Content uploaded by Robert Pieper
Author content
All content in this area was uploaded by Robert Pieper on Mar 07, 2016
Content may be subject to copyright.
ORIGINAL ARTICLE
Influence of medium-chain fatty acids and short-chain organic
acids on jejunal morphology and intra-epithelial immune cells in
weaned piglets
F. Ferrara
1,2
, L. Tedin
1
, R. Pieper
1
, W. Meyer
3
and J. Zentek
1
1 Institute of Animal Nutrition, Department of Veterinary Medicine, Freie Universit€
at Berlin Berlin, Germany
2 Institute of Vegetative Physiology, Charit
e University Hospital and Center of Cardiovascular Research Berlin, Germany, and
3 Institute of Anatomy, Foundation, University of Veterinary Medicine Hannover, Germany
Summary
Medium-chain fatty acids (MCFA) and short-chain organic acids (SOA) are often used as feed additives in
piglet diets. There are limited studies in pigs describing the impact of MCFA or SOA on gut morphology and
the local immune system. The aim of this study was to investigate whether the supplementation of SOA
(0.41% fumaric acid and 0.32% lactic acid), or the combination of SOA with MCFA (0.15% caprylic and
capric acid) would have effects on gut morphology and intestinal immune cells in weaned piglets. A total
number of 72 weaned piglets were randomly allocated into three experimental groups. Tissue samples of six
animals per group were used to investigate the potential impact of the feed additives on villus length and
crypt depth of the jejunum and to quantify intra-epithelial lymphocytes (IEL). CD3-positive IEL were deter-
mined via immunohistochemistry (IHC) and flow cytometry (FC), whereas CD2, CD5, CD8b, CD16
and cd TCR-positive IEL were only analysed by FC. The supplementation of MCFA and SOA did not signifi-
cantly affect morphometric data. The FC data indicated that SOA significantly increased the quantity of
CD2
CD8
cd T cells in the jejunum epithelium. Both IHC and FC analyses of pig jejunum confirmed that
the majority of IEL expressed the surface marker CD3 and could be classified as cytotoxic T lymphocytes. In
conclusion, the data indicated that SOA increased the proportion of CD2
CD8
cd T cells in the jejunal
epithelium. Thus, SOA might enable a beneficial effect on the local immunity by increasing the constitutive
number of potential effector cells to defeat infectious diseases.
Keywords piglets, small intestinal morphology, local immunity, immunohistochemistry, flow cytometry
Correspondence F. Ferrara, Institute of Vegetative Physiology, Charit
e University Hospital, Hessische Straße 3-4, 10115 Berlin, Germany.
Tel: +49 30 528 447; Fax: + 49 30 528 972; E-mail: fabienne.ferrara@charite.de
Received: 21 April 2015; accepted: 26 January 2016
Introduction
Weaning is a challenge for the piglet including separa-
tion from the mother, environmental and nutritional
changes. This may set the scene for digestive disorders,
reduced performance and even increased mortality
(Campbell et al., 2013). Feed composition is an impor-
tant driving factor for feed intake, energy and nutrient
supply to enterocytes and subsequently affects epithe-
lial structure and integrity (Le Dividich and S
eve,
2000; Lall
es et al., 2004, 2007; Zabielski et al., 2008).
In addition, nutritional factors and changes in the gut
microbial ecosystem may affect the local immune
response in the piglet gut during the weaning period
(Le Dividich and S
eve, 2000; Lall
es et al., 2004, 2007;
Zabielski et al., 2008). Transient changes in intestinal
morphology and function (e.g. reduction of villus
length and enzyme activity) within the first 5 days
following weaning may therefore result from the com-
plex interaction between diet, intestinal microbiota
and the local immune system (Pluske et al., 1997;
Stokes et al., 2004). The mucosa-associated immune
system is an important factor determining the balance
between tolerance against food antigens or autochtho-
nous intestinal bacteria or local defence reactions
against putative pathogens or other antigens. Local
intestinal immune response is mainly driven by com-
plex interactions between innate and acquired mecha-
nisms of the intestinal associated local immune system
(Pabst and Rothk€
otter, 1999; Vega-Lopez et al., 2001;
Stokes et al., 2004). Intra-epithelial lymphocytes
(IEL) as part of the gut-associated immune system are
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH 1
DOI: 10.1111/jpn.12490
located between the enterocytes, and in pigs, they are
relatively equally distributed along the villus axis
(Vega-Lopez et al., 2001). Based on the close proxim-
ity to the intestinal lumen, it is hypothesized that IEL
may play an important role within the local immune
defence (Pabst, 1987). Maintaining the intestinal
physiological and intestinal barrier function in wean-
ing piglets is of high importance for the prevention of
digestive disorders.
Medium-chain fatty acids (MCFA) and short-chain
organic acids (SOA) constitute rapidly available
energy sources for intestinal and extra-intestinal tis-
sues and could be useful particularly in feeding
weaned piglets (Lueck, 1980; Borum, 1992; Heo et al.,
2002). MCFA and SOA can be utilized by enterocytes
as energy source and can attenuate negative effects of
weaning on villus length and crypt depth in piglets
(Dierick et al., 2002a,b, 2003; Lee et al., 2007). Addi-
tionally, previous studies suggest that MCFA or SOA
can modulate the number of intra-epithelial lympho-
cytes (IEL), the lymphocyte proliferation rate and
jejunal cytokine expression (Dierick et al., 2002a,b,
2003; Lee et al., 2007; Kuang et al., 2015).
The focus of this study was to investigate the effect
of SOA and the combination of SOA with MCFA on
gut morphology and IEL populations in weaned pig-
lets, with focus on synergistic effects of both feed addi-
tives.
Material and methods
This study was performed in accordance with the Ger-
man ethical and animal care guidelines and approved
by the State Office of Health and Social Affairs ‘Lan-
desamt f€
ur Gesundheit und Soziales Berlin’ (LAGeSo)
(registration number G0293/09).
Animals, housing and diets
A total number of 72 castrated piglets (Duroc x Pi
e-
train) (weaned at 25 1 days of age) were used in
three consecutive experimental feeding trials to evalu-
ate the influence of the feed additives on performance
parameters. In each trial, n=24 piglets were ran-
domly allocated into three experimental groups with
eight animals each and two piglets per flat deck pen
resulting in a total of n=72 piglets. Performance data
have been described in a previous publication (Zentek
et al., 2013). The room temperature was 28 °C at the
beginning of the feeding trial and was gradually
decreased to 22 °C within 10 days. The humidity was
65%. A light programme provided a 16-h light and
8-h dark period. Dietary treatments were as follows
(Table 1): a commercial starter diet (CON), the starter
diet containing 1.05% of a SOA (with 39.8% fumaric
acid, 31.% lactic acid and silicium dioxide as a carrier,
SOA), and the starter diet containing SOA and 0.3%
of a MCFA premix (with 50% of caprylic and capric
acid).
Sampling
After a 28-days experimental feeding trail, three pig-
lets per group and per consecutive period, respec-
tively, were euthanized by intracardial application of
a tetracainhydrochloride, mebezoniumiodide and
embutramide (T61
â
; Intervet, Unterschleissheim,
Germany), after a deep anaesthesia with a combina-
tion of ketamine hydrochloride (Ursotamin
â
, 10%;
Serumwerk Bernburg AG, Bernburg, Germany) and
azaperon (Stresnil
â
, Jansen-Cilag, Neuss, Germany)
(1/0.2 mg per kg BM). In total, six animals (two
animals each feeding trail) per group were used to
investigate potential effects of the feed additives on
Table 1 Analysed and calculated nutritional composition of the piglet
diet (as-fed basis)*
Item CON SOA SOA+MCFA
Analysed
Dry matter, g/kg 888.8 887.5 888.6
Crude ash. g/kg 62.9 64.5 67.3
Crude protein, g/kg 207.3 202.6 204.0
Crude fat, g/kg 61.2 59.1 62.3
Crude fibre, g/kg 46.5 49.5 50.3
Ca, g/kg 9.7 9.6 9.8
P, g/kg 5.5 5.5 5.5
Na, g/kg 2.6 2.8 2.9
K, g/kg 8.1 8.1 8.0
Mg, g/kg 1.6 1.6 1.6
Fe, mg/kg 451 445 467
Zn, mg/kg 142 130 128
Cu, mg/kg 114 123 126
Buffering capacity, mEq/kg 545 510 495
pH 4.69 4.41 4.38
L-lactic acid, g/kg 0.45 3.97 3.74
D-lactic acid, g/kg 0.09 0.13 0.11
Fumaric acid, g/kg <0.05 <0.05 <0.05
Caprylic acid, g/kg 0.09 0.08 0.52
Capric acid, g/kg 0.06 0.08 0.52
Calculated
Lys, g/kg 12.5 12.5 12.5
Met, g/kg 4.69 4.69 4.69
Thr, g/kg 8.45 8.45 8.45
Trp, g/kg 2.63 2.63 2.63
ME, MJ/kg 13.8 13.8 13.8
*or +OA, without or with organic acid; or +MCFA, without or with
medium-chain fatty acid.
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH2
MCFA and SOA in piglets F. Ferrara et al.
gut morphology and intra-epithelial cells (IEL). Tissue
samples were taken at approximately 50% of the
entire length of the jejunum avoiding taking tissue
with discrete or parts of the continuous payers patch.
Histological preparation
A first jejunum tissue sample with the length of 5 cm
was removed immediately post-mortem and sliced on
the antimesenterial side, stretched on cork plates and
washed with phosphate-buffered saline (PBS) at room
temperature. The tissue sample was then quickly
transferred (mucosal surface facing downwards) into
the fixation medium 4% neutral buffered formalin
solution (37% formol +29PBS in aqua dest.). In
general, all samples were embedded for histological
examinations according to Hornickel (2009), with
minor modifications, that is during the last step,
incubation of tissue was performed with xylol-I for
30 min (combination of 1:1 xylol and 96% alcohol),
followed by three times washing with pure xylol and
transfer into warm paraffin (60 °C) overnight (12–
14 h).
Morphometric analyses
Haematoxylin–eosin staining was performed by a
standard protocol (Hornickel, 2009). Well-orientated
villus length and crypt depth were measured by an
investigator blinded to outcome with a standard light
microscope (Photomicroscope III; Zeiss, Jena, Ger-
many) with a digital microscope camera (DP 72;
Olympus, Hamburg, Germany)and the related analy-
sis software (Cell Sense Software; Olympus). Morpho-
metric analyses were performed on four slices from
each animal. Villus length was defined as a line
between the villus top and base, whereas crypt depth
was defined as a conjunction line between the open-
ing of the crypt and their base. Per slice, five villi and
crypts were measured.
