roughage source throughout the experiment, and its availability was kept constant across treatments. Animals also had free access to clean drinking water and mineral salt blocks. Total dry matter intake was recorded daily to determine actual nutrient consumption.

The samples were centrifuged at 16,000× g for 15 min using a tabletop centrifuge (PLC-02, Thermo Fisher Scientific, Waltham, MA, USA). A 4 × 4 Latin square design was employed, in which T1 served as the control diet containing concentrate only, while T2, T3, and T4 represented diets in which 50% of the concentrate was replaced with cassava leaf pellets, chaya leaf pellets, and mulberry leaf pellets, respectively. The chemical composition of the experimental diets is presented in

Table 1. Chemical composition of pellet feed, concentrate, and rice straw (% DM) used in the experiment.

Composition, Kilograms of Dry Matter	T2	T3	T4	Concentrate	Rice Straw
Soybean meal	10	10	0
Rice bran	30	30	30
Cassava leaves/chaya leaves/mulberry leaves	10	10	10
Leucaena leaf meal	18	18	18
Cassava mash	28	28	28
Molasses	2	2	2
Sulfur	1	1	1
Lime	0.5	0.5	0.5
Mineral premix *	0.5	0.5	0.5
Chemical composition
DM, %	94.53	95.55	96.63	95.23	96.71
% dry matter
Ash	7.47	6.85	7.57	11.12	10.11
OM	92.53	93.15	92.43	88.88	89.89
CP	16.09	16.73	17.37	16.03	2.77
AIA	1.23	1.35	1.44	1.55	5.45
NDF	49.79	46.54	48.26	45.52	70.32
ADF	39.11	35.97	36.59	21.62	47.22
Price, baht/kilogram	6.93	6.93	6.93	14.33	2.67
Price, USD/kilogram	0.21	0.21	0.21	0.44	0.08

CC = Concentrate; CPRFLK = Cassava pulp and rain tree pods fermented with loog-pang kaomark; RS = Rice straw; DM = Dry matter; OM = Organic matter; CP = Crude protein; AIA = Acid insoluble ash; NDF = Neutral detergent fiber; ADF = Acid detergent fiber. * Each kg contains the following: Vitamin A, 10,000,000 IU; Vitamin E, 70,000 IU; Vitamin D, 1,600,000 IU; Fe, 50 g; Zn, 40 g; Mn, 40 g; Co, 0.1 g; Cu, 10 g; Se, 0.1 g; I, 0.5 g. T2 = Pellet feed mixed with cassava leaves; T3 = Pellet feed mixed with chaya leaves; T4 = Pellet feed mixed with mulberry leaves.

. The experiment consisted of four periods, each lasting 21 days. During each period, the first 14 days were used for dietary adaptation, followed by a 7-day sample collection phase. Although the use of a 4 × 4 Latin square design is appropriate for metabolism and digestibility studies, the relatively small number of animals (n = 4) represents a limitation and may restrict the generalization of the findings beyond the experimental conditions.

2.3. Data Collection and Samples Analysis

Animals were measured for body weight gain at the beginning and end of each period. The amount of feed that the animals refused was subtracted from the amount given to them, and any remaining feed was thrown out before morning feeding. Over the last seven days of each period, adjustments were made to the concentration, rice straw, and pellet feed samples. Urine samples were collected as spot samples at a fixed time (morning, prior to feeding) and were adjusted for urinary volume using creatinine concentration to estimate daily purine derivative excretion. Fecal samples were obtained by means of rectal sampling twice daily during the collection period, composited per animal, thoroughly homogenized, and subsampled to ensure representativeness [6]. Daily excretion of purine derivatives was calculated according to established equations.

Following the guidelines of the Association of Official Analytical Chemists [7], the obtained samples were dried at 72 °C by means of continuous drying, ground on a 1 mm screen using a Cyclotech Mill, Tecator, Höganäs, Sweden, and determined for dry matter (DM; ID 967.03), ash (ID 492.05), ether extract (EE; ID 455.08), and crude protein (CP; ID 984.13). Van Soest et al. [8] state that amylase was used to analyze NDF and ADF but that sodium sulfite was not used. Instead, acid-insoluble ash (AIA) and residual ash were added to the expression of NDF and ADF [9].

