- Email: [email protected]
Bovine lactoferrin regulates cell survival, apoptosis and inflammation in intestinal epithelial cells and preterm pig intestine
Bovine lactoferrin regulates cell survival, apoptosis and inflammation in intestinal epithelial cells and preterm pig intestine Duc Ninh Nguyen, Pingping Jiang, Allan Stensballe, Emøke Bendixen, Per T. Sangild, Dereck E.W. Chatterton PII: DOI: Reference:
S1874-3919(16)30081-1 doi: 10.1016/j.jprot.2016.03.020 JPROT 2459
To appear in:
Journal of Proteomics
Received date: Revised date: Accepted date:
17 December 2015 14 February 2016 11 March 2016
Please cite this article as: Nguyen Duc Ninh, Jiang Pingping, Stensballe Allan, Bendixen Emøke, Sangild Per T., Chatterton Dereck E.W., Bovine lactoferrin regulates cell survival, apoptosis and inflammation in intestinal epithelial cells and preterm pig intestine, Journal of Proteomics (2016), doi: 10.1016/j.jprot.2016.03.020
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Bovine lactoferrin regulates cell survival, apoptosis and inflammation in
PT
intestinal epithelial cells and preterm pig intestine Duc Ninh Nguyena,b, Pingping Jianga, Allan Stensballec, Emøke Bendixend, Per T. Sangilda, Dereck
Comparative Pediatrics and Nutrition, Department of Veterinary Clinical and Animal Sciences, and b
SC
a
RI
E.W. Chattertona,b*
NU
Department of Food Science, University of Copenhagen, DK-1958, Denmark Department of Health Science and Technology, Aalborg University, DK-9220, Denmark
d
Department of Molecular Biology and Genetics, Aarhus University, DK-8000, Denmark
MA
c
TE
D
* Corresponding author: Dereck E.W. Chatterton, Department of Food Science, and Department of Veterinary Clinical and Animal Sciences, University of Copenhagen, Rolighedsvej 30, Frederiksberg
AC CE P
DK-1958, Denmark. Tel: +45. 35 33 35 90; Fax: +45. 35 33 31 90; Email: [email protected]
Running title: LF regulates the fate of IECs Abbreviations: iTRAQ, isobaric tag for relative and absolute quantitation; LF, lactoferrin; NEC, necrotizing enterocolitis.
1
ACCEPTED MANUSCRIPT
Abstract
PT
Bovine lactoferrin (bLF) may modulate neonatal intestinal inflammation. Previous studies in intestinal epithelial cells (IECs) indicated that moderate bLF doses enhance proliferation whereas high doses
RI
trigger inflammation. To further elucidate cellular mechanisms, we profiled the porcine IEC proteome
SC
after stimulation with bLF at 0, 0.1, 1 and 10 g/L by LC-MS-based proteomics. Key pathways were
NU
analyzed in the intestine of formula-fed preterm pigs with and without supplementation of 10 g/L bLF. Levels of 123 IEC proteins were altered by bLF. Low bLF doses (0.1-1 g/L) up-regulated 11 proteins
MA
associated with glycolysis, energy metabolism and protein synthesis, indicating support of cell survival. In contrast, a high bLF dose (10 g/L) up-regulated three apoptosis-inducing proteins, down-
D
regulated five anti-apoptotic and proliferation-inducing proteins and 15 proteins related to energy and
TE
amino acid metