Relationship Between Growth Performance and Metabolite Profile of Broiler Chickens Supplemented with Probiotics Bacillus coagulans and Lactobacillus Pradhika EI1.
Astuti RI2.
Meryandini A2 Biotechnology Study Programme.
Graduate School.
IPB University.
Jl.
Agatis.
Darmaga.
Bogor.
West Java.
Indonesia 16680 Biology Department.
Faculty of Mathematics and Natural Sciences.
IPB University.
Jl.
Agatis.
Darmaga.
Bogor.
West Java.
Indonesia 16680 Email: ameryandini@apps.
eceived 24-09-2024.
revised 04-03-2025.
accepted 07-03-2.
ABSTRAK Pradhika EI.
Astuti RI.
Meryandini A.
Hubungan performa tubuh dan profil metabolit pada ayam broiler dengan suplementasi probiotik Bacillus coagulans and Lactobacillus plantarum.
JITV 30.
:115-125.
DOI:http://dx.
14334/jitv.
Suplementasi probiotik merupakan alternatif dari Antibiotic Growth Promotor.
Probiotik L.
plantarum dan B.
diketahui dapat meningkatkan performa pertumbuhan ayam broiler.
Informasi mengenai hasil metabolit kedua probiotik dengan inangnya masih terbatas.
Penelitian ini bertujuan untuk mengidentifikasi metabolit pembeda antara Bacillus coagulans dan Lactobacillus plantarum dan metabolit yang berhubungan dengan peningkatan performa pertumbuhan ayam broiler dengan hasil suplementasi probiotik .
Sebanyak 120 ekor Ayam Broiler unsexRoss 308 diberikan pakan perlakuan dengan Kontrol Negatif (NC).
plantarum (LP).
coagulans (BC), dan B.
coagulans dicampur dengan L.
plantarum (BCLP).
Parameter kinerja pertumbuhan yang dievaluasi adalah rata-rata bobot badan .
verage Body Weight/ avg BW), konversi pakan .
djustment Feed Conversion Ratio/adjFCR), konsumsi pakan kumulatif.
umulative Feed Intake/cumFI) dan faktor efisiensi performa (Performance Efficiency Factor/PEF).
Analisis metabolik dilakukan dengan metode pemprofilan metabolit tidak tertarget pada sampel sekum yang terdiri dari analisis senyawa spektrum luas dan analisis senyawa volatil.
Hasil penelitian menunjukkan bahwa kinerja pertumbuhan ( avg BW) yang berbeda nyata .
O0,.
Sedangkan parameter lainnya yaitu, adjFCR, cumFI, dan PEF, tidak memberikan perbedaan yang signifikan (P>0,.
Metabolit pembeda yang penting antar perlakuan adalah asam asetat, asam laktat, asam butanoat, 1-oktadekanol, dan asam palmitat.
Metabolit yang dapat dinyatakan sebagai metabolit pembeda antara LP dan BC adalah asam asetat, asam laktat, dan asam butanoat.
Sedangkan metabolit yang dapat dinyatakan sebagai metabolit pembeda adalah asam laktat sebagai pembeda kinerja avgBW yang baik dan 1-oktadekanol dan asam palmitat sebagai pembeda tanpa suplementasi probiotik.
Kesimpulan dari studi ini adalah asam asetat, asam laktat dan asam butanoat adalah metabolit pembeda antara probiotik B.
coagulans dan L.
plantarum dan asam laktat sebagai metabolit pembeda kinerja yang baik.
Kata Kunci: Ayam Broiler.
Metabolit Pembeda.
Metabolit.
Performa.
Probiotik
ABSTRACT
Pradhika EI.
Astuti RI.
Meryandini A.
Relationship of growth relationship between growth performance and metabolite profile of broiler chickens supplemented with probiotics Bacillus coagulans and Lactobacillus plantarum.
JITV 30.
: 115-125.
DOI:http://dx.
org/10.
14334/jitv.
Probiotic supplementation is an alternative to Antibiotics Growth Promotor.
The probiotics L.
plantarum and B.
are known to improve the growth performance of broiler chickens.
Information regarding the metabolite results of these two probiotics with their hosts is still limited.
This study aims to identify differentiating metabolites between Bacillus coagulans and Lactobacillus plantarum and metabolites associated with enhanced growth performance in chickens supplemented with A total of 120 unsexed Ross 308 Broilers were given a treated diet with Negative Control (NC).
plantarum (LP), coagulans (BC), and B.
coagulans mixed with L.
plantarum (BCLP).