On the last day of data collection, 10 mL of blood was taken from the jugular vein and poured into tubes with 12 mg of EDTA. This was performed at 0, 3, and 6 h after feeding. Thereafter, the plasma was separated by means of spinning it at 5000 g for 10 min (Table Top Centrifuge PLC-02, Thermo Fisher Scientific, Waltham, MA, USA). The obtained plasma was stored at 20 °C until it was examined for hemoglobin content (Hct) and blood urea nitrogen (BUN) using the procedures outlined in [10] and [11], respectively. Rumen fluid samples were collected at 0, 3, and 6 h after feeding on the final day of the data collection period. Rumen fluid was obtained using an orogastric tube technique, which was selected as a minimally invasive and practical method for repeated sampling in beef cattle. The oral stomach collection system consisted of a 90 cm polyvinyl chloride orogastric tube fitted with a 15 mL perforated plastic conical tube serving as a rumen sieve, connected to an electrical vacuum pump. To minimize saliva contamination, the first 50–100 mL of rumen fluid was discarded before sample collection. The tube was inserted to a standardized depth corresponding to the ventral sac of the rumen, and sampling was performed carefully to avoid excessive suction. All equipment was rinsed between animals to prevent cross-contamination. Each time, about 50 cc of rumen fluid was extracted from the rumen center using a stomach tube connected to a vacuum pump. A portable pH and temperature meter (HANNA HI-8424 Portable pH/ORP Meter, Woonsocket, RI, USA) was used to rapidly measure the temperature and pH of rumen fluid. Four thicknesses of cheesecloth were then used to filter the rumen fluid samples. A total of 45 mL of the rumen fluid sample was taken and placed in a plastic bottle with 5 mL of sulfuric acid solution (1 M) to halt the microbial activity fermentation process. The combination was centrifuged at 16,000× g for 15 min (Table Top Centrifuge PLC-02, Thermo Fisher Scientific, Waltham, MA, USA). Volatile fatty acids (VFAs) were separated using high-performance liquid chromatography (HPLC) following the method described by Samuel et al. [12]. Methane (CH₄) production was estimated according to the method of Moss et al. [13] using the equation CH₄ = 0.45 (acetate) − 0.275 (propionate) + 0.40 (butyrate). The ammonia nitrogen concentration (NH₃-N) in the supernatant was subsequently determined. For microbial analysis, nine milliliters of 10% formalin solution was mixed with one milliliter of ruminal fluid to preserve microbial cells. Microbial populations were enumerated using a hemocytometer (Boeco, Hamburg, Germany) under a light microscope. Prior to counting, samples were stained with appropriate differential stains (e.g., methylene blue for bacteria and lactophenol cotton blue for fungi) to facilitate identification. Microorganisms were identified based on morphological characteristics, including cell size, shape, and motility. All counts were performed in duplicate, and mean values were used for statistical analysis [14]. Prior to statistical analysis, the assumptions of parametric testing were evaluated. Data normality was assessed using the Shapiro–Wilk test, and homogeneity of variance among treatments was examined using Levene’s test. These preliminary tests confirmed that the data satisfied the assumptions required for parametric analysis.

Calculations

Digestible organic matter fermented in the rumen (DOMR) was calculated as follows: DOMR (kg/d) = digestible organic matter intake (DOMI, kg/d) × 0.65, where DOMI = [digestibility of organic matter (kg/kg DM) × organic matter intake (kg/d)]/100, 1 kg DOMI = 15.9 MJ ME/kg [15,16].

For determining microbial population, protozoa and fungi were enumerated under a microscope and a hemocytometer [17].

Microbial purine concentration and the efficiency of microbial N synthesis were measured using purine derivative excretions according to the equation developed by Chen and Gomez [18]

Y = 0.12X + (0.20 BW^0.75)

Microbial N synthesis was estimated based on urinary excretion of purine derivatives (PD) according to the equation of [18]:

MN (g/d) = 70X/(0.116 × 0.83 × 1000) = 0.727X

where X and Y are, respectively, absorption and excretion of PD in mmol/d. Efficiency of microbial N synthesis (EMNS) was calculated using the following formula:

EMNS = microbial N (g/d)/DOMR

where DOMR = digestible OM apparently fermented in the rumen.

2.4. Statistical Analyses

A 4 × 4 Latin square design was employed to evaluate treatment effects using SAS software (Version 9.4; SAS Institute Inc., Cary, NC, USA) [19]. Data were analyzed using the General Linear Model (PROC GLM). Prior to model fitting, residuals were assessed for normality using the Shapiro–Wilk test and for homogeneity of variance through Levene’s test to ensure compliance with ANOVA assumptions.

The statistical model was specified as follows:

$${\mathit{Y}}_{\mathit{i}\mathit{j}\mathit{k}}=\mathit{\mu }+{\mathit{M}}_{\mathit{i}}+{\mathit{A}}_{\mathit{j}}+{\mathit{P}}_{\mathit{k}}+{\mathit{\epsilon }}_{\mathit{i}\mathit{j}\mathit{k}}$$

where ${\mathit{Y}}_{\mathit{i}\mathit{j}\mathit{k}}$ represents the observed response from animal $\mathit{j}$ receiving treatment $\mathit{i}$ during period $\mathit{k}$; $\mathit{\mu }$ is the overall mean; ${\mathit{M}}_{\mathit{i}}$ is the fixed effect of dietary treatment (i = 1–4); ${\mathit{A}}_{\mathit{j}}$ is the random effect of animal (j = 1–4), accounting for individual variability; ${\mathit{P}}_{\mathit{k}}$ is the fixed effect of period (k = 1–4); and ${\mathit{\epsilon }}_{\mathit{i}\mathit{j}\mathit{k}}$ is the residual error term, assumed to be independently and normally distributed with constant variance.

