Effect of Dietary Non-phytate Phosphorus Levels on the Diversity and

Structure of Cecal Microbiota in Meat Duck from 1 to 21 d of age

S. J. Dai, K. Y. Zhang, X. M. Ding, S. P. Bai, Y. H. Luo, J. P. Wang, and Q. F. Zeng1

Institute of Animal Nutrition, Key Laboratory for Animal Disease-Resistance Nutrition of China, Ministry of
Education, Sichuan Agricultural University, Chengdu, Sichuan, China, 611130

ABSTRACT The study was conducted to distinguish microbiota clustering based on dietary NPP levels oc-
the effect of dietary non-phytate phosphorus (NPP) lev- cured, with 0.22% NPP groups distinctly different from
els on the community diversity and structure of the the 0.46% and 0.58% NPP group samples. Moreover, di-
cecal microbiota in meat duck based on 16S rDNA etary NPP levels could change the relative abundance
high-throughput sequencing. In total, 525 1-d-old duck- of the phylum Proteobacteria (linear, P < 0.05), genera
lings were fed diets (105 ducklings, 7 pens of 15 duck- Eubacterium coprostanoligenes (quadratic, P < 0.05),
lings, on each diet) containing five levels of NPP (0.22, Ruminococcaceae UCG-014 (quadratic, P < 0.05)
0.34, 0.40, 0.46, and 0.58%) for 21 days. The results and Subdoligrannulum (linear, P < 0.05), and Lach-
showed that dietary NPP levels linearly and quadrat- nospiraceae family (quadratic, P < 0.05) in cecal mi-
ically increased (P < 0.05) 21 d body weight, 1 to crobiota of ducks. Increasing the dietary NPP level in-
21 d feed intake and NPP intake, and contrarily, lin- fluenced the cecal microbiota and positively affected the
early decreased (P < 0.05) β -diversity of cecal microbial growth of meat ducks.
population in ducks. ß-diversity analyses showed that
Key words: cecal microbiota, dietary non-phytate phosphorus, meat duck, performance

2018 Poultry Science 97:2441–2450

http://dx.doi.org/10.3382/ps/pey090

INTRODUCTION species (Stanley et al., 2014), which play a crucial role

in feed digestion, toxins degradation, pathogens exclu-
Dietary phosphorus (P) is an essential nutrient sion, stimulation of the immune system, and endocrine
not only for normal growth, skeletal system devel- activity (Zhu et al., 2002). Current exciting research is
opment, and maintenance of animals (Shastak and beginning to unravel how the composition of food mod-
Rodehutscord, 2013) but also for bacteria, since it is ulates the gut microbiota (Rajoka et al., 2017). There
needed for a variety of metabolic processes in bacte- is a promising dietary strategy, which has recently re-
rial cell (Durand and Komisarczuk, 1988). Moreover, it ceived much attention in human research, using a rat
was demonstrated that bacterial P and calcium (Ca) or pig model with dietary Ca and P content to modu-
assimilation and metabolic activity depend on P and late the intestinal eubiosis. Changes in Ca and P sup-
Ca availability in the large intestine of pigs (Mann et plementation affected the composition and activity of
al., 2014a). Mann et al. (2014b) found that high versus the microbial community in the digestive tract of broil-
adequate dietary Ca-P content significantly promoted ers (Ptak et al., 2015). Dietary higher P and Ca lev-
Lactobacillus by 14.9% units at the gastric Pars nong- els led to a shift in relative abundance of Lactobacillus
landularis of weaned pigs. An improved colonization re- (L.) reuteri and L. crispatus to L. salivarius and L. tai-
sistance against intestinal pathogens and promotion of wanensis in broiler’s crop (Witzig et al., 2015). Daniel
lactobacilli in ileal digesta and mucosa has been ob- et al. (2016) also found that diets supplemented with P
served with Ca-P rich diets in rats (Bovee-Oudenhoven affected the cecal microbiota and improved the growth
et al., 1999). These results suggested that dietary P con- of broilers. However, research on the interaction be-
centration could affect gut health partly by modulating tween dietary NPP levels and cecal microbiota of ducks
the composition of gut microbiota. is poorly understood. In a qualitative comparison, there
The microbial community present in the broiler’s gas- were clear distinctions in the structure and composition
trointestinal tract (GIT) has more than 900 bacterial of gut microbiome between Pekin duck cecal contents
and those of broiler chickens or turkeys. Therefore, we

C 2018 Poultry Science Association Inc. suppose that it is not advisable to extrapolate the re-
Received November 21, 2017. sults of studies on affecting the microbiomes from one
Accepted February 22, 2018.
1
Corresponding author: [email protected] bird order to another (Best et al., 2017).