The growth performance parameter evaluated was the average Body Weight .
verage BW), adjustment Feed Conversion Ratio .
djFCR), cumulative Feed Intake .
umFI), and Performance Efficiency Factor (PEF).
Metabolomic analysis was carried out using the untargeted metabolite profiling method on cecum samples, consisting of broad-spectrum and volatile compound analyses.
The study shows that analysis of differences in growth performance resulted in only the avgBW parameter being significantly different (PO0.
Meanwhile, other performance parameters, adjFCR, cumFI, and PEF, do not provide any significant difference (P>0.
The important differentiating metabolites between treatments are acetic acid, lactic acid, butanoic acid, 1-octadecanol, and palmitic acid.
Metabolites that can be stated as differentiating metabolites between LP and BC are acetic acid, lactic acid, and butanoic acid.
Meanwhile, metabolites that can be declared differentiating metabolites are lactic acid as a differentiator for good avgBW performance and 1-octadecanol and palmitic acid as differentiators without probiotic supplementation.
It can be concluded that JITV Vol.
30 No.
2 Th.
2025:115-125
acetic acid, lactic acid, and butanoic acid are the metabolites that differentiate the probiotics B.
coagulans and L.
plantarum and lactic acid as a differentiating metabolite of good performance.
Key Words: Broiler Chicken.
Differentiating Metabolites.
Metabolite.
Performance.
Probiotics INTRODUCTION The chicken farming industry continues to develop to meet the increasing need for animal protein.
One important factor in achieving optimal growth and health of chickens is using feed enriched with AGP (Antibiotic Growth Promote.
The prohibition of AGP dramatically influences the productivity of broiler chickens in Indonesia as an implementation of Law (UU) Number 41 of 2014 concerning Amendments to Law Number 18 of 2009 concerning Livestock and Animal Health concerning the ban on using antibiotics and certain hormones as feed additives.
this encourages researchers to develop safe and effective AGP One solution to this problem, called AGP replacer, is probiotic supplementation.
Probiotics have been widely used in the feed industry today as AGP Some of the advantages of using probiotics in the digestive tract are stimulating beneficial microbes, preventing pathogen colonization by competition for attachment to the epithelium, reducing pH conditions, producing organic acids, forming compounds with systemic effects, and modulating the immune system (Abd El-Hack et al.
Tran et al.
One of the LAB (Lactic Acid Bacteri.
based probiotics is the L.
plantarum strain.
Feed supplemented with L.
plantarum strain B1 was shown to reduce the number of E.
coli, increase other LAB bacteria, produce SCFA (Short Chain Fatty Aci.
, and improve broiler performance (Peng et al.
One type of SFB (Spore Former Bacteri.
used as probiotics is B.
coagulans (Gu et al.
coagulans strain R11 was shown to prevent oxidative damage and reduce the abundance of pathogens such as E.
aeruginosa, and Salmonella (Xing et al.
One approach to studying the efficacy of probiotics on the health of their hosts is through the metabolomics Metabolite profiling of untargeted compounds allows for identifying compounds that undergo significant concentration changes under different treatment conditions (Frainay and Jourdan Liu et al.
explained that the metabolite results from probiotics could generally be lactic acid, hydrogen peroxide, secreted proteins .
xtracellular protein.
, organic acids, indole, bacteriocins, and antimicrobial peptides.
Wang et al.
provide an overview of the characteristics of LAB metabolites as probiotic functions, including being able to produce short-chain fatty acids, amines, bacteriocins, vitamins and exopolysaccharides.
According to Elshaghabee et .
Bacillus spp.
stimulate the immune system and produce several antimicrobial substances, e.
bacteriocins like inhibitory substances and antibiotics.
Probiotics based on L.
plantarum and B.
improve broiler growth performance (Khajeh Bami et Peng et al.
However, this characteristic is unclear in identifying metabolite compounds that distinguish the two types of probiotics.
The metabolomic analysis provides an overview of the diversity of metabolite compounds from probiotics.
Therefore, it is necessary to know the profile of metabolite diversity between the two probiotics, which is associated with growth performance.
This study aimed to identify the differences in metabolite profiles between B.
coagulans and L.
plantarum and to determine the metabolites that distinguish good growth
MATERIALS AND METHODS
This research was conducted at the research farm (AME House/Digestibility Assay Unit.
House 7.
Room A & B, closed hous.
of PT Nugen Bioscience Indonesia.
Malingping.
Banten and applied chemistry department laboratory of PT Nugen Bioscience Indonesia.
Ancol.
North Jakarta.
The Animal Ethics Committee School of Veterinary Medicine and Biomedical Science IPB University approved the 070/KEH/SKE/VII/2023.