Analysis of variance (ANOVA) was conducted to determine treatment effects. When a significant main effect of treatment was detected, multiple comparisons among treatment means were performed using Duncan’s novel multiple range test [20]. This procedure was selected to provide pairwise separation of means while maintaining control over experiment-wise Type I error under balanced design conditions. Statistical significance was declared at p < 0.05.

A double 4 × 4 Latin square (n = 8) was not adopted because the study was constrained by animal availability, housing capacity, and the labor- and cost-intensive nature of repeated measurements and laboratory analyses across periods. While a single 4 × 4 Latin square (n = 4) effectively accounts for between-animal heterogeneity and period-related variation, the limited sample size reduces the precision of estimated treatment effects and statistical power, thereby increasing the risk of Type II error for outcomes with modest effect sizes. Accordingly, inference should be restricted to the experimental conditions evaluated, and external validity (generalizability) to broader populations and production environments may be limited. Replication with an increased number of experimental units (e.g., a double Latin square or additional squares) is warranted to improve the robustness of effect estimates and strengthen population-level inference.

3. Results and Discussions

3.1. Growth Performance and Feed Intake

The use of leaf-based pelleted diets in place of concentrate affected the growth performance of beef cattle. As shown in

Table 2. Effects of leaf-based pelleted diets on production efficiency and feed intake in beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
BWG, kg/d	0.50 b	0.79 a	0.57 b	0.54 b	0.03	0.002
Total DM intake
Kg/d	2.92	3.28	3.23	3.29	0.12	0.22
% BW	1.16 b	1.39 a	1.29 ab	1.30 ab	0.04	0.04
g kg−1BW0.75	46.33 b	54.32 a	51.12 ab	51.78 ab	1.59	0.05
Rice straw intake
Kg/d	5.56	5.90	5.73	5.90	0.19	0.58
% BW	2.22	2.49	2.28	2.32	0.08	0.19
g kg−1BW0.75	88.27	97.46	90.79	92.70	2.78	0.22
Concentrate intake
Kg/d	8.48	9.19	8.96	9.19	0.29	0.34
% BW	3.38 b	3.87 a	3.57 ab	3.62 ab	0.10	0.06
g kg−1BW0.75	134.60 b	151.78 a	141.91 ab	144.48 ab	3.76	0.08

^a,b Values on the same row with different superscripts differ (p < 0.05). BWG = body weight gain; SEM = standard error of mean; T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate.

, cattle in the cassava leaf pellet group (T2) exhibited the highest body weight gain (BWG) at 0.79 kg/d, which was significantly greater (p < 0.05) than that of the control diet (T1; 0.50 kg/d) and the other leaf-based diets, including mulberry pellets (T4; 0.54 kg/d) and chaya pellets (T3; 0.57 kg/d). Total dry matter intake (DMI), expressed in absolute amounts (kg/d), did not differ significantly among treatments (p > 0.05). However, when expressed as a proportion of body weight, animals in the cassava pellet group consumed more feed (1.39% BW) than those in the control group (1.16% BW). A similar tendency was observed when intake was expressed relative to metabolic body weight (g/kg BW^0.75), where T2 showed the highest values.

The higher BWG observed in the cassava pellet group may be related to the higher crude protein content of cassava leaves (16.73% DM) compared with the control diet (14.03% DM). Increased intake relative to body weight may also indicate improved nutrient utilization and rumen fermentation associated with the inclusion of cassava leaves. Similar findings have been reported in studies indicating that cassava leaves can enhance feed intake and rumen function by providing fermentable nitrogen and bypass protein [1,21].

The improvement in nutrient digestibility observed in the leaf-pellet treatments provides a plausible explanation for the favorable production responses observed in this study. Animals receiving cassava, chaya, and mulberry leaf pellets (T2–T4) exhibited significantly higher apparent digestibility of dry matter and organic matter compared with the control treatment. These improvements indicate a greater availability of nutrients for absorption and metabolic utilization. In addition, the digestibility of structural carbohydrates, particularly neutral detergent fiber (NDF) and acid detergent fiber (ADF), was significantly enhanced in the leaf-pellet diets. Improved fiber degradation is particularly important in ruminant nutrition because it increases the availability of fermentable substrates for rumen microorganisms, thereby supporting more efficient energy utilization by the host animal. This is why they are good for tropical cattle systems. The findings corroborate the notion that utilizing pelleted cassava, chaya, and mulberry leaf resources can enhance resource sustainability and diminish the necessity for costly concentrates in tropical regions [22].