Immunohistochemical analysis
To detect CD3-positive IELs by immunohistochem-
istry (IHC), the monoclonal antibody PPT3 (Table 2)
was applied to the 5 lm slices after antigen retrieval
within 30 min in citrate buffer (96–98 °C). To avoid
non-specific binding of the Fc part of the respective
PPT3 primary antibody, an isotype control was con-
ducted with a non-specific antibody (DAKO, Golstrup,
Denmark), at which the isotype control was applied in
the same protein concentration as the primary anti-
body (Fig. 1). Both antibodies were diluted with 1%
bovine serum albumin (BSA) and PBS. In summary,
an indirect IHC method was performed. To visualize
the primary antibody, a two-step indirect method was
used by the EnVision
â
mouse system (EnVision+Sys-
tem-HRP, mouse K4007, Dako, Golstrup, Denmark)
whereas the secondary antibody was conjugated with
a horseradish peroxidase (HRP)-labelled dextran poly-
mer (goat anti-mouse IgG
1
). In general, the IHC pro-
tocol was performed according to a general procedure
which was published previously (Hornickel, 2009).
For the evaluation of the counting accuracy, a double-
blinded quantification of CD3-positive IEL was per-
formed only on complete and intact villi (four slices
per animal, five villi per slice) and expressed among
100 enterocytes.
Flow cytometric analysis
For the later phenotypic analysis of the IELs via flow
cytometry (FC), an additional 20 cm of the small intes-
tine was taken, adjacent to the jejunal sample that was
chosen for the IHC. The monoclonal primary and fluo-
rescence-labelled secondary antibodies used to identify
IEL populations are listed in Tables 2 and 3.
Cell isolation
The isolation of the cells from the middle of jejunum
tissue sample was performed as described previously
(Mafamane et al., 2011). Briefly: tissues were imme-
diately transferred into fresh PBS solution and
everted to expose the intestinal mucosa. Following
that, tissues were incubated in Hanks0balanced salt
solution (HBSS)-DTT (dithiothreitol) medium and
(5 min and 37 °C) and HBSS-EDTA medium. Cell
suspensions were filtered through a sterile synthetic
gauze (210-mm nylon mesh) and centrifugated at
600 gfor 10 min at 4 °C. Cell pellets were resus-
pended with approximately 10 ml of Roswell Park
Memorial Institute Medium No. 1640 (RPMI) med-
ium and incubated with 50 ll DNase for 5 min at
37 °C and centrifuged (same conditions as described
above). The obtained cell pellet was mixed with Per-
coll (25%) solution and centrifuged (600 g, 20 min).
Finally, the cell pellet was resuspended with RPMI
medium and centrifuged (300 g,4°C, for 10 min)
and suspended with lysis buffer (ammoniumchloride
solution).
Flow cytometry
The staining of the cell preparation for the unla-
belled primary antibodies CD3, CD2, CD8band
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH 3
F. Ferrara et al. MCFA and SOA in piglets
Po-TcR1-N4 (Table 2) was performed as previously
published (Mafamane et al., 2011). For every reac-
tion, 1 910
6
cells were solved to saturating concen-
trations of the antibody in a volume of 100 llof
PBS and 0.2% BSA for 20 min on ice. Cells were
incubated with the fluorescence-labelled antibody
(Table 3). Finally, cells were analysed by the
FACSCalibur flow cytometer (Becton Dickinson Bio-
science, San Jos
e, CA, USA), using the computer
software CellQuest pro
â
(Becton Dickinson Bio-
science), by evaluating the relative distribution of
the diverse IEL phenotypes in the middle of jeju-
num. Per cent values were calculated within the
lymphocyte gate (=100%).
Additionally, the relation of IEL phenotypes among
100 enterocytes was calculated using the counted
number of CD3-positive IEL using the following calcu-
lation formula:
Number of CD3þcells=100enterocytes
ð% of population measured per FC=% of CD3þIEL in FCÞ
Statistical analyses
The statistical evaluation was performed by SPSS 19.0
(SPSS Inc., Chicago IL, USA). The normal distribution
of the data was tested with the Kolmogorov–Smirnov
test including Lilliefors test. Normally distributed data
were expressed as mean and standard deviation; non-
normally distributed data are expressed as median,
minimum and maximum (villus length and crypt
depth). Single variance analyses were performed on
normally distributed data, followed by the LSD post
hoc test in case of p <0.05. The Kruskal–Wallis test
was performed on non-normally distributed data
followed by Wilcoxon–Mann–Whitney test in case of
(a) (b)
(c) (d)
Fig. 1 Brown-stained CD3-positive IEL in the mid-jejunum (control group); LM, 9400; (a, b) isotype control of primary antibody PPT3 without (a) and
with counter staining (b); (c, d) antigen retrieval with citrate buffer (30 min, 96–98 °C) without (c) and with counter staining (d).
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH4
MCFA and SOA in piglets F. Ferrara et al.
significance to determine the differences between
groups. A p-value of ≤0.05 was considered significant
for all data, whereas a p-value of ≤0.1 was considered
as tendency. p-values which are listed in all tables
based on a single variance analyses (normally dis-
tributed data) or Kruskal–Wallis test (non-normally
distributed data).
Results
Histological and Immunohistological measurements
Villus length and crypt depth in the mid-jejunum
were not significantly influenced by dietary treat-
ment. However, a trend towards longer villi was
determined with the SOA+MCFA group as compared
to the control (Table 4).
Table 4 Immunohistochemical counts of CD3
+
-
IEL and villus length and crypt depth in the mid-
dle of jejunum
Group
CON SOA SOA+MCFA p-Value
CD3
+
-IEL*60.4 6.6 61.5 4.7 58.7 7.7 0.549
Villus length†400 (230–500) 410 (300–530) 420 (260–540) 0.098
Crypt depth‡130 (70–260) 140 (60–300) 140 (80–290) 0.600
CON, control; SOA, short-chain organic acids (fumaric and lactic acid); MCFA, medium-chain fatty
acids (caprylic acid and capric acid); CD, cluster of differentiation; IEL, intra-epithelial lymphocytes;
p, probability value from an analysis of variance (CD3 measurement) or Kruskal–Wallis test (villus
length and crypt depth).