2441
2442 DAI ET AL

Therefore, in the present study, we investigated the Table 1. Ingredients and compositions of the
hypothesis that the diversity and composition of the basal diets (%, dry matter basis).
intestinal microbiota is linked with dietary NPP levels Ingredients contents
in ducks. The objective was to determine if there were
any differences in microbial diversity and/or relative Corn 59.50
Soybean Oil 1.87
abundance of bacterial taxa, at the phylum, class, order, Soybean meal 33.85
family, and genus levels, within the cecal microbiota in L-Lysine-HCL 0.024
meat ducks from 1 to 21 d of age. DL-Methionine 0.197
Limestone 1.905
Monocalcium phosphate 0.395
Tryptophan 0.061
Bentonite3 1.168
MATERIALS AND METHODS Sodium chloride 0.35
Choline chloride 0.15
The Institutional Animal Care and Use Committee Vitamin premix1 0.03
of Sichuan Agricultural University approved all proce- Mineral premix2 0.50
dures used in the study. Total 100.00
Calculated nutrients levels (%)
Metabolizable energy (Kcal/kg) 2,900.00
Crude protein 19.5
Calcium 0.90
Birds, Diets, and Management non-phytate phosphorus 0.22
Digestible lysine 0.96
A total of 525 1-d-old ducks (Cherry Valley, Sichuan Digestible methionine 0.46
Mianying Breeding Duck Co. Ltd. Mianyang, China) Digestible threonine 0.66
Digestible tryptophan 0.26
were randomly allocated to one of five treatments with Total phosphorus (analyzed) 0.54
seven replicate cages of 15 birds per cage in a com- 1
Provided per kilogram of diet vitamin A, 8,000 IU;
pletely randomized design and fed a basal corn-soybean cholecalciferol, 2,000 IU; vitamin E, 5 IU; vitamin K,
meal diet (NPP 0.13%) supplemented with 0.09, 0.21, 1 mg; thiamine, 0.4 mg; riboflavin, 3.2 mg; pyridoxine,
0.27, 0.33, or 0.45% of inorganic phosphorus in the form 1.2 mg; vitamin B12 , 6 μ g; folic acid, 100 μ g; niacin,
7 mg; calcium pantothenate, 5 mg.
of monobasic sodium phosphate (CaH2 PO4 ·2H2 O). The 2
Provided per kilogram of diet: Fe (FeSO4 ·H2 O)
analyzed total P (TP) content in experimental diets 80 mg, Cu (CuSO4 ·5H2 O) 8 mg, Mn (MnSO4 ·H2 O)
was 0.54, 0.72, 0.79, 0.84, or 0.92%, and the correspond- 70 mg, Zn (ZnSO4 ·H2 O) 90 mg, I (KI) 0.4 mg, Se
(Na2 SeO3 ) 0.3 mg.
ing dietary non-phytate phosphorus (NPP) levels were 3
Regulation of limestone, monocalcium and ben-
0.22, 0.34, 0.40, 0.46, and 0.58%, respectively. Each diet tonite to formulate the other 4 diets containing 0.34,
contained a constant calcium content of approximately 0.40, 0.46, and 0.58% NPP. The other ingredients in all
diets keep the same.
0.9%. The period of the experiment was from 1 to 21 d
of age.
The basal diet (1 to 21 d) (Table 1, 2-mm-diameter- DNA extraction and high-throughput
pellet) was formulated under digestible amino acid basis sequencing and analysis of 16S rDNA gene
to meet the nutrient requirements of Pekin ducks sug-
amplicons
gested in the NRC (1994) and Han et al. (2017). Ducks
were reared in pens (2.2 m × 1.2 m × 0.9 m) in a DNA extraction and high-throughput sequencing and
temperature- and humidity-controlled room with analysis of 16S rDNA gene amplicons were performed
24-h constant light schedule and free access to water using the Illumina Hiseq platform Novo gene (Novo
and feed. gene Bioinformation Technology, Beijing, China). The
total genome DNA from the cecal samples was ex-
tracted using the sixteen alkyl three ethyl ammo-
Data Collection and Sampling nium bromide (CTAB)/ sodium alkyl sulfonate twelve
(SDS) method (Bo et al., 2017). DNA was quantified in
On d 21, the body weight (BW) and feed consump- a NanoDrop ND-1000 spectrophotometer (Thermo Sci-
tion of ducks were recorded on basis. Average daily gain entific, Waltham, MA) and stored at −20◦ C (Stanley
(ADG), average daily feed intake (ADFI), feed con- et al., 2012).
version ratio (FCR), NPP, and TP intake were calcu- According to the concentration, DNA was diluted
lated. Feed waste was recorded daily, and the data were to 10 ng/μ L using sterile water. The 16S rDNA
used in the calculations of feed consumption. Then, one genes of distinct regions (16S V4) were amplified us-
bird per pen was euthanized by carbon dioxide asphyx- ing specific primer (515F GTGCCAGCMGCCGCG-
iation following anesthesia in a gas mixture (35%CO2 , GTAA; 806R GGACTACHVGGGTWTCTAAT) with
35%N2 , and 30% O2 ) (Zeller et al., 2015). The GIT was the unique barcodes. All PCR reactions were car-
immediately dissected after euthanization, the two ceca ried out with Phusion High-Fidelity PCR Master Mix
were opened longitudinally and digesta samples were (New England Biolabs). PCR products were mixed
collected with a sterile spoon. Samples were stored at in equal density ratios. Then, mixture PCR products
−80◦ C. were purified with Qiagen Gel Extraction Kit (Qiagen,
 DIETARY NON-PHYTATE PHOSPHORUS ON CECAL MICROBIOTA 2443
Table 2. Effect of dietary non-phytate phosphorus levels on growth performance and total phosphorus intake
in meat duck during 1 to 21 d of age.