Although crude protein digestibility was numerically higher in the leaf-pellet treatments, the differences were not statistically significant. This result suggests that the replacement of concentrate with tropical leaf pellets did not negatively affect protein utilization in the rumen–intestinal system. Maintaining similar protein digestibility while improving overall dry matter and fiber digestibility indicates that the pelleted leaf resources can function as a nutritionally balanced feed component rather than simply a roughage substitute.

The improved digestibility of nutrients may partly explain the favorable intake and growth responses observed in animals receiving cassava leaf pellets. Higher digestibility generally enhances the efficiency of feed utilization by increasing the proportion of nutrients that become available for microbial fermentation and post-ruminal absorption. Consequently, the inclusion of tropical leaf pellets may contribute to improved feed conversion efficiency and support animal productivity under tropical production conditions. However, it should be noted that the present findings are primarily based on digestibility measurements; therefore, further studies incorporating long-term growth performance, rumen fermentation characteristics, and metabolic indicators would help clarify the mechanisms underlying these responses.

Patterns of concentrate intake also provide additional context for these findings. Animals in T2 consumed a slightly greater proportion of concentrate relative to body weight compared with those in the control treatment, although total concentrate intake did not differ significantly. This pattern may indicate that cassava leaf pellets interacted with the remaining concentrate portion in the diet to support efficient nutrient utilization and maintain voluntary feed intake. Taken together, these results suggest that partial replacement of concentrate with cassava leaf pellets can be a practical feeding strategy for beef cattle in tropical regions, potentially reducing reliance on commercial concentrates while maintaining efficient nutrient utilization and acceptable production performance.

From a physiological standpoint, the improved body weight gain (BWG) and enhanced voluntary feed intake observed in cattle fed cassava leaf pellets (T2) suggest that cassava leaves provided a balanced supply of fermentable nitrogen and readily digestible carbohydrates. The synergistic balance between energy and nitrogen likely optimized microbial growth and rumen fermentation, facilitating improved fiber degradation and energy capture [23]. These findings are consistent with Polyorach et al. [24], who reported that cassava foliage enhances ruminal microbial protein synthesis and improves nitrogen utilization efficiency in beef cattle. Similarly, Ampapon et al. [25] indicated that tropical foliage containing moderate tannin levels can stabilize rumen fermentation by reducing proteolysis and enhancing microbial growth, ultimately leading to greater feed efficiency.

3.2. Blood Metabolites and Hematological Indices

Blood biochemical measurements provide informative details regarding the animal’s general health and protein metabolism. The study found no significant difference in mean blood urea nitrogen (BUN) values between treatments (10.68–11.13 mg/dL; p > 0.05). The BUN values’ consistency across treatments indicated that dietary protein was utilized efficiently and that none of the leaf-based diets resulted in an increased nitrogen load that would have compromised renal function. The outcomes are consistent with previous research showing that supplementing with cassava and mulberry leaves does not adversely affect nitrogen metabolism and falls within the physiological range for beef cattle [26].

Additionally, hematocrit (HCT) values, which ranged from 29.33 to 31.42% (p > 0.05), stayed constant between treatments. These results demonstrate that leaf-based diets did not impair blood volume or oxygen-carrying capacity since they are within the typical physiological ranges for beef cattle. At the measured inclusion levels, the stability of both BUN and HCT highlights the safety of incorporating tropical leaves into cow meals and shows no negative hematological consequences.

The absence of notable variations in blood metabolites (BUN, HCT) among treatments reinforces the nutritional safety of the leaf-based diets. The maintenance of BUN within the physiological range (10–12 mg/dL) demonstrates effective nitrogen metabolism and suggests well-balanced dietary protein degradation and microbial capture [27]. A stable hematocrit (HCT) value also suggests that nutrient absorption and overall physiological function were unaffected by the inclusion of leaf materials, confirming that none of the experimental diets induced stress or metabolic imbalance. These results are comparable to the findings of Gunun et al. [28], who observed no detrimental effects on blood chemistry or rumen fermentation when replacing up to 50% of concentrate with cassava or mulberry leaves in dairy cattle (

Table 3. Effects of leaf-based pelleted diets on BUN and HCT in beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
BUN, mg/dL
0 h-post feeding	10.05	10.45	10.08	10.35	0.35	0.81
3 h-post feeding	10.80	11.40	10.88	11.03	0.44	0.78
6 h-post feeding	11.30	11.53	11.08	11.15	0.43	0.88
Mean	10.72	11.13	10.68	10.84	0.39	0.85
HCT, %
0 h post feeding	28.50	29.38	30.88	31.25	2.87	0.89
3 h post feeding	30.00	32.75	31.75	29.88	1.45	0.48
6 h post feeding	29.50	32.13	28.88	28.50	1.59	0.43
Mean	29.33	31.42	30.50	29.88	1.85	0.87

SEM = standard error of mean; BUN = blood urea nitrogen; HCT = hematocrit; T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate.