*Mean of CD3-positive IELs determined from individual data from both microscopy counts,
expressed among 100 enterocytes; results of measured mid-jejunum.
†Villus length (lm).
‡Crypt depth (lm): median (minimum–maximum).
Table 2 Primary antibodies to differentiate IEL
in the small intestine in piglets Primary
Antibody Isotype Specificity Dilution Method Vendor
PPT3* Mouse
IgG
1
K
Porcine CD3e1:100
1:200
IHC
FC
SouthernBiotech.,
Birmingham,
AL, USA
Anti-mouse
IgG
1
†
Mouse
IgG
1k
Non 1:100 IHC Dako, Golstrup,
Denmark
CD2 Mouse
IgG
2a
Porcine
CD2
1:200 FC VMRD Inc.
Pullman,
WA, USA
PGBL22A‡Mouse
IgG
1
Porcine
TcR1-N4
1:100 FC VMRD Inc.
PG164A Mouse
IgG
2a
Porcine CD8b1:100 FC VMRD Inc.
G7 (CD16) Mouse
IgG
1
Porcine CD16 1:100 FC Acris Antibodies
Herford Germany
CD5 Mouse
IgG
1
Porcine CD5 1:100 FC VMRD Inc.
IHC, immunohistochemistry; FC, flow cytometry.
*Primary antibody was used for IHC and FC.
†Isotype control.
‡Primary antibody for the detection of cd
+
T cells, binds on d-chain; all listed antibodies are mono-
clonal.
Table 3 Secondary antibodies to differentiate IEL by FC
Secondary antibody* Primary antibody Dilution Stain
Goat Anti-mice PPT3 1:100 FITC
IgG
1
PGBL22A 1:200 PE
G7 (CD16) 1:200 PE
CD5 1:200 PE
Goat Anti-mice CD2 1:100 FITC
IgG
2a
CD8b1:100 FITC
FITC, fluoresceine isothiocyanate; PE, phycoerythrin.
*All secondary antibodies were produced by SouthernBiotech.
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH 5
F. Ferrara et al. MCFA and SOA in piglets
Similarly, no significant differences for CD3-positive
IELs were observed by IHC in jejunal tissue samples
(Table 4).
Characterization of IELs by flow cytometry
The majority of IEL expressed the surface marker CD3
(Tables 5 and 6). The majority of CD3-positive cells
could be classified based on the data obtained by FC as
cytotoxic T lymphocytes (CD3
+
CD8
+ab
). There were
no significant differences in CD3 cell count between
CON and the treatment groups or within the
treatment groups (Tables 5 and 6). Additionally, a low
number of (CD2
CD8
)cd T cells could be detected
by FC (Tables 5 and 6). The data indicate a significant
(p =0.027) increase of the quantity of (CD2
CD8
)
cd T cells in the epithelium in the jejunum in the
piglets from group SOA.
Discussion
The hypothesis of this study was that SOA and MCFA
would affect gut morphology and the local intestinal
immune system in weaned piglets due to the effects
on the intestinal microbiota and their role as energy
source for the epithelial cells. Animal performance
data and changes in gastrointestinal microbial ecology
have been described previously (Zentek et al., 2013).
Structural and functional changes of the small intes-
tine have been often characterized by a reduction of
villus length and increased crypt depth and enzyme
activity, or higher intestinal permeability, thus indi-
cating disturbed mucosa integrity during the first 3–
5 days post-weaning (Boudry et al., 2004; Montagne
et al., 2007; Smith et al., 2010). This has been shown
to predispose piglets to gastrointestinal diseases and is
associated with lower growth rate post-weaning
(Pluske et al., 1997; Zabielski et al., 2008). Weaning
resulted in a decrease of villus length in the jejunum
from approximate 740–390 lm (3 days post-weaning,
15 days of age), whereas the crypt depth increased
from ~180 lmupto210lm (Tang et al., 1999). The
regeneration stage, from days 5–15 post-weaning, is
characterized by the progressive reconstruction of
physiological structures and digestive capacity (Tang
et al., 1999; Spreeuwenberg et al., 2001; Montagne
et al., 2007). Within the first 34 days of age, villus
length increased to 400–500 lm (Tang et al., 1999).
Generally, the obtained data regarding villus length in
this study are in good agreement with previous studies
in weaned piglets (Tang et al., 1999; Montagne et al.,
2007; Martin et al., 2012), at which villus length was
in a range between 400–420 lm and crypt depth
between 130–140 lmat53(1) days of age. How-
ever, the assumption that dietary supply with SOA or
SOA+MCFA would affect jejunal morphology could
be confirmed only in part because no significant dif-
ferences between the dietary treatments were deter-
mined. The slightly longer villi without differences in
crypt depth in the SOA+MCFA group may indicate a
reduced rate of apoptosis at the tops of the villi. Simi-
larly, the administration of medium-chain triglyc-
erides from Cuphea seed (C. lancelota und C. ignea)in
combination with lipase in the diet of weaned piglets
has been shown to lead to a significant increase in vil-
lus length at the distal small intestine and a decrease
of depth of the crypts (Dierick et al., 2003). In con-
trast, the administration of formic acid resulted in
Table 5 Phenotypes of IEL analysed by FC in the mid-jejunum of the pig-
lets (data as % of total epithelial lymphocytes; mean and standard devia-
tion)
Group
CON SOA SOA+MCFA p-Value
CD3
+
72.0 8.7 82.2 6.9 77.9 8.1 0.117
CD8b
+
46.34 8.3 54.0 8.4 51.5 13.2 0.318
CD2
+
/CD5
33.3 9.6 23.2 9.2 27.6 7.6 0.081
CD16
+
28.7 8.9 27.5 6.9 28.4 5.9 0.905
CD2
+
/cd
+
22.6 4.4 20.7 5.5 19.0 6.3 0.440
CD2
/cd
+
0.6 0.3
b
1.1 0.5
a
0.6 0.3
b
0.027
CON, control group; SOA, short-chain organic acids (fumaric and lactic
acid); MCFA, medium-chain fatty acids (caprylic acid and capric acid);
IEL, intra-epithelial lymphocytes; CD, cluster of differentiation; FC, flow
cytometry; p, probability value from analysis of variance.
a,b
Different indices within a line shows significant differences (p ≤0.05).