Dietary non-phytate phosphorus levels, % P-Value

Parameter1 0.22 0.34 0.40 0.46 0.58 SEM2 ANOVA Linear Quadratic

ADG(g) 59.72a 68.6b 69.0b 70.8b 69.0b 0.85 0.001 0.001 0.001
ADFI(g) 102.8a 114.5b 116.6b 116.9b 116.9b 1.33 0.001 0.001 0.001
FCR(g/g) 1.72 1.67 1.69 1.65 1.68 0.02 0.066 0.079 0.061
NPP intake (g) 4.75a 8.18b 9.79c 11.3d 14.1e 0.10 0.001 0.001 0.001
TP intake(g) 11.7a 17.3b 19.3c 20.6d 22.4e 0.19 0.001 0.001 0.001
a-e
Means in the same row with no common superscript are significantly different (P < 0.05).
1
ADG: Average Daily Gain; ADFI: Average Daily Feed Intake; FCR: Feed Conversion Ratio; NPP: non-phytate phos-
phorus; TP: Total Phosphorus.
2
Means represent 7 pens of 15 ducks per pen.

Hilden, Germany). Sequencing libraries were generated Statistical Analysis

using TruSeq DNA PCR-Free Sample Preparation Kit
(Illumina, San Diego, CA) following the manufacturer’s The effects of dietary NPP levels on BW, FI, FCR,
recommendations, and index codes were added. The li- TP intake, NPP intake, and the structure and diver-
brary quality was assessed on the Qubit 2.0 Fluorom- sity of the cecal microbiota were determined by a one-
eter (Thermo Scientific) and Agilent Bioanalyser 2100 way ANOVA using the GLM procedure in SAS software
system. Finally, the eligible libraries were sequenced on (SAS Institute Inc., Cary, NC, 2004). When the dietary
an Illumina HiSeq 2500 platform and 250 bp paired-end NPP levels effect was significant (P < 0.05), polynomial
reads were generated. contrasts and the linearity of response were examined to
analyze dietary NPP levels using linear and quadratic
regressions. The R2 was provided to compare these re-
gressions when the linear or quadratic effect was signif-
Data Analysis icant (P < 0.05) (Pesti et al., 2009). Probability values
Paired-end reads were assigned to samples based on < 0.05 were considered significant.
their unique barcode and truncated by cutting off the
barcode and primer sequence. Paired-end reads were RESULTS
merged using FLASH (V1.2.7), a very fast and accurate
analysis tool, which was designed to merge paired-end Growth Performance
reads when at least some of the reads overlap the read
generated from the opposite end of the same DNA frag- Dietary NPP levels linearly and quadratically in-
ment, and the splicing sequences were called raw tags. creased (P < 0.05) ADG, ADFI, TP intake, and NPP
Quality filtering on the raw tags was performed under intake of ducks during 1 to 21 d of age (Table 2). When
specific filtering conditions to obtain the high-quality compared to other groups, ducks fed diets containing
clean tags, according to the QIIME (V1.7.0) quality- 0.22% NPP had lower (P < 0.05) ADG and ADFI. Di-
control process. Reads were filtered by QIIME quality etary NPP levels significantly affected (P < 0.05) TP
filters. Sequences with ≥97% similarity were assigned intake and NPP intake among all treatments. However,
to the same optimal taxonomic units (OTUs). Then, dietary NPP levels had no marked effect on FCR (P >
a representative sequence was chosen for each OTU to 0.05).
annotate the taxonomic information of that unit. All
OTUs were subsequently analyzed for abundance and
diversity. For α-diversity analysis, rarefaction curves
Alpha-Diversity
and rank abundance curves were generated by R project Illumina Hiseq high-throughput sequencing was per-
(Version 2.15.3). formed to evaluate the bacterial abundance. On aver-
Sequences (clean data) were analyzed using age, we obtained 81,488 total reads. Reads were then
the Quantitative Insights into Microbial Ecology merged based on the overlapped sequences to gen-
(QIIME, Version 1.7.0) software package, and in- erate tags. On average, 75,767 total tags were ob-
house Perl scripts were used to analyze α-diversity tained. We clustered the tags using a 97% similarity
(Observed-species, Shannon, Simpson, Chao I, ACE, cutoff to obtain an average of 1,319 OTUs. Rarefac-
Goods coverage, and PD whole tree). A jackknifed tion curves were analyzed together to test the suffi-
β -diversity analysis was conducted to assess the ciency of the sequence collection (Figure 1). In rarefac-
statistical variation of sample location in principal tion curves, the curve of the observed species number
coordinate analysis (PCoA) plots based on unweighted plateaued with the increase in samples/sequences num-
UniFrac distances (Lozupone et al., 2011). ber, which indicated that enough samples/sequences
2444 DAI ET AL