).

3.3. Digestibility of Nutrients

The apparent nutrient digestibility of beef cattle fed diets in which 50% of the concentrate was replaced with pelleted leaf feed is presented in

Table 4. Effect of leaf-based pelleted diets on the digestibility of beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
Apparent digestibility, %
Dry matter	47.28 b	58.00 a	57.36 a	59.88 a	1.50	0.01
Organic matter	42.21 b	53.98 a	53.30 a	56.29 a	1.68	0.01
Crude protein	50.31	57.97	55.62	57.31	1.72	0.07
Neutral detergent fiber	36.82 b	50.19 a	49.50 a	57.94 a	3.56	0.03
Acid detergent fiber	27.99 b	44.10 a	40.79 a	43.66 a	3.21	0.04

^a,b Values on the same row with different superscripts differ (p < 0.05). SEM = standard error of mean; T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate.

. The results showed that dry matter (DM) digestibility ranged from 47.28% in the control group (T1) to 59.88% in the mulberry leaf pellet group (T4). Similarly, organic matter (OM) digestibility increased from 42.21% in T1 to 56.29% in T4. The digestibility of crude protein (CP) ranged from 50.31% to 57.97% across treatments and did not differ significantly among dietary groups (p > 0.05). In contrast, the digestibility of fiber fractions improved markedly in the leaf-pellet diets. Neutral detergent fiber (NDF) digestibility increased from 36.82% in the control treatment to 57.94% in the mulberry leaf pellet treatment, while acid detergent fiber (ADF) digestibility increased from 27.99% to values ranging between 40.79% and 44.10% in the leaf-pellet groups. These results indicate that replacing 50% of concentrate with tropical leaf-based pellets improved the digestibility of dry matter, organic matter, and fiber components without negatively affecting crude protein utilization in beef cattle.

Although the statistical differences were significant in the pelleted diet, a numerical improvement in DM, OM, and fiber digestibility was observed between animals fed pelleted diets containing cassava, chaya, or mulberry leaves. This suggests that the inclusion of such leaf materials could enhance ruminal fermentation efficiency and feed utilization. According to Wanapat et al. [29], feed materials rich in plant protein and secondary metabolites can improve microbial colonization and fiber degradation in the rumen, resulting in better digestive performance. Norrapoke and Pongjongmit [30] also reported that the pelleting process enhances nutrient uniformity and feed intake consistency, which in turn supports more stable fermentation and improved nutrient digestibility.

Moreover, the CP digestibility was not significantly different, and it could be linked to the higher crude protein concentration and favorable amino acid profile of mulberry leaves compared to other tropical forages. These characteristics promote microbial protein synthesis and rumen fermentation balance, leading to improved fiber breakdown [31]. Additionally, bioactive compounds such as tannins and flavonoids in mulberry and chaya leaves may modulate ruminal microbial populations, enhancing fermentation efficiency and reducing energy loss through methane production [13].

Overall, the findings are consistent with previous studies showing that partial substitution of concentrate with high-protein leaf pellets maintains or slightly improves nutrient digestibility without compromising animal performance [29,30]. These results indicate that using pelleted leaf diets, particularly those containing mulberry leaves as a 50% concentrate replacement, is a viable strategy for improving feed cost efficiency and maintaining nutrient digestibility in beef cattle production systems.

The digestibility results further validate the potential of leaf-based pelleted diets to sustain nutrient utilization efficiency. Although statistical differences were significant, the numerical improvement in dry matter (DM), organic matter (OM), and fiber digestibility indicates that pelleting enhanced feed uniformity and improved nutrient accessibility for rumen microbes. Pelleting reduces particle size variation and heat-treats the material, which can denature anti-nutritional factors and increase nutrient availability [32]. Moreover, the fermentation-friendly compounds present in mulberry and chaya leaves, such as polyphenols and flavonoids, might have modulated microbial communities and reduced methanogenesis, leading to more efficient rumen fermentation [33]. These bioactive compounds can inhibit methanogenic archaea and redirect hydrogen toward propionate production, thereby improving energy efficiency and reducing energy loss through methane [34].