Treatment, experimental One hundred and twenty DOC Broiler .
A0.
Ross 308 grade A3 unsex (PT Charoen Pokphand Jaya Farm.
Tangerang.
Indonesi.
were randomly divided into four dietary treatments: LP (L.
BC (B.
BCLP (B.
coagulans and L.
and NC .
o probiotic.
with 6 replicates per treatment and 5 bird per replicates.
Twenty-four cages were arranged on racks randomly divided into two rooms .
ooms A and B).
Each room contained 12 cages with three replications of each treatment.
Each cage .
39 m, 0.
145 m.
consists of 2 nipple drinkers/cage, 1 bell drinker/cage, 1 feeder/cage, and 1 lamp/cage.
Each room .
5 m, 100 m .
contains 4 cage racks/room, 1 fan unit/room, and 1 water tank unit/room.
Cages were cleaned with disinfectant .
eracetic acid-hydrogen peroxide-acetic acid (Cid 2000.
PT SHS International.
Jakarta.
Indonesi.
Pradhika et al.
Relationship of growth relationship between growth performance and metabolite profile of broiler chickens supplemented Rooms were fumigated .
g peroxyacetic acid (Forcent Fumigant.
PT Indovetraco Makmur Abadi.
Jakarta.
Indonesi.
added to 150 ml formaldehyde 37 % (Formac.
PT Indovetraco Makmur Abadi.
Jakarta.
Indonesi.
for an area of 5 m.
before use to prevent DOCs were weighed before being put into the cages.
Feeding was done using an ad libitum feeding system according to the treatment.
In the starter/brooding phase .
-10 day.
DOC was fed with S10 feed, and in the grower and finisher phases .
-30 day.
, they were fed with S11 feed, which has been supplemented with probiotic products according to the The nutritional content of the feed can be seen in Table 1.
Room temperature was maintained according to Ross 308 guidelines (Aviagen, 2.
by adding a heat source lamp in the brooding phase, adjusting the frequency of ventilation opening, and setting the fan switch.
Chicken performance was determined by measuring avgBW .
verage Body Weight at 10, 21, and 30 day.
, cumFI .
umulative Feed Intak.
, adjFCR .
djusted Feed Conversion Rati.
PEF (Performance Efficiency Facto.
, and mortality parameters calculated according to Ross 308 Aviagen .
and Martynez & Valdiviy .
standard guidelines which can be seen in the Table 2.
All chickens were then slaughtered, and cecum content samples were aseptically removed from the chickens using scissors and tweezers and placed into labeled sterile tubes.
Samples were frozen with dry ice in an ice box and then stored in an ultra-low temperature freezer (Kaltis 390.
Taipei.
Taiwa.
at -80AC, according to Zhou et al.
Feed preparation Probiotic products consist of L.
plantarum N1A1 or coagulans BR04 mixed in a carrier .
orn starch and CaCO.
(PT Nugen Bioscience Indonesia.
Jakarta.
Indonesi.
with concentration >106 CFU/g.
Starter (S.
and grower-finisher (S.
phase feeds (PT Charoen Pokphand Indonesia.
Balaraja.
Indonesi.
were each mixed with 1.
5 % probiotic product to produce a minimum concentration of 108 CFU/kg feed as recommended by Wang et al.
Feed is mixed with a mini-feed mixer for 15 minutes at room The NC treatment was supplemented with products without probiotics .
nly carrier.
Metabolomic analysis The extraction and derivatization step for metabolomic analysis of untargeted broad-spectrum compounds was adopted from Fiehn .
of standard mix QC.
Acetonitrile:Isopropanol: Water (AIW) solution (Merck.
Darmstadt.
German.
with a ratio of 3:3:2 was purged with N2 gas from gas generator (Proton N341M.
Proton OnSite.
USA) for 5 min and then cooled at -20 AC.
25-30 mg of cecum content sample was weighed with analytical balance (Precisa XB 220A.
Dietikon.
Switzerlan.
into a 2 ml microtube, and 1 ml of AIW was added.
The microtube was mixed with a vortex (Heidolph REAX control.
Schwabac.
for 10 s followed by one h sonication (Elmasonic P300H.
Singen.
German.
at 35AC and then centrifuged (Biofuge Fresco Sorvall.
Thermo Fisher Scientific.
Waltham.
USA) at 13,000 yg for 2 min.
AAl of supernatant was separated into a new microtube and concentrated with a vacuum concentrator (Concentrator 5301.
Eppendorf.
Hamburg.
German.
for 2 hours at 45AC.