Figure 1. Alpha diversity analysis of different groups. A, Rarefaction curve is generated by setting the number of sequence as x-axis, and
the number of observed species as y-axis. The curve reflected the relationship between the quantity of observed species and sequences. The
“plateaued” shape of the curve indicated that enough samples/sequences were obtained to cover the majority of species; B, each circle in the
Venn diagram represented one group noted by the name of same color. The numbers located in the overlapping area represented the number of
OTUs share with respective groups. The numbers located in the individual area represented the number of OTUs peculiar to the representative
group. Caecum 1, Caecum 2, Caecum 3, Caecum 4, and Caecum 5 represent dietary non-phytate phosphorus contents were 0.22, 0.34, 0.40, 0.46,
and 0.58%, respectively.

were obtained to cover the majority of species. A Venn In general, there was a significant linear decrease (P <
diagram was generated based on the OTU classifica- 0.05) in the diversity of the microbial populations by
tion (Figure 1). The five groups shared 570 OTUs in all metrics as dietary NPP levels increased (Table 3).
common. When considering the Shannon diversity metric, which
Multiple α-diversity metrics revealed clear differences takes into account the richness and evenness of species,
among microbial populations in ducks fed diets with dif- a pattern of dietary NPP levels emerges. This was sup-
ferent NPP levels with respect to richness and evenness. ported by statistically significant differences between
 DIETARY NON-PHYTATE PHOSPHORUS ON CECAL MICROBIOTA 2445
Table 3. Effects of dietary non-phytate phosphorus levels on alpha diversity metrics of cecal microbiota in meat
duck at 21 d of age.

Dietary non-phytate phosphorus levels, % P-Value

Parameters 0.22 0.34 0.40 0.46 0.58 SEM ANOVA Linear Quadratic

Observed species 743.21 802.8 797.0 513.2 496.8 97.15 0.077 0.019 0.251
Shannon 6.33a 6.39a 6.80a 5.80a,b 5.68b 5.39 0.014 0.001 0.783
Simpson 0.97 0.96 0.93 0.94 0.92 0.02 0.262 0.053 0.632
Chao1 817.0 941.0 881.5 548.5 551.3 111.9 0.056 0.017 0.232
ACE 830.5 944.1 888.7 558.4 559.3 113.6 0.064 0.018 0.253
Goods coverage 0.998 0.997 0.997 0.999 0.999 0.0004 0.079 0.032 0.237
PD whole tree 55.70a 58.90a 55.41a 34.88b 36.88b 6.21 0.028 0.005 0.413
a,b
Means in the same row with no common superscript are significantly different (P < 0.05).
1
Means represent 5 ducks per pen.

the 0.22%, 0.34%, or 0.40% NPP groups and the 0.58% (phylum Firmicutes, 50.7%), Clostridiales (50.7%),
NPP group (Table 2, ANOVA, P < 0.05), the same as Ruminococcaceae (phylum Firmicutes, 37.5%), and
PD whole tree metrics. Bacteroides (28.8%) were dominant.
Certain taxa were identified as differentially abun-
dant, according to dietary NPP levels (Figure 4).
Beta-Diversity First, we found that the relative abundance of the
phylum Proteobacteria linearly decreased (P < 0.05)
To deeply illustrate the differences in cecal flora be- with increasing dietary NPP levels. Other remarkable
tween these groups, β -diversity analysis was performed distinctions occurred in the genera Eubacterium
to analyze the phylogenetic relationship and relative coprostanoligenes, Ruminococcaceae UCG-014, and
abundance of species. The structure of the duck ce- Subdoligrannulum. The increase of dietary NPP
cal microbial populations is distinct and clearly corre- levels, quadratically increased (P < 0.05) the rela-
lated with dietary NPP levels on β -diversity measures. tive abundance of Eubacterium Coprostanoligenes, and
In the unweighted UniFrac principal coordinates anal- linearly increased (P < 0.05) the relative abundance
ysis (PCoA), the first principal coordinate (PC1) ex- of Subdoligrannulum, while Ruminococcaceae UCG-014
plained 29.89% of the variation among samples, with quadratically decreased (P < 0.05). Ducks fed diets
PC2 explaining 12.32% of the variation. The PCoA with 0.46% and 0.58% NPP had a higher (P < 0.05)
can estimate the main discrepancies in the OTUs be- abundance of Subdoligrannulum than that of ducks fed
tween groups using the distance of sample dots. Then, the other three experimental diets. Some species also
PCoA was calculated based on samples’ distance matri- showed some interesting variations influenced by di-
ces, which were generated based on their group species etary condition. The abundance of the family Lach-
phylogenic and evolutionary relationships. In our study, nospiraceae present a quadratic increase (P < 0.05)
unweighted UniFrac PCoA showed that the species dis- with the increase of dietary NPP levels.
crepancy was obvious between the high dietary NPP
contents (0.46% and 0.58%) and the low dietary NPP
contents (0.22% and 0.34%) (Figure 2, A). According DISCUSSION
to weighted UniFrac PCoA (Figure 2, B), the species
discrepancy was also distinguished between the high di- The actual NPP requirement of meat ducks aged 1
etary NPP contents (0.46% and 0.58%) and the low to 21 d of age was 0.40% (NRC, 1994). A lower level
dietary NPP contents (0.22%). of NPP in the diet (0.22%) reduced ADG and ADFI
in ducks aged 1 to 21 d in the current study. Simi-
lar to the result of the present study, Rama Rao et al.
The Relative Abundance of Some Bacteria (2006) found that depression in weight gain and FI was
observed at higher levels of Ca (0.8% and 0.9%) with
The relative abundance of bacteria was analyzed at lower levels of NPP in diet (0.3% and 0.35%) of broil-
the levels of phylum, class, order, family, and genus. The ers. Zeng et al. (2015) also observed that meat ducks
structure and stability of the gut microenvironment can fed diet with 0.23% NPP had a lower FI than ducks fed
be denoted by the relative abundance of different types diet with 0.28% NPP.
of bacteria. The heat map was generated using the rela- One major finding of the present study was that
tive abundance of different phylum (Figure 3). In meat higher dietary NPP levels (0.46% and 0.58%) decreased
ducks at 21 d of age, the cecal microbiota was mainly α-diversity, and the species discrepancy was obvious
composed of Firmicutes (55%), Bacteroidetes (33.4%), when compared to a lower dietary NPP level (0.22%).
Actinobacteria (3.3%), Tenericutes (4.38%), Proteobac- These results agreed with Palacios et al. (2008), who
teria (2.89%), and unclassified bacteria (1.03%). At the observed that in the ileal and cecal sections, the
class, order, family, and genus levels, the Clostridia more diverse microbial communities are responsible for
2446 DAI ET AL