Table 6 The relation of IEL phenotypes among 100 enterocytes, calcu-
lated from the counted number of CD3-positive IEL and FC data (mean
and standard deviation)
Group
CON SOA SOA+MCFA p
CD3
+
-IEL*60.4 6.3 60.3 5.7 58.0 8.0 0.867
CD8b
+
-IEL 36.2 7.8 41.4 5.1 42.3 6.8 0.434
CD2
+
/CD5
-IEL 28.0 7.6 15.4 9.0 20.3 9.5 0.211
CD16
+
-IEL 25.0 4.9 17.8 6.7 20.70 7.6 0.377
CD2
+
/cd
+
-IEL 17.8 3.7 12.4 4.4 11.4 5.6 0.176
CD2
/cd
+
-IEL 0.50 0.3 0.70 0.3 0.4 0.2 0.406
CON, control group; SOA, short-chain organic acids (fumaric and lactic
acid); MCFA, medium-chain fatty acids (caprylic acid and capric acid););
IEL, intra-epithelial Lymphocytes; CD, cluster of differentiation; FC, flow
cytometry;IHC, immunohistochemistry; p, probability value from analy-
sis of variance.
*Mean of CD3-positive IELs determined from combining individual data
from both microscopy counts, expressed per 100 enterocytes.
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH6
MCFA and SOA in piglets F. Ferrara et al.
shorter jejunal villi of early weaned piglets (Man-
zanilla et al., 2004). Consequently, our findings sug-
gest that there might be synergistic positive effects of
combined feeding with MCFA+SOA in weaned pig-
lets. The performance of the piglets was not affected
by MCFA or SOA (Zentek et al., 2013), thus indicat-
ing a more related effect of MCFA directly towards
enterocytes. Consequently, the mechanisms and pos-
sible triggers driving small intestinal cell proliferation
and apoptosis under the influence of SOA and MCFA
are of interest and require further investigations.
In addition, epithelial changes in response to SOA
and MCFA during the early post-weaning period
might be even more important to assess small intesti-
nal functionality during this time.
For the assessment of the impact of both additives
on mucosal immune function, morphometric mea-
surements were combined with quantification of IELs
by IHC and FC. The identical PPT3 antibody was used
to allow the evaluation of results from IHC and FC in
two adjacent intestinal segments. In general, the
number of CD3-positive IELs was not significantly dif-
ferent between the treatment groups. Obviously, the
percentage of CD3-positive IELs analysed by FC was
higher than determined with IHC. This could be either
due to methodological effects, for example different
thicknesses of paraffin blocks or counting procedures
within the IHC method, or the loss of epithelial cells
during the tissue preparation for FC analyses. The
later was emphasized by the decreased total number
of calculated IEL by FC among 100 enterocytes based
on labelled CD3
+
IELs by IHC. However, false-positive
results of IHC were excluded by isotype control anti-
body of the primary antibody, and additionally, to
evaluate cell counting accuracy, a double-blinded trial
was performed. Finally, based on our implementation
of FC preparation, it is not possible to recommend one
of these used methods, for that purpose, it is necessary
to use a validate step, for example histological exami-
nation of the jejunal tissue sample of the number of
left epithelial cells following cell isolation.
There are limited studies in pigs describing the
mode of action of MCFA or SOA on IEL populations
in piglets. A significant decrease in the number of IEL
in the small intestine was observed in piglets fed
Cuphea seeds (Dierick et al., 2003). In the present
study, the number of CD3
+
IELs among 100 entero-
cytes did not differ between treatment groups. In the
literature, there is a controversal discussion regarding
the impact of IEL number as an indicator of mucosal
integrity. A reduced IEL number combined with
longer villi and decreased crypt depth, based on
reduced apoptotic rate, was considered as protective
effect of MCFA (Dierick et al., 2003). On the other
hand, a lower number of intestinal IEL may indicate a
downregulation of local immune responses, due to
the reduced number of effector cells (Vega-Lopez
et al., 1995; Rothk€
otter et al., 1999; Gu et al., 2002).
In fact, additional investigations could be useful for
further descriptions of potential effects of MCFA and
SOA on the intestinal local immunity, as indicated by
Kuang et al. (2015), at which both additives downreg-
ulated systemic and jejunal pro-inflammatory cyto-
kine expression (TNF-a) and stimulated the tissue
expression of the regulatory cytokine TGF-b.