MeOX solution was prepared by mixing 20 mg methoxyamine HCl (Sigma Aldrich.
Massachusetts.
USA) and 1 ml pyridine (Merck.
Darmstadt.
German.
, then sonicated for 15 min at 60 AC.
The concentrated microtube was added with 50 AAl MeOX, followed by 1.
5 hours of sonication at 30AC.
Then 100 AAl of MSTFA (N-methyl-N-.
Merck.
Darmstadt.
German.
was added and sonicated for one h at 37AC and then centrifuged at 13,000 yg for 10 min at 18AC.
The supernatant obtained was then transferred to a vial insert and placed into a GC vial.
The samples in the vials were then randomly arranged and analyzed using a gas chromatography system with specifications according to Jain et al.
GC system:
Agilent 7000C Triple Quadrupole GC/MS System (Agilent.
Santa Clara.
USA).
column: HP-5MS Ultra Table 1.
Nutritional content of S10 and S11 feed Parameter S10 feed S11 feed Moisture (%) Fat (%) Fiber (%) Protein (%) Ash (%) ME .
kal/k.
ME= Metabolism Energy JITV Vol.
30 No.
2 Th.
2025:115-125
Table 2.
Calculation formula for performance parameter Parameter Formula avgBW .
/bir.
total bird weight/number of birds.
cumFI .
/bir.
Average daily Feed Intake .
vgDFI) y number of birds y number of days adjFCR actFCR .
arget body weight Ae actual body weight / 4500 .
actFCR
total feed consumed / total bird weight
PEF
livability y bird weight .
/ age .
y FCR livability (%) The final number of birds/initial number of birds y 100 mortality (%) total death or culling/number of birds y 100 Inert .
25 mmy0.
25 AA.
gas: He .
25 ml/mi.
injection volume: 1 AAl.
delay: 4 min.
inlet: splitless, 250AC, 14.
7 psi.
oven: 75AC, 280AC .
AC/min, 1.
detector: MS, source: 230AC, 40-600 m/z, scan time: 0.
2 s.
Extraction and derivatization methods in metabolomics analysis for volatile targeted compounds were adopted from Hsu et al.
Partially frozen samples stored at -80AC were freeze-dried with a freeze dryer for 24 hours.
Samples that were not analyzed immediately could be stored again at -80AC.
A dry sample of 0.
02 g was weighed in a microtube, and 1 ml 5% phosphoric acid (Merck.
Darmstadt.
German.
was added.
The sample was vortexed for 30 s and then centrifuged at 3,000yg for 10 min.
60 AAl of supernatant was removed, and 240 AAl of 0.
5% phosphoric acid and 300 AAl of butanol (Merck.
Darmstadt.
German.
were The sample was vortexed for 30 s, shaken for 5 min, followed by sonication for 5 min.
Then, the microtube was centrifuged at 3,000 yg for 10 min.
AAl of supernatant .
rganic laye.
was transferred to a vial insert, and 20 AAl of butanol was added.
The samples in the vials were then randomly arranged and analyzed using a gas chromatography system according to the following specifications.
GC system: Agilent 7000C Triple Quadrupole GC/MS System (Agilent.
Santa Clara.
USA).
column: DB-WAXms .
25 AA.
gas: He .
25 ml/mi.
volume: 1 AAl.
delay: 4 min.
inlet: splitless, 250AC, 14.
oven: 70AC, 170AC .
AC/min, 0 mi.
, 240AC .
AC/min, 15 mi.
detector: MS, source: 230AC, 40-550 m/z, scan time: 0.
2 sec.
Solvent blank, reagent .
blank, and method blank were selected as quality control for each batch analysis (Fiehn 2016.
Broadhurst et al.
Eurachem 2.
Data analysis Raw growth performance data were processed, and the significance of performance was determined by statistical analysis on MinitabA 16.
1 (Minitab Ltd.
Pennsylvania.
USA).
The statistical analysis stages performed were .
outlier identification (NIQR boxplo.
, .
assumption checking .
ata normality test:
Shapiro-Wilk test, homogeneity or homoscedasticity of data: Bartlett tes.
, .
omnibus test .
arametric test:
ANOVA or non-parametric test: Kruskal-Walli.
, and .
post hoc test .
arametric test: Tukey test or nonparametric test: Dunn tes.
(Granato et al.
Nonparametric tests are performed for samples <15 data.
Chromatogram data from the metabolomic analysis was processed using Masshunter Qualitative Analysis 00 software (Agilent.
Santa Clara.
USA).
Chromatogram peaks with a minimum height of 10 5 mAU .
ili Absorbance Uni.
were identified from the TIC (Total Ion Chromatogra.
, and then the deconvolution process was performed.
The detected peaks were then matched to the National Institute of Standards and Technology (NIST) database with a similarity score of at least 80 %.
Raw data in peak intensity height.
RT (Retention Tim.
, and compound name were processed in MS Excel by adopting the procedure from Fiehn .
Curation data from the analysis of broad-spectrum and volatile compounds were combined into one, then outlier identification, compound name filtering and box-plot generation using MS Excel.
Multivariate analysis and compound categorization were performed with MetaboAnalyst 5.
(Wishart Research Group.
Alberta.
Canad.
The compounds obtained were grouped by class using the 'Enrichment Analysis' feature.
All compounds identified by NIST from the two metabolomic analyses were confirmed by .
he Human Metabolome Databas.
HMDB library-based matching.
Compounds identified but not matched and indicated not to be metabolites were excluded from further analysis.
The proportion .
of data was determined by calculating the number of metabolites that appeared .
per number of samples .
Identified metabolites that have a proportion >0.
are then processed using the 'Statistical Analysis .
ne facto.
' feature on MetaboAnalyst 5.
0 with the stages of .
data upload, .
data integrity checking, .
data filtering, .
data normalization .
statistical processing.
This statistical process is divided Pradhika et al.
Relationship of growth relationship between growth performance and metabolite profile of broiler chickens supplemented into three: .
Principal Component Analysis (PCA) .
core plot and loading plo.
, .
PLS-DA (Partial Least-Squares Discriminant Analysi.
(Variable Importance in Projection (VIP) scor.
RESULTS AND DISCUSSION Growth performance The chicken used in this study used the Ross 308 This Ross 308 strain performs better than other strains in BW and FCR parameters (Martynez & Valdiviy 2.
The total number of samples was 30 for each treatment except BCLP, which had 29 samples due to one bird being affected by the Runting Stunting Syndrome (RSS).
RSS in broiler chickens is observed on 4-7 days with shorter shanks, lower body weight, pale, distention of the abdomen, poor feather development, listlessness, and diarrhea (Li et al.
Aviagen 2.
Assumption tests were conducted to ensure that the data followed a normal distribution pattern and that data variance was homogeneous (Kozak & Piepho 2018.
Orcan 2.
The RSD of the avgBW parameter ranges 08 to 11.
07 %.
Data uniformity is acceptable if the ARSD value is <10 % (Aviagen 2.
The performance profile of the chickens showed that only avgBW was significantly different after ANOVA and Tukey tests with the highest to lowest weights in order:
0-10 days (LPa.
BCLPab.
NCb.
BC.
, 0-21 days: (LPa.
BCLPb.
BCb.
NC.
, 0-30 days (LPa.
BCab.
BCLPab.
NC.
The data indicate that treatment with L.
plantarum yields the best avgBW performance across all rearing periods.
Table 3.
Growth performance for parameters avgBW, cumFI, adjFCR, mortalityPEF with RSD values and letter notations from post hoc tests for data significance Parameters avgBW .
/bir.
ARSD (%) (PO0.
cumFI .
/bir.
ARSD (%) (PO0.
adjFCR
ARSD (%)
(PO0.
mortality (%) PEF ARSD (%) .
O0.
Treatment 0-10 d 0-21 d 0-30 d 57bA10.
67bA10.
57bA11.
5aA9.
1057aA8.
43aA9.
5bA8.
33bA7.
87abA10.
BCLP
67abA10.
69bA9.
96abA7.
57A8.
77A10.
53A7.
30A6.
37A10.
00A7.
43A4.
17A4.
63A3.
BCLP
17A2.
43A6.
88A4.
05A7.
33A9.
38A6.
04A8.
25A8.
33A3.
06A1.
30A3.
35A2.
BCLP
05A4.
28A8.
36A7.
BCLP
62A9.
64A2.
88A4.
BCLP
57A6.
NC= negative control.
LP = Lactobacillus plantarum.
BC = Bacillus coagulans.
BCLP= Bacillus coagulans & Lactobacillus plantarum, avgBW= average body weight, cumFI= cumulative feed intake, adjFCR= adjusted feed conversion ratio.
PEF= performance efficiency Factor on 0-30 day observation.
RSD= relative standard deviation JITV Vol.
30 No.
2 Th.
2025:115-125
The non-parametric significance difference test (Kruskal-Walli.
was conducted on several other performance parameters .
umFI, adjFCR, and PEF).
These parameters stated that they were not significantly different, as indicated by a P>0.
However, when viewed from the average data, the LP treatment still has the best value compared to other treatments.
One bird in the BCLP treatment was excluded due to stunting, resulting in a mortality rate of 3.
FCR measures feed utilization efficiency or production efficiency.
the smaller the FCR value, the better or more efficient (Prakash et al.
, 2.
Meanwhile.
PEF is used to measure overall growth performance, which indicates that the higher the PEF value, the better the growth performance (Aviagen Ross 308 broilers at 28 and 35 days will have FCR .
5 and 1.
and PEF .
31 and 405.
respectively (PetriseviN et al.
The results of this study showed that all treatments had FCR and PEF values better than the performance in large-scale Comparisons can also be made with the parameter values of the Ross 308 growth standard (Aviagen 2.
Compared to this standard, avgBW values were higher in all treatments for 0-30 days.
While cumFI was higher than the standard in all treatments and all rearing days.
However, adjFCR had worse results for all treatments and rearing days.
Other studies have found that L.
plantarum can significantly improve chicken growth performance (Banu et al.
Peng et al.
Humam et al.
Wang et al.
Separately.
coagulans has also been shown to improve chicken performance (Zhang et al.
Zhen et al.
A comparison between L.
plantarum and another endospore-forming probiotic (B.
in broilers .
-98 days ol.
studied by Nam et al.
showed that treatment with L.
plantarum resulted in better BW compared to B.
subtilis while the FCR parameters demonstrate no significant differences, with the best value is observed with L.
plantarum treatment.
At the same time, the study on the effects of L.
plantarum and B.
coagulans on broilers .
-42 day.
shows that the treatment with L.
plantarum yields the most favorable outcomes for the Average Daily Gain (ADG) and Average Daily Feed Intake (ADFI) but not FCR parameters (Yu et al.
Metabolite profile Principal Component Analysis (PCA) is a method to reduce the dimensionality of specific datasets (Debik et al.
, 2.
It improves interpretability without losing much information (Hasan & Abdulazeez, 2.
The PCA score plot between the overlapping treatment groups in Figure 1 shows no significant difference between the treatments.
If there is no clear separation between groups on the PCA graph, then there is no significant effect between treatments, and it can be considered indistinguishable (Fiehn 2016.
Jiang et al.
However, the LP and BC treatment groups visually provide a more oval cluster than the other PCA loading plot graph serves to visualize the loading contribution of each metabolite to the variance observed in the data between treatments (Withers et al.
The further away from the center, the more influential the metabolite is to the treatment (Ren et al.
In summary, the loading plot illustrates the direction of projection of the metabolite features of the PCA score plot in space where it has the most extended vector for the highest variation in the data (Van Dyk.
Metabolites that contribute strongly, as seen in the loading plot in Figure 1, are palmitic acid, 1octadecanol, and 5-oxoproline.
Metabolites that indicated a negative correlation were 1-octadecanol and palmitic acid.
One comparative study examining the metabolites of the probiotics B.
coagulans and L.
plantarum through untargeted metabolomic analysis is reported by Cukkemane et al.
This study utilized various probiotics to ferment different milk media, including four lactic acid bacteria (LAB) and one spore-forming bacterium (SFB), specifically B.
coagulans ATCC 12425 and L.
plantarum NRC 716.
The PCA and heatmap analysis results of each class of detected compounds indicated that B.
coagulans and L.
plantarum metabolites differed significantly.
However, while utilizing the same bacteria, this study does not detail the chicken host's metabolite conditions, as it employs milk for fermentation.
Zhang et al.
reported significant differences in the PCA analysis of cecum samples from chickens undergoing LAB probiotic treatment in response to heat stress.
The PLS-DA score plot in Figure 2 does not demonstrate a clear separation between treatments.
However, the clustering observed indicates that the BC.
LP, and NC treatments exhibit distinct patterns and Worley and Powers .
state that PLSDA aggressively enforces separations between experimental groups and is often employed as an alternative method when PCA fails to reveal group However, this practice carries significant Without proper validation.
PLS-DA can quickly produce statistically unreliable group separations.
QA is the estimated value of a model's predictive ability, calculated through cross-validation.
A strong prediction will yield a high QA value.
conversely, if QA is negative, the model is deemed non-predictive (SzymaEska et al.
This study's PLS-DA model demonstrates positive QA values for three principal components (PC.
, precisely 0.
10, 0.
13, and 0.
Therefore, it can be concluded that the model has good predictive VIP (Variable Importance in Projectio.
is a parameter used to calculate a cumulative measure of the Pradhika et al.
Relationship of growth relationship between growth performance and metabolite profile of broiler chickens supplemented Figure 1.