Figure 2. Beta diversity analysis (PCoA) plot (based on OTUs) according to dietary non-phytate phosphorus levels. A, unweighted unifrac
PCoA; B, weighted unifrac PCoA. PC1 and PC2 in x-and y-axis represented two principle discrepancy components between groups, and the
percentage in bracket means contribution value to the discrepancies by the component. Dots represent samples. Samples in same group share
same color. Caecum 1, Caecum 2, Caecum 3, Caecum 4, and Caecum 5 represent dietary non-phytate phosphorus contents were 0.22, 0.34, 0.40,
0.46, and 0.58%, respectively.

phytate degrading activities. This might be due to the Furthermore, P is important for bacterial proliferation,
lack of sufficient enzymes (such as endogenous mucosal and contributes considerably to bacterial structure and
phytase and phosphatase) in the proximal GIT, which metabolic processes (Lengeler et al., 1999). According
lead to an incomplete hydrolysis of phytate in the body to the results of the present study, ducks fed diets with
of non-ruminant animals (Maenz and Classen, 1998; 0.22% NPP had poor growth performance but had a
Onyango and Adeola, 2009; Selle et al., 2012). This higher α-diversity of cecal microbial population. One
phenomenon caused most of the phytate-P in the diet reason for this result may be because the ducks fed di-
to enter the hindermost ileum and cecum to change ets with 0.22% NPP need to spend more energy to com-
bacterial structure and metabolic processes. Therefore, pete with the gut microbial population to obtain more
when ducks are fed a diet with available P deficiency, P and other nutrients. In addition, increasing diversity
their gut microbial communities need to obtain P from in cecal microbiota in ducks fed diets with 0.22% NPP
the degradation of phytate. Moreover, according to may imply the increase of several enteric pathogens.
Wood and Clark (1988), a surplus of P can be stored Some enteric pathogens, for example, have developed
as polyphosphates in bacterial cells and may be used strategies to metabolically utilize myo-inositol from
as an energy and P source for metabolic processes. phytate-P degradation as a carbon and energy source.
 DIETARY NON-PHYTATE PHOSPHORUS ON CECAL MICROBIOTA 2447

Figure 3. The heat map of relative abundance of different phylum. Samples information is transverse listed, and species annotations are
longitudinal shown. The left clustering tree is species-related clustering tree, and the upper tree is sample-related clustering tree. The heat map
was performed by discrepancies of species-relative abundance between samples, with colors gradually changed from deep red to deep blue, in
accordance with high relative abundance to low. Caecum 1, Caecum 2, Caecum 3, Caecum 4, and Caecum 5 represent dietary NPP contents were
0.22, 0.34, 0.40, 0.46, and 0.58%, respectively.