In the adult pig, it has been shown that over 50% of
the total T cells in the small intestine were located at
the epithelium (Vega-Lopez et al., 2001). The highest
proportion of IELs is located close to the basement
membrane (53%), whereas most of these cells bear
the CD2
+
CD4
CD8
+
phenotype (Vega-Lopez et al.,
2001). A limited number of double-negative
CD2
+
CD4
CD8
cells are located on the same level of
the enterocyte cell nucleus (Vega-Lopez et al., 1993,
2001; Rothk€
otter et al., 1994). CD4
+
, CD16
+
and
TcR1
+
cd cells have been found at low density in the
epithelium (Vega-Lopez et al., 1993; Solano-Aguilar
et al., 2001). In general, T lymphocytes of the pigs can
be differentiated by their T-cell receptor into ab T cells
and cd T cells. The first group can be furthermore
divided into four different subtypes (Yang and Park-
house, 1996; Charerntantanakul and Roth, 2007):
native T-helper cells (CD4
+
CD8
cells), T memory
cells (CD4
+
CD8
+aalow
) and cytotoxic T cells
(CD4
CD8
+ablow
and CD4
CD8
+abhigh
cells). CD8
+
cells are a predominant T-cell subset in antiviral
responses, whereas CD4
+
T cells predominantly occur
during bacterial and parasitic infections (Charerntan-
tanakul and Roth, 2007).
To achieve a wide differentiation of IELs, an addi-
tional jejunum tissue sample was further investigated
by FC. The FC data demonstrated that the majority of
IELs expressed the surface marker CD3. Furthermore,
CD3 and CD5 are not expressed at the surface of natu-
ral killer (NK) cells and may be useful to separate
CD8-positive cd T lymphocytes from NK cells
(Saalm€
uller et al., 1994). In summary, the FC data
indicated that SOA significantly increased (p =0.027)
the quantity of CD2
CD8
cd T cells in the epithelium
in the jejunum. Phenotypically, the total porcine cd T
cells are divided into three subsets based on the
surface expression of CD2 and CD8
aa
(Wen et al.,
2012). The non-classical group of cd T cells
(CD2
CD4
CD8
) represents the main proportion of
T cells in young pigs (Yang and Parkhouse, 1996;
Stepanova and Sinkora, 2013) with 15–26% (4 weeks
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH 7
F. Ferrara et al. MCFA and SOA in piglets
of age) and 24–39% (4 month old piglets) (Yang and
Parkhouse, 1996). The number of non-classical cd T
cells in the blood decrease with the age of 8 months
(Yang and Parkhouse, 1996). All functions of
non-classical cd T cells are until now not fully
understood (Charerntantanakul and Roth, 2007;
Stepanova and Sinkora, 2013). However, the high
number of non-classical group of cd T cells is in
contrast to humans and mice, thus indicating a por-
cine specific lineage (Stepanova and Sinkora, 2013).
In fact, a fraction of non-classical cd T cells has the
ability to downregulate TCRcd within hours in
spleen and thymus and additionally to upregulate
their receptor only in thymus (Stepanova and Sin-
kora, 2013). The proliferation rate of non-classical T
cells in spleen is very low, compared to those of
classical cd T cells (CD2
+
CD4
CD8
,
CD2
+
CD4
CD8
+aa
) (Stepanova and Sinkora, 2013).
Equally to the CD2
+
CD4
CD8
cd T subset (classical
cd T cells), the non-classical cd T cells have mainly
pro-inflammatory function as visible by directly
secreting of IFN-cor by promoting CD4
+
ab T-cell
proliferation (Wen et al., 2012). In contrast to the
examinations of Stepanova and Sinkora (2013), the
study of Wen et al. (2012) revealed that the CD8
subsets can differentiate into CD8
+
by acquiring
CD8 expression in ileum and blood following exper-
imental virus infection, at which the frequencies of
the non-classical cd T cells were reduced in these
compartments. According to the later study, our
results may be indicating a beneficial effect of SOA
by increasing a constitutive higher number of non-
classical T cells to defend pathogen infection, which
appears to be an additional beneficial effect within
the context of prevention of post-weaning (Zentek
et al., 2013).
The cellular processes of antigen identification by cd
T cells appear to be assumed: Unlike ab T cells, these
cells are known to recognize foreign antigens in a
non-MHC-restricted manner directly by their receptor
or like unspecific immune cells with the aid of patho-
gen-associated patterns (PAMPs), by Toll-like recep-
tors, or scavenger receptors (Hedges et al., 2005;
Charerntantanakul and Roth, 2007; Wen et al.,
2012). However, the results of our study indicated
that cytotoxic ab T cells, cd T cells and NK cells repre-
sent a large proportion of IEL; this may indicate an
important role of IEL in local gut immunity, at which
they play important roles in antigen recognition and
immune response (Wen et al., 2012).
Conclusion
In conclusion, the supplementation of SOA or SOA
+MCFA to a starter diet for piglets did not affect sig-
nificantly jejunal morphology. However, a tendency
towards higher villi without changes in crypt depth
may indicate some effects on epithelial cell turnover,
which requires further investigation. Using IHC and
FC revealed that the majority of IEL expressed the
surface marker CD3 and could be classified as cyto-
toxic T lymphocytes without significant differences
between treatments. In summary, feeding SOA
increased the quantity of non-classical cd T cells
which may confer protection against putative patho-
gens or antigens through a higher abundance of
effector cells.
Acknowledgement
The authors wish to thank Mrs. Petra Huck for her
technical assistance with flow cytometry analyses,
Mrs. Karin Briest-Forch for her technical assistance
with histological preparations and Damian McLeod
for help in preparing the manuscript.
References
Borum, P. R., 1992: Medium-chain
triglycerides in formula for preterm neo-
nates: implications for hepatic and
extrahepatic metabolism. Journal of Pedi-
atrics 120, 139–145.
Boudry, G.; P
eron, V.; Hu€
erou- Luron,
I.; Lall
es, J. P.; S
eve, B., 2004:
Weaning induces both transient and
long- lasting modifications of absorp-
tive, and barrier properties of piglet
intestine. Journal of Nutrition 134,
2256–2262.