3.4. Rumen Fermentation Characteristics

The characteristics of rumen fermentation offer concrete proof of how diet affects microbial activity and energy metabolism. The temperature and pH of the rumen remained constant throughout treatments (6.73–6.86 and 39.12–39.83 °C, respectively), suggesting that the rumen environment was consistent across all diets. Conversely, higher ammonia-nitrogen (NH₃-N) concentrations were observed in the leaf-pellet treatments, particularly T4 (14.20 mg/dL), compared with the control group (9.14 mg/dL). This increase is consistent with the higher crude protein content and greater ruminal degradability of leaf-based diets, which provide more nitrogen substrates for microbial growth. However, while elevated NH₃-N can enhance microbial protein synthesis, the benefit occurs only when concentrations fall within the optimal range of approximately 8–20 mg/dL. Levels above the microbial utilization capacity may instead reflect inefficient nitrogen capture, leading to excess ruminal ammonia and increased urea formation. Therefore, although the higher NH₃-N in T4 suggests improved protein degradability, it also highlights the need to balance degradable protein supply with available fermentable energy to ensure efficient nitrogen utilization and prevent unnecessary N losses. The addition of leaf pellets significantly raised the quantities of total volatile fatty acids (VFAs), which in T4 reached 110.41 mmol/L rather than 106.67 mmol/L in T1. While acetate acid production dropped (56.51 mol/100 mol in T4 vs. 64.46 in T1), VFA profiles changed in favor of increased propionate acid production (32.65 mol/100 mol in T4 vs. 25.81 in T1). The estimated methane production index, calculated from VFA molar proportions using the equation proposed by Moss et al., decreased from 25.71 to 20.69 mol/100 mol TVFA in the mulberry leaf treatment (T4) compared with the control group. This reduction corresponds with the observed shift in rumen fermentation patterns. However, it is important to note that this value represents a predicted methane production potential derived indirectly from VFA profiles rather than a direct measurement of methane emissions. Therefore, the results should be interpreted cautiously as an indication of possible changes in rumen methanogenesis under the experimental conditions. Direct measurements using methods such as respiration chambers or the SF₆ tracer technique would be required to confirm actual methane emission responses. By rerouting fermentation pathways toward propionate, our findings imply that tropical leaf pellets can function as natural methane mitigators, increasing energy efficiency and lowering greenhouse gas emissions [13,35].

From an environmental perspective, the shift in volatile fatty acid (VFA) profiles toward a slightly higher proportion of propionate and a lower acetate-to-propionate ratio is noteworthy, even though the overall increase in total VFA concentration was modest (+3.7 mmol/L). Such changes, while not dramatic, may still hold biological relevance because propionate serves as a key gluconeogenic precursor in ruminants and contributes to more efficient energy utilization. A reduced acetate-to-propionate ratio may also indicate a shift in hydrogen utilization toward propionate formation rather than methane production. Based on the predictive equation proposed by Moss et al., this shift in fermentation pattern may be associated with a lower estimated methane production under certain conditions. Although the magnitude of methane reduction predicted for the mulberry treatment (T4) was moderate, it should be interpreted as a model-based estimation rather than direct evidence of methane mitigation, since methane emissions were not measured directly in this study [13]. Nevertheless, the results may suggest that leaf-based diets containing bioactive compounds such as tannins could influence rumen fermentation pathways in a way that potentially affects methane formation. These findings are consistent with Leng et al. [36], who reported that tropical tree foliage containing tannins can influence rumen fermentation characteristics and may moderately reduce methane production potential without adversely affecting digestibility. Overall, the results indicate a possible environmental benefit, although further studies involving direct methane measurements are required to confirm this effect.

3.5. Rumen Microbial Populations

Rumen fermentation’s properties provide verifiable evidence of how nutrition influences microbial activity and energy metabolism.

The results in

Table 5. Effects of leaf-based pelleted diets on ruminal microbial populations in beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
Total direct count (cell/mL)
Protozoa (×105)	4.79 a	4.67 ab	4.48 b	4.45 b	0.09	0.01
Anaerobic fungi (×107)	6.78	6.76	6.74	6.76	0.03	0.44
Bacteria (×1011)	5.32	5.32	5.32	5.38	0.05	0.47

^a,b Values on the same row with different superscripts differ (p < 0.05). SEM = standard error of mean; T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate.

show that replacing 50% of the concentrate with cassava, chaya, or mulberry leaf pellets affected rumen microbial populations differently. Protozoal counts were significantly higher in the control group (T1) than in the leaf-pellet treatments, with chaya (T3) and mulberry (T4) showing the lowest protozoal populations (p = 0.01). This pattern is consistent with the presence of plant bioactive constituents (e.g., tannins and/or saponins) that have been associated with reduced rumen protozoa. Because protozoa can prey on rumen bacteria, a lower protozoal population may be linked to improved microbial protein flow and nitrogen utilization; however, this interpretation should be treated cautiously without direct measures of microbial protein synthesis.