This feature applies to Gram-positive enteropathogens et al., 2015). These results indicated that dietary NPP
such as Enterococcus faecalis, Bacillus cereus, Listeria levels did not change the composition of gut microbiota
monocytogenes, and Clostridium perfringens, as well in meat ducks, suggesting that the complexity and the
as Gram-negative pathogens, such as Salmonella ty- anaerobic environment of the ceca did not allow impor-
phimurium, which may under certain circumstances tant changes in the bacterial community (Ley et al.,
cause infections in pigs (Staib and Fuchs, 2014). 2008).
Therefore, intervention studies are required to further Moreover, in the current study, we first found that di-
study dietary P deficiency or the increase of dietary etary NPP levels could change the relative abundance
phytate-P to determine whether they increase intestinal of the phylum Proteobacteria, the genera Eubacterium
pathogenic bacterial colonization and impair gut health coprostanoligenes, Subdoligrannulum, and Ruminococ-
of poultry. caceae UCG-014, and the family Lachnospiraceae in the
As previously noted, at the phylum level, Firmicutes cecal digesta of ducks. Individuals fed on the Mediter-
and Bacteroidetes (more than 80% of sequences) were ranean diet (a balanced intake of fruits, grains, monoun-
major bacteria in the duck gut microbiota (Vasaı̈ et al., saturated fat, vegetables, and polyunsaturated fat) had
2014). Best et al. (2017) also found barn-raised ducks lower numbers of Bacillaceae, Proteobacteria and acute
contain four major phyla found most often in other ani- phase C-reactive proteins (Marlow et al., 2013; De
mal systems as follows: Firmicutes (57%), Bacteroidetes Filippis et al., 2016), suggesting the decrease of the rela-
(30%), Proteobacteria (6%), and Actinobacteria (3%). tive abundance of the phylum Proteobacteria represents
Consistently, in the current study, majority of the mi- a healthier condition. Ruminococcaceae UCG-014 is a
croorganisms colonizing the cecum of ducks belonged to common genus reported in the chicken ceca (Bjerrum
the phyla Firmicutes and Bacteroidetes (more than 80% et al., 2006; Mohd et al., 2015), which has been asso-
of sequences), which were commonly described in pre- ciated with the maintenance of gut health and has the
vious studies that characterized the microbial commu- enzymatic capability to degrade cellulose and hemicel-
nities of the chicken GIT (Stanley et al., 2013; Deusch lulose (Biddle et al., 2013). The family Lachnospiraceae
2448 DAI ET AL

Figure 4. Significant difference in relative abundance analysis of some bacterial species according to dietary non-phytate phosphorus levels.
A, relative abundance of Phylum Proteobacteria; B, relative abundance of Genera Ruminococcaceae UCG-014; C, relative abundance of Genera
Eubacterium Coprostanoligenes; D, relative abundance of Genus Subdoligrannulum; E, relative abundance of Family Lachnospiraceae.