Campbell, J. M.; Crenshaw, J. D.; Polo, J.,
2013: The biological stress of early
weaned piglets. Journal of Animal Science
and Biotechnology 4,1–4.
Charerntantanakul, W.; Roth, J. A., 2007:
Biology of porcine T lymphocytes.
Animal Health Research Reviews 7,
81–92.
Dierick, N. A.; Decuypere, J. A.; Molly,
K.; van Beek, E.; Vanderbeke, E.,
2002a: The combined use of triacyl-
glycerols containing medium-chain
fatty acids (MCFAs) and exogenous
lipolytic enzymes as an alternative for
nutritional antibiotics in piglet nutri-
tion I. In vitro screening of the release
of MCFAs from selected fat sources by
selected exogenous lipolytic enzymes
under simulated pig gastric conditions
and their effects on the gut flora of
piglets. Livestock Production Science 75,
129–142.
Dierick, N. A.; Decuypere, J. A.; Molly, K.;
van Beek, E.; Vanderbeke, E., 2002b: The
combined use of triacylglycerols (TAGs)
containing medium chain fatty acids
(MCFAs) and exogenous lipolytic enzymes
as an alternative to nutritional antibiotics
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH8
MCFA and SOA in piglets F. Ferrara et al.
in piglet nutrition. II. In vivo release of
MCFAs in gastric cannulated and slaugh-
tered piglets by endogenous and exoge-
nous lipases; effects on the luminal gut
flora and growth performance. Livestock
Production Science 76,1–16.
Dierick, N. A.; Decuypere, J. A.; Degeyter,
I., 2003: The combined use of whole
Cuphea seeds containing medium chain
fatty acids and an exogenous lipase in
piglet nutrition. Archiv f€
ur Tieren€
ahrung
57,49–63.
Gu, X.; Li, D.; She, R., 2002: Effect of
weaning on small intestinal structure
and function in the piglet. Archives of
Animal Nutrition 56, 275–286.
Hedges, J. F.; Lubick, K. J.; Jutila, M. A.,
2005: cd T cells respond directly to
pathogen-associated molecular patterns.
Journal of Immunology 174, 6045–6053.
Heo, K. N.; Lin, X.; Han, I. K.; Odle, J.,
2002: Medium-chain fatty acids but not
L-carnitine accelerate the kinetics of
14C triacylglycerol utilization by colos-
trum-deprived newborn pigs. Journal of
Nutrition 132, 1989–1994.
Hornickel, N., 2009: Investigations on the
innate immunity of the esophagus
epithelium of domesticated mammals.
Thesis, Tier€
arztliche Hochschule Han-
nover, Hannover
Kuang, Y.; Wang, Y.; Zhang, Y.; Song, Y.;
Zhang, X.; Lin, Y.; Che, L.; Xu, S.; Wu,
D.; Xue, B.; Fang, Z., 2015: Effects of
dietary combinations of organic acids
and medium chain fatty acids as a
replacement of zinc oxide on growth,
digestibility and immunity of weaned
pigs. Animal Feeding Science and Technol-
ogy 208, 145–157.
Lall
es, J. P.; Boudry, G.; Favier, C.; Le Floc
̓
h, N.; Luron, I.; Montagne, L.; Oswald,
I.; Pie
́, S.; Piel, C.; Sève, B., 2004: Gut
function and dysfunction in young pigs.
Animal Research 53, 301–316.
Lall
es, J. P.; Bosi, P.; Smidt, H.; Stokes, C.
R., 2007: Nutritional management of
gut health in pigs around weaning. Pro-
ceeding of the Nutrition Society 66, 260–
268.
Le Dividich, J.; S
eve, B., 2000: Effects of
underfeeding during the weaning per-
iod on growth, metabolism, and hor-
monal adjustments in the piglet.
Domestic Animal Endocrinology 19,63–74.
Lee, D. N.; Liu, R.; Chen, Y. T.; Wang, R.
C.; Lin, S. Y.; Weng, C. F., 2007: Effects
of diets supplemented with organic acids
and nucleotides on growth, immune
response and digestive tract develop-
ment in weaned pigs. Journal of Animal
Physiology and Animal Nutrition 91, 508–
518.
Lueck, E., 1980: Antimicrobial Food Addi-
tives. Springer-Verlag, Berlin Heidelberg,
New York.
Mafamane, H.; Szabo, I.; Schmidt, F. G.;
Filter, M.; Walk, N.; Tedin, K.; Scharek-
Tedin, L., 2011: Studies on the effect of
an Enterococcus faecium probiotic on T cell
populations in peripheral blood and
intestinal epithelium and on the suscep-
tibility to Salmonella during a challenge
infection with Salmonella Typhimurium
in piglets. Archives of Animal Nutrition 65,
415–430.
Manzanilla, E. G.; Perez, J. F.; Martin, M.;
Kamel, C.; Baucells, F.; Gasa, J., 2004:
Effect of plant extracts and formic acid
on the intestinal equilibrium of early-
weaned pigs. Journal of Animal Science
82, 3210–3218.
Martin, L.; Pieper, R.; Kr€
oger, S.; Goodarzi
Boroojeni, F.; Vahjen, W.; Neumann,
K.; Van Kessel, A. G.; Zentek, J., 2012:
Influence of age and Enterococcus fae-
cium NCIMB 10415 on development of
small intestinal digestive physiology in
piglets. Animal Feed Science and Technol-
ogy 175,65–75.
Montagne, L.; Boudry, G.; Favier, C.; Le
Hu€
erou- Luron, I.; Lall
es, J. P., 2007:
Main intestinal markers associated with
the changes in gut architecture and
function in piglets after weaning. British
Journal of Nutrition 97,45–57.