In contrast, anaerobic fungal and total bacterial counts did not differ statistically among treatments (p > 0.05) and appeared nearly identical across diets. Given that these values were obtained by means of microscopy-based enumeration, the apparent uniformity may reflect reporting precision (e.g., rounding to an insufficient number of decimal places) and/or methodological limitations, including the inherent variability of microscopic counting, detection limits for low-abundance groups, and constraints in discriminating morphotypes rather than true biological equivalence. To strengthen confidence in these results, the manuscript should report counting precision explicitly (appropriate decimal places and units), describe technical replication (e.g., duplicate/triplicate counts per sample and averaging procedures), and provide additional methodological detail: sample preservation and fixation, dilution scheme, staining method (if applied), counting chamber type (e.g., hemocytometer/Sedgewick–Rafter), magnification, number of fields counted and the rule for field selection, criteria used to identify protozoa/bacteria/fungal zoospores or sporangia, and procedures to reduce observer bias (e.g., blinded counting and/or inter-observer agreement). With these clarifications, the data would more robustly support the conclusion that leaf pellets primarily reduced protozoal abundance while total bacterial and fungal counts remained broadly unchanged within the sensitivity of the microscopy method under the conditions tested (

Table 6. Effect of pelleted plant leaf feed on the rumen fermentation process in beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
Ruminal, pH	6.73	6.84	6.85	6.86	0.06	0.52
Ruminal temperature, °C	39.25	39.55	39.83	39.12	0.29	0.39
NH3-N, mg/dL	9.14 c	11.63 b	12.52 b	14.20 a	0.48	<0.01
Total VFA, mmol/L	106.67 b	107.91 b	108.13 b	110.41 a	0.67	0.01
	VFA profiles, mol/100 mol
Acetic acid	64.46 a	59.50 b	59.53 b	56.51 b	1.15	0.01
Propionic acid	25.81 b	29.83 ab	29.08 ab	32.65 a	1.45	0.01
Butyric acid	9.73 b	10.68 ab	11.39 a	10.85 ab	0.45	0.04
CH4 production A, mol/100 mol TVFA	25.71 a	22.75 ab	23.25 ab	20.69 b	1.03	0.01

^a,b,c Values on the same row with different superscripts differ (p < 0.05). SEM = standard error of mean, T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate. ^A Calculated according to the method of Moss et al. (2000) [13]: CH₄ production = 0.45 (acetate) − 0.275 (propionate) + 0.4 (butyrate).

).

3.6. Microbial Protein Synthesis

Leaf supplementation greatly increased microbial protein synthesis, a crucial measure of rumen microbial efficiency. Microbial nitrogen supply (MNS) increased from 55.44 gN/d in T1 to 67.18 gN/d in T4, and absorbed purine derivatives (PD) increased from 76.29 mmol/d in T1 to 92.44 mmol/d in T4 (p < 0.01). Likewise, the mulberry group’s microbial protein synthesis (EMPS) efficiency increased from 21.67 gN/kg OMDR in the control group to 30.70 gN/kg OMDR (

Table 7. Effects of leaf pelleted diets on microbial protein synthesis in beef cattle.

Parameters	T1	T2	T3	T4	SEM	p-Value
Microbial protein synthesis
PD excreted, mmol/d	23.69	25.06	25.51	25.64	0.75	0.10
PD absorbed, mmol/d	76.29 b	87.17 a	90.52 a	92.44 a	2.73	0.01
MNS, gN/d	55.44 b	63.35 ab	65.78 a	67.18 a	3.05	0.02
EMPS, gN/kg OMDR	21.67 b	28.61ab	27.60 ab	30.70 a	2.18	0.03

^a,b Values on the same row with different superscripts differ (p < 0.05). SEM = standard error of mean; PD = purine derivative; MNS = microbial nitrogen supply; EMPS = efficiency of microbial protein synthesis; OMDR = organic matter digestible in the rumen; T1 = concentrate (control group); T2 = cassava leaf pellets replace 50 percent of concentrate; T3 = chaya leaf pellets replace 50 percent of concentrate; T4 = mulberry leaf pellets replace 50 percent of concentrate.

).

Microbial protein synthesis (MPS) is one of the most critical indicators of rumen fermentation efficiency. The substantial increases in microbial nitrogen supply (MNS), absorbed purine derivatives (PD), and efficiency of microbial protein synthesis (EMPS) in the leaf-supplemented treatments confirm that the inclusion of tropical leaf pellets provides sufficient nitrogen sources for microbial growth. The enhancement of EMPS from 21.67 to 30.70 gN/kg OMDR in the mulberry group demonstrates improved synchronization between nitrogen and energy supply, which is essential for microbial protein yield [37]. Leng et al. [36] reported similar improvements, noting that tropical forages with balanced amino acid composition enhance microbial efficiency and animal performance. Increased MPS also translates into better protein availability in the small intestine, improving the amino acid supply for growth and production.