was reported to be associated with corn-based diets and acid. Heyer et al. (2015) considered that P deficiency
is mainly composed by anaerobes and some Clostrid- caused a reduction in SCFA synthesis due to reduce
ium members (Munyaka et al., 2015). OTUs belong- fermentation of cellulose in pig, which suggested that
ing to Lachnospiraceae are known to degrade complex the activity of bacterial fibrolytic enzymes is modulated
polysaccharides to SCFA. In the Lachnospiraceae fam- by available P of the surrounding medium (Metzler and
ily, OTUS accounted for almost 60% of the mucin- Mosenthin, 2008). Moreover, several authors have con-
adhered microbiota (Roos et al., 2000), and there were cluded from the results of in vitro studies (Francis et al.,
more abundant in the diet supplementation with P 1978; Lee et al., 1985) that P, as a coenzyme, is essential
(Daniel et al., 2016). Supplementation of Ca in the diet for bacterial degradation of dietary fiber. Also, the bac-
enhanced the presence of an uncultured Subdoligran- terial synthesis of fibrolytic enzymes is strongly depen-
ulum sp., which was previously found in the ceca dent on a sufficient P supply. These results were in ac-
of turkeys (Scupham, 2007) and can produce butyric cordance with current study. According to the findings
 DIETARY NON-PHYTATE PHOSPHORUS ON CECAL MICROBIOTA 2449
above, dietary P content could affect the abundance of Durand, M., and S. Komisarczuk. 1988. Influence of major minerals
bacteria related with the production of SCFA and the on rumen microbiota. J. Nutr. 118:249–260.
Francis, G. L., J. M. Gawthorne, and G. B. Storer. 1978. Factors
maturation of gut microbiota, which suggested that affecting the activity of cellulases isolated from the rumen digesta
regulation of the dietary P supply could improve of sheep. Appl. Environ. Microbiol. 36:643–649.
growth performance of meat ducks by regulating gut Han, H. Y., K. Y. Zhang, X. M. Ding, S. P. Bai, Y. H. Luo, J.
health. P. Wang, and Q. F. Zeng. 2017. Effect of dietary fiber levels
on performance, gizzard development, intestinal morphology, and
nutrient utilization in meat ducks from 1 to 21 days of age. Poult.
Sci. 96:4333–4341
CONCLUSIONS Heyer, C. M., E. Weiss, S. Schmucker, M. Rodehutscord, L. E.
Hoelzle, R. Mosenthin, and V. Stefanski. 2015.The impact of
In conclusion, this study uses a high-throughput py- phosphorus on the immune system and the intestinal microbiota
rosequencing based on the 16S rDNA gene in cecal sam- with special focus on the pig. Nutr. Res. Rev. 28:67–82
ples from ducks and provides a first inventory of the Lee, S. F., C. W. Forsberg, and L. N. Gibbins. 1985
microbial community and the effect of dietary NPP lev- Cellulolytic activity of Clostridium acetobutylicum. Appl.
Environ. Microbiol. 50:220–228.
els on the abundance of major different groups. Higher Lengeler, J. W., G. Drews, and H. G. Schlegel. 1999. Biology of the
(0.58%) or lower (0.22%) dietary NPP contents affected Prokaryotes, Stuttgart: Georg Thieme.
the diversity of cecal samples and had a significant ef- Ley, R. E., M. Hamady, C. Lozupone, P. J. Turnbaugh, R. R. Ramey,
and J. S. Bircher. 2008. Evolution of mammals and their gut
fect in modifying the bacterial community in the ceca. microbes. Science 320:1647–1651.
The increase in the dietary NPP supply influenced the Lozupone, C., M. E. Lladser, D. Knights, J. Stombaugh, and R.
ceca microbiota and positively affected the growth per- Knight. 2011. UniFrac: an effective distance metric for microbial
formance of meat ducks. Further investigations are nec- community comparison. ISME J 5:169–172.
Maenz, D. D., and H. L. Classen. 1998. Phytase activity in the
essary to understand the functional microbiota in ducks small intestinal brush border membrane of the chicken. Poult.
to improve the gut health and the availability of dietary Sci. 77:557–563.
P, and to thus reduce P loss by excretion. Mann, E., M. Dzieciol, B. U. Metzler-Zebeli, M. Wagner, and S.
Schmitz-Esser. 2014. Microbiomes of unreactive and patholog-
ically altered ileocaecal lymph nodes of slaughter pigs. Appl.
ACKNOWLEDGMENTS Environ. Microbiol. 80:193–203.
Mann, E., S. Schmitz-Esser, Q. Zebeli, M. Wagner, M. Ritzmann,
This research was supported by grants from and B. U. Metzler-Zebeli. 2014b. Mucosa-associated bacterial mi-
crobiome of the gastrointestinal tract of weaned pigs and dynam-
the National Key R & D Program of China ics linked to dietary calcium-phosphorus. PLoS One 9:e86950.
(2017YFD0502004), and Sichuan Agricultural Univer- Marlow, G., S. Ellett, I. R. Ferguson, S. Zhu, N. Karunasinghe, and
sity 211 Foundation of China. A. C. Jesuthasan. 2013. Transcriptomics to study the effect of a
Mediterranean-inspired diet on inflammation in Crohn’s disease
patients. Hum Genomics. 7:24.
REFERENCES Metzler, B. U., and R. Mosenthin. 2008. A review of interactions
between dietary fiber and the gastrointestinal microbiota and
Best, A. A., A. L. Porter, S. M. Fraley, and G. S. Fraley. 2017. their consequences on intestinal phosphorus metabolism in grow-
Characterization of Gut Microbiome Dynamics in Developing ing pigs. Asian Australas. J. Anim. Sci Sci. 21:603–615.
Pekin Ducks and Impact of Management System. Front. Micro- Mohd, S. M. A., C. C. Sieo, C. W. Chong, H. M. Gan, and Y. W.
biol. 7:2125. Ho. 2015. Deciphering chicken gut microbial dynamics based on
Biddle, A., L. Stewart, J. Blanchard, and S. Leschine. 2013. Un- high-throughput 16S rRNA metagenomics analyses. Gut Pathog.
tangling the genetic basis of fibrolytic specialization by Lach- 7:4.
nospiraceae and Ruminococcaceae in diverse gut communities. Munyaka, P. M., N. K. Nandha, E. Kiarie, C. M. Nyachoti, and E.
Diversity 5:627–640. Khafipour. 2016. Impact of combined β -glucanase and xylanase
Bjerrum, L., R. M. Engberg, T. D. Leser, B. B. Jensen, K. Finster, enzymes on growth performance, nutrients utilization and gut mi-
and K. Pedersen. 2006. Microbial community composition of the crobiota in broiler chickens fed corn or wheat-based diets. Poult.
ileum and cecum of broiler chickens as revealed by molecular and Sci. 95:528–540.
culture-based techniques. Poult. Sci. 85:1151–1164. National Research Council. 1994. Nutrient Requirements of Poultry.
Bo, T., S. Shao, D. Wu, S. Niu, J. Zhao, and L. Gao. 2017. Relative 9th rev. ed. Natl. Acad. Press,Washington, DC.
variations of gut microbiota in disordered cholesterol metabolism Onyango, E. M., and O. Adeola. 2009. Dietary phytate (inositol
caused by high-cholesterol diet and host genetics. Microbiology- hexaphosphate) regulates the activity of intestinal mucosa phy-
Open 00:e491. tase. J. Anim. Physiol. Anim. Nutr. 93:639–646.
Bovee-Oudenhoven, I. M., M. L. Wissink, J. T. Wouters, and R. Van Palacios, M. C., M. Haros, Y. Sanz, and C. M. Rosell. 2008.Selection
der Meer. 1999. Dietary calcium phosphate stimulates intestinal of lactic acid bacteria with high phytate degrading activity for
lactobacilli and decreases the severity of a Salmonella infection in application in whole wheat breadmaking. LWT - Food Science
rats. J. Nutr. 129:607–612. and Technology. 41:82–92.
Daniel, B.-M., V. Marius, S. Vera, R. Markus, and C.-S. Amélia Pesti, G. M., D. Vedenov, J. A. Cason, and L. Billard. 2009. A
2016. Insights into Broilers Gut Microbiota Fed with Phospho- comparison of methods to estimate nutritional requirements from
rus, Calcium, and Phytase Supplemented Diets. Front. Microbiol. experimental data. Br. Poult. Sci. 50:16–32.
7:2033. Ptak, A., M. R. Bedford, S. Światkiewicz, K. Żyla, and D. Józefiak.
De Filippis, F., N. Pellegrini, L. Vannini, I. B. Jeffery, A. La Sto- 2015. Phytase modulates ileal microbiota and enhances growth
ria, and L. Laghi. 2016. High-level adherence to a Mediter- performance of the broiler chickens. PLoS One. 10:e0119770.
ranean diet beneficially impacts the gut microbiota and associ- Rajoka, M. S. R., J. L. Shia, M. M. Hafiza, J. Zhua, Q. Lia, D. Y.
ated metabolome. Gut 65:1812–1821. Shao, Q. S. Huang, and Y. K. Hui. 2017. Interaction between diet
Deusch, S., B. Tilocca, A. Camarinha-Silva, and J. Seifert. 2015. composition and gut microbiota and its impact on gastrointesti-
News in livestock research-use of Omics-technologies to study the nal tract health. Food Sci. Hum. Wellness. 6:121–130.
microbiota in the gastrointestinaltract of farm animals. Comput. Rama Rao, S. V., M. V. L. N. Raju, M. R. Reddy, and P. Pa-
Struct. Biotechnol. J. 13:55–63. vani. 2006. Interaction between dietary calcium and non-phytate
2450 DAI ET AL