Pabst, R., 1987: The anatomical basis for
the immune function of the gut. Anat-
omy and Embryology 176, 135–144.
Pabst, R.; Rothk€
otter, H. J., 1999: Postna-
tal development of lymphocyte subsets
in different compartments of the small
intestine of piglets. Veterinary Immunol-
ogy and Immunopathology 72, 167–173.
Pluske, J. R.; Hampson, D. J.; Williams, I.
H., 1997: Factors influencing the struc-
ture and function of the small intestine
in the weaned pig: a review. Livestock
Production Science 51, 215–236.
Rothk€
otter, H. J.; Kirchhoff, T.; Pabst, R.,
1994: Lymphoid and nonlymphoid cells
in the epithelium and Lamina propria of
the intestinal mucosa in pigs. Gut 35,
1582–1589.
Rothk€
otter, H. J.; M€
ollhof, S.; Pabst, R.,
1999: The influence of age and breeding
conditions on the number and prolifera-
tion of intraepithelial lymphocytes in
pigs. Scandinavian Journal of Immunology
50,31–38.
Saalm€
uller, A.; Hirt, W.; Maurer, S.;
Weiland, E., 1994: Discrimination
between two subsets of porcine CD8
+
cytolytic T lymphoytes by the expres-
sion of CD5 antigen. Immunology 81,
578–583.
Smith, F.; Clark, J. E.; Overman, B. L.;
Tozel, C. C.; Huang, J. H.; Rivier, J. E.;
Blikslager, A. T.; Moeser, A. J., 2010:
Early weaning stress impairs develop-
ment of mucosal barrier function in the
porcine intestine. American Journal of
Physiology Gastrointestinal and Liver Physi-
ology 298, 352–363.
Solano-Aguilar, G. I.; Vengroski, K. G.;
Beshah, E.; Douglass, L. W.; Lunney, J.
K., 2001: Characterization of lympho-
cyte subsets from mucosal tissue in
neonatal swine. Developmental & Compar-
ative Immunology 25, 245–263.
Spreeuwenberg, M. A. M.; Verdonk, J. M.
A. J.; Gaskins, H. R.; Verstegen, M. W.
A., 2001: Small intestine epithelial bar-
rier function is compromised in pigs
with low feed intake at weaning. Journal
of Nutrition 131, 1520–1527.
Stepanova, K.; Sinkora, M., 2013: Porcine
cd T Lympohocytes can be categorized
into two functionally and developmen-
tally distinct Subsets according to
expression of CD2 and level of TCR. The
Journal of Immunology 190, 2111–2120.
Stokes, C. R.; Bailey, M.; Haverson, K.;
Harris, C.; Jones, P.; Inman, C.; Pie, S.;
Oswald, I. P.; Williams, B. A.; Akker-
mans, A. D. L.; Sowa, E.; Rothk€
otter, H.
J.; Miller, B. G., 2004: Postnatal devel-
opment of the intestinal immune system
in piglets: implications for the process of
the weaning. Animal Research 53, 325–
334.
Tang, M.; Laarveld, B.; Van Kessel, A. G.;
Hamilton, D. L.; Estrada, A.; Patience, J.
F., 1999: Effect of segregated early
weaning on postweaning small intesti-
nal development in pigs. Journal of Ani-
mal Science 77, 3191–3200.
Vega-Lopez, M. A.; Bailey, M.; Stevens,
K.; Stokes, C. R., 1993: Immune cell dis-
tribution in the small intestine of the
pig: immunhistological evidence for an
organized compartmentalization in the
lamina propria. Veterinary Immunology
and Immunopathology 37,49–60.
Vega-Lopez, M. A.; Bailey, M.; Telemo, E.;
Stokes, C. R., 1995: Effect of early
weaning on the development of
immune cells in the pig small intestine.
Veterinary Immunology and Immunopathol-
ogy 44, 319–327.
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH 9
F. Ferrara et al. MCFA and SOA in piglets
Vega-Lopez, M. A.; Arenas-Contreras, G.;
Bailey, M.; Gonzales-Pozos, S.; Stokes,
C. R.; Ortega, M. G.; Mondragon-Flores,
R., 2001: Development of intraepithelial
cells in the porcine small intestine.
Developmental & Comparative Immunology
8, 147–158.
Wen, K.; Bui, T.; Li, G.; Liu, F.; Li, Y.;
Kocher, J.; Yuan, L., 2012: Characteriza-
tion of immune modulating functions of
cd T cell subsets in a gnotobiotic pig
model of human rotavirus infection.
Comparative Immunology, Microbiology and
Infectious Diseases 35, 289–301.
Yang, H.; Parkhouse, R. M., 1996:
Phenotypic classification of porcine
lymphocyte subpopulations in blood
and lymphoid tissues. Immunology 89,
45–52.
Zabielski, R.; Godlewski, M. M.; Guil-
loteau, P., 2008: Control of develop-
ment of gastrointestinal system in
neonates. Journal of Physiology and Phar-
macology 59(Suppl. 1), 35–54.
Zentek, J.; Ferrara, F.; Pieper, R.; Tedin,
L.; Meyer, W.; Vahjen, W., 2013:
Effects of dietary combinations of
organic acids and medium chain fatty
acids on the gastrointestinal microbial
ecology and bacterial metabolites in
the digestive tract of weaning piglets.
Journal of Animal Science 93,
3200–3210.
Journal of Animal Physiology and Animal Nutrition ©2016 Blackwell Verlag GmbH10
MCFA and SOA in piglets F. Ferrara et al.