These enhancements demonstrate how providing leaf-based diets to rumen bacteria increases their ability to synthesize protein. Since microbial protein is the primary source of amino acids absorbed in the small intestine, such findings are crucial for the development of cattle. In addition to enhancing growth, a higher microbial protein source also lessens the requirement for pricey protein supplements like soybean meal. The study’s encouraging findings are consistent with observations that supplementing ruminants with tropical foliage promotes nitrogen retention and microbial protein synthesis [1,2].

Economically, the partial substitution of concentrate with cassava, chaya, or mulberry leaves may offer potential opportunities to reduce feeding costs in tropical beef production systems. These leaves are widely available agricultural resources in many tropical regions, including Thailand, and have been reported as alternative protein sources in ruminant diets. Their use may reduce reliance on conventional protein ingredients such as soybean meal and promote more efficient utilization of locally available biomass. In this context, the incorporation of plant-based feed resources could support broader sustainable livestock production strategies, which align with the principles of the bio-circular-green (BCG) economy that emphasize resource efficiency and circular agricultural practices [38]. However, the economic and sustainability implications should be interpreted as potential benefits rather than direct outcomes of the present study.

Mulberry leaf pellets may promote superior digestibility and microbial fermentation because their favorable amino acid composition and higher soluble protein fraction provide readily fermentable nitrogen sources for rumen microbes. In the rumen, amino acids can be deaminated and fermented by rumen bacteria, supplying ammonia, volatile fatty acids, and carbon skeletons that support microbial growth and fiber degradation. This biochemical pathway helps stabilize microbial colonization and enhances fermentation efficiency. Furthermore, from a sustainability perspective, integrating tropical leaf-based feed resources into ruminant production contributes to environmental stewardship and circular agriculture. Using locally available by-products such as cassava, mulberry, and chaya leaves reduces feed cost and minimizes feed–food competition between livestock and humans [39]. These leaves are typically residual materials from household gardens or agricultural pruning; thus, repurposing them into high-protein feed reduces waste accumulation and supports zero-waste farming systems. In addition, leaf pellet processing adds value to underutilized biomass and enhances feed shelf life, facilitating storage and transport efficiency under tropical conditions [40].

From a nutritional ecology viewpoint, the moderate concentration of secondary metabolites such as condensed tannins in tropical foliage plays a dual role. At appropriate inclusion levels, tannins can form protein–tannin complexes, slowing ruminal protein degradation and enhancing post-ruminal amino acid absorption [41]. This mechanism not only improves nitrogen retention but also mitigates ruminal ammonia accumulation, explaining the balanced BUN levels across treatments in the present study. However, excessive tannin intake could lead to reduced palatability or mineral bioavailability, as reported by Makkar [42]; therefore, optimal formulation remains critical for maximizing animal performance.

In alignment with global low-carbon livestock initiatives, feeding strategies that improve feed conversion efficiency and lower methane emissions, such as those demonstrated herein, offer a practical route to achieve climate-smart agriculture [43]. Incorporating tropical leaf-based pellets can therefore support national targets for reducing greenhouse gas emissions while sustaining animal productivity. In general, the results show that finding the right balance between nutrient availability, microbial efficiency, and environmental outcomes is important for both economic and ecological sustainability in tropical ruminant production systems.

In conclusion, the present study provides compelling evidence that pelleted tropical leaf diets, especially those based on cassava and mulberry leaves, can replace up to 50% of concentrate in beef cattle diets without compromising performance or health. The inclusion of these leaf materials improved growth performance, maintained nutrient digestibility, enhanced microbial protein synthesis, and reduced methane emission potential. These outcomes are consistent with sustainable livestock production goals aimed at reducing feed costs, optimizing nutrient utilization, and minimizing environmental impacts. Therefore, leaf-based pelleted feeds represent a feasible and ecologically sound strategy for intensifying tropical beef production systems in the era of climate-conscious agriculture [44].

4. Conclusions

This study investigated the potential of mulberry, cassava, and chaya leaf pellets as partial replacements for concentrate in diets of native Thai beef cattle. Under the experimental conditions of a 4 × 4 Latin square design involving four animals and 21 days per period, replacing 50% of the concentrate with tropical leaf pellets maintained feed intake, growth performance, and key rumen fermentation characteristics without negative effects on the measured animal health indicators. These findings indicate that tropical leaf pellets derived from locally available plant resources may serve as alternative feed ingredients in beef cattle diets under controlled experimental conditions. However, the results should be interpreted within the limitations of the present study, including the small sample size and relatively short experimental duration.

Further research involving larger animal populations, longer feeding periods, and additional measurements such as carcass characteristics, direct methane emission assessments, and economic evaluations is required to better determine the broader applicability and practical feasibility of using tropical leaf pellets in beef production systems.