The PCA score plot .
and loading plot .
illustrate the differences between treatments and influential metabolite Figure 2.
The PLS-DA score plot illustrates the enforced separation between treatments.
Clustering indicates that the BC.
LP, and NC treatments exhibit distinct patterns and directions but do not demonstrate clear separation influence of individual variables on the model (Galindo-Prieto et al.
, 2.
This analysis reflects the loading weight for each component and the response variability explained by the PLS-DA components that can be used for feature selection (Thevenot 2016.
Zheng et al.
Metabolites .
with VIP values >1 in PLS-DA models are identified as important differential metabolites (Deng et al.
Gromski et Mapping metabolites between treatments on a heatmap provides an overview of the hierarchical clustering of metabolite profiles (Vacanti 2.
Heat maps allow users to easily visualize changes in metabolite concentration patterns across samples and treatments, displaying actual data values using color gradients (Chong & Xia 2.
The heatmap dendrogram in Figure 3 shows that LP treatments are grouped with BC and continue to be further grouped with NC.
BC provided the most distinct profile compared to the other treatments.
Separation of important metabolites is done using VIP analysis in PLS-DA.
Metabolite screening based on VIP score >1 in Figure 3 resulted in palmitic acid, 1-octadecanol, acetic acid, lactic acid, and butanoic acid as important Broiler chickens with poor performance are indicated by the increase of several metabolites in the cecum, namely D-mannose, hexadecanoic acid, cholesterol.
L-valine.
L-leucine, glutamic acid, glucopyranose, -D-allopyranose and phosphoric acid (Chen et al.
In this study, it was described as increasing 1-octadecanol and glycolic acid.
Relationship between chicken performance and metabolite profile Rinttily & Apajalahti .
summarize that metabolites derived from microbiota composition can influence growth performance and suggest that the JITV Vol.
30 No.
2 Th.
2025:115-125
cecal microbial profile may reflect the efficiency of feed digestion and nutrient absorption in the intestine.
The relationship between chicken performance and metabolite profiles can be summarized in Table 4.
and BC profiles have significant metabolite differences and significant avgBW performance differences, particularly for 0-21 rearing days.
The primary metabolites differentiating between LP and BC quite far from the white mid-spectrum were acetic acid, lactic acid, and butanoic acid.
The BC treatment also had a NC.
Differentiating metabolites that are indicators of unsupplemented by probiotics are the decrease of palmitic acid and the increase of 1-octadecanol.
Metabolites expressed as differentiating metabolites .
iomarker candidate.
are lactic acid as a good avgBW performance distinguisher and 1-octadecanol and palmitic acid as distinguishers without probiotic Xing et al.
reported changes in unique compounds that could serve as biomarkers in the digestive tract of laying hens supplemented with B.
coagulans and exposed to lead (P.
These changes included the presence of antioxidant and antibacterial compounds, such as 4-acetamido butanoic acid, dodecanoic acid.
L-3-phenylacetic acid, apigenin, and Zhang et al.
found an increase in SCFA compounds such as acetic acid, propionic acid, butyrate, isobutyric acid, and valeric acid in the digestive tract of broiler chickens administered B.
Additionally.
Ito et al.
noted that while the concentrations of certain SCFAs, such as propionate and butyrate, would increase, other types, including acetate and lactate, would decrease.
Analysis of differences in growth performance characteristics in the administration of probiotics L.
plantarum and B.
coagulans resulted in only avgBW parameters significantly different with the highest to lowest weights in order LP.
BCLP.
BC.
NC.
Figure 3.
The heatmap .
and VIP score .
of identified metabolites with a proportion >0.
8 illustrate the hierarchical grouping of treatments and metabolites based on their relative levels.
Table 4.
Relationship mapping between significant performance parameters .
vgBW) and important metabolites with a VIP score >1 0-10 day LPa > BCLPab > BCb > NCb 0-21 day LPa > BCLPb > BCb > NCb 0-30 day LPa > BCLPab > BCab > NCb Metabolite Palmitic acid 1-Octadecanol Acetic acid Lactic acid Butanoic acid NC= negative control.
LP= Lactobacillus plantarum.
BC= Bacillus coagulans.
BCLP= Bacillus coagulans & Lactobacillus plantarum.
Diferent superscript letters mean significant different.
Treatments that do not have the same letter notation are significantly different (A= 0.
Pradhika et al.
Relationship of growth relationship between growth performance and metabolite profile of broiler chickens supplemented Other performance parameters .
djFCR, cumFI, and PEF) did not differ significantly.
Metabolite profile analysis on the administration of probiotics L.
plantarum and B.
coagulans in the digestive tract of broiler chickens with NC.
LP.
BC, and BCLP treatments gave results that were not significantly different after PCA analysis.
Important metabolites with VIP score >1 are acetic acid, lactic acid, butanoic acid, 1-octadecanol and palmitic acid.
Metabolites expressed as distinguishing metabolites between LP and BC are acetic acid, lactic acid, and butanoic acid.
At the same time, metabolites expressed as distinguishing metabolites of biomarker candidates are lactic acid as a good avrBW performance distinguisher and 1octadecanoic and palmitic acid as a distinguisher without probiotic supplementation.
CONCLUSION
The administration of probiotics Lactobacillus plantarum (LP) and Bacillus coagulans (BC) significantly influenced broiler growth performance, with average body weight .
vgBW) being the only parameter showing significant differences.
The highest to lowest avgBW values were observed in the order of LP.
BCLP.
BC, and NC treatments.
Other performance parameters, including adjFCR, cumFI, and PEF, showed no significant differences.
Metabolite profile analysis indicated no significant differences between treatments based on PCA and PLS-DA results.
However, important metabolites with a VIP score >1 were identified, including acetic acid, lactic acid, butanoic acid, 1-octadecanol, and palmitic acid.
Acetic acid, lactic acid, and butanoic acid were key distinguishing metabolites between LP and BC.
Additionally, lactic acid was identified as a potential biomarker for good avgBW performance, while 1-octadecanol and palmitic acid were differentiating metabolites in treatments without probiotic supplementation.
These findings suggest that probiotic supplementation can selectively influence broiler growth performance and metabolite profiles, providing valuable insights for optimizing broiler nutrition strategies.
ACKNOWLEDGEMENT
Aviagen.
Ross 308 Ross 308 FF broiler: performance Huntsville (USA): Aviagen.
Banu LA.
Mustari A.
Ahmad N.
Efficacy of probiotics on growth performance and hemato-biochemical parameters in broiler.
Res in Agri Livest and Fish, 6:
91Ae100.
DOI: 10.
3329/ralf.
Broadhurst D.
Goodacre R.
Reinke SN.
Kuligowski J.
Wilson ID.
Lewis MR.
Dunn WB.
Guidelines and considerations for the use of system suitability and quality control samples in mass spectrometry assays applied in untargeted clinical metabolomic studies.
Metabol.
14: 1-17.
DOI: 10.
1016/j.
Chen X.
Zhao W.
Liu YZ.
Aschalew ND.
Sun Z.
Zhang XF.
Zhen YG.
Caecal microbiome and metabolites associated with different growth performances of Indian J Anim Res.
55: 109-114.
DOI:
18805/ijar.
B-1062
Chong J.
Xia J.
Using MetaboAnalyst 4.
0 for metabolomics data analysis, interpretation, and integration with other omics data.
Comput Met Data Anal Metabol.
Cukkemane A.
Kumar P.
Sathyamoorthy B.
metabolomics footprint approach to understanding the benefits of synbiotics in functional foods and dietary therapeutics for health, communicable, and noncommunicable diseases.
Food Res Int.
128:108679.
DOI: 10.
1016/j.
Debik J.
Sangermani M.
Wang F.
Madssen TS.
Giskeydegyurd GF.
Multivariate analysis of NMRAabased metabolomic data.
NMR Biomed.
35:e4638.
DOI:
1002/nbm.
Elshaghabee FM.
Rokana N.
Gulhane RD.
Sharma C.
Panwar Bacillus as potential probiotics: status, concerns, and future perspectives.
Front Microb.
DOI: 10.
3389/fmicb.
Eurachem.
Blanks in Method Validation Supplement to Eurachem Guide The Fitness for Analytical Methods.
Bucharest (Romani.
: Eurachem.
Fiehn O.
Metabolomics by gas chromatography-mass spectrometry: Combined targeted and untargeted Cur Prot Mol Bio.
114: 30.
1Ae30.
DOI:10.
1002/0471142727.
Frainay C.
Jourdan F.
Computational methods to identify metabolic sub-networks based on metabolomic Briefings in Bioinformatics, 18: 43-56.
DOI:
1093/bib/bbv115 PT Nugen Bioscience Indonesia funded the study.
This company provides in-vivo experiments, probiotics, and metabolomic analysis.
GalindoAaPrieto B.
Eriksson L.
Trygg J.
Variable influence on projection (VIP) for orthogonal projections to latent structures (OPLS).
J Chemometrics.
28: 623632.
DOI: 10.
1002/ cem.
REFERENCES