phosphorus levels on growth, bone mineralization and mineral Shastak, Y., and M. Rodehutscord. 2013. Determination and
excretion in commercial broilers. Anim. Feed Sci. Technol. 131: estimation of phosphorus availability in growing poultry
135–150. and their historical development. Worlds Poult. Sci. J. 69:
Roos, S., F. Karner, L. Axelsson, and H. Jonsson. 2000. Lactobacil- 569–586.
lus mucosae sp. nov., a new species with in vitro mucus-binding Vasaı̈, F., K. Brugirard Ricaud, M. D. Bernadet, L. Cauquil, O.
activity isolated from pig intestine. Int. J. Syst. Evol. Microbiol. Bouchez, and S. Combes. 2014. Overfeeding and genetics affect
50:251–258. the composition of intestinal microbiota in Anasplatyrhynchos
SAS Institute Inc. 2004. SAS User’s Guide. Statistics. Version 9.1ed. (Pekin) and Cairinamoschata (Muscovy) ducks. FEMS Microbiol.
SAS Institute Inc., Cary, NC. Ecol. 87:204–216.
Scupham, A. J. 2007.Succession in the intestinal microbiota of pread- Witzig, M., A. CamarinhaDaSilva, R. Green-Engert, K. Hoelzle, E.
olescent turkeys. FEMS Microbiol. Ecol. 60:136–147. Zeller, and J. Seifert. 2015. Spatial variation of the gut microbiota
Selle, P. H., A. J. Cowieson, and N. P. Cowieson. 2012. Protein in broiler chickens as affected by dietary available phosphorus and
phytate interactions in pig and poultry nutrition: a reappraisal. assessed by T-RFLP analysis and 454 pyrosequencing. PLoS One
Nutr. Res. Rev. 25:1–17. 10:e0143442.
Staib, L., and T. M. Fuchs. 2014. From food to cell: nutrient exploita- Wood, H. G., and J. E. Clark. 1988. Biological aspects of inorganic
tion strategies of enteropathogens. Microbiology. 160:1020–1039. polyphosphates. Annu. Rev. Biochem. 57:235–260.
Stanley, D., S. E. Denman, R. J. Hughes, M. S. Geier, T. M. Crow- Zeller, E., M. Schollenberger, M. Witzig, Y. Shastak, I. Kühn, and
ley, and H. Chen. 2012. Intestinal microbiota associated with L. E. Hoelzle. 2015. Interactions between supplemented mineral
differential feed conversion efficiency in chickens. Appl Microbiol phosphorus and phytase on phytate hydrolysis and inositol phos-
Biotechnol. 96:1361–1369. phates in the small intestine of broilers. Poult. Sci. 94:1018–1029.
Stanley, D., M. S. Geier, S. E. Denman, V. R. Haring, T. M. Crowley, Zeng, Q., X. Huang, Y. Luo, X. Ding, S. Bai, J. Wang, X. Yue,
and R. J. Hughes. 2013. Identification of chicken intestinal micro- Z. Su, Y. Liu, and K. Zhang. 2015. Effects of a Multi-enzyme
biota correlated with the efficiency of energy extraction from feed. Complex on Growth Performance, Nutrient Utilization and Bone
Vet. Microbiol. 164:85–92. Mineralization of Meat Duck. J Animal Sci Biotechnol. 6:12–20.
Stanley, D., R. J. Hughes, and R. J. Moore. 2014. Microbiota of the Zhu, X. Y., T. Zhong, Y. Pandya, and R. D. Joerger. 2002. 16S
chicken gastrointestinal tract: influence on health, productivity rRNA-based analysis of microbiota from the cecum of broiler
and disease. Appl Microbiol Biotechnol. 98:4301–4310. chickens. Appl. Environ. Microbiol. 68:124–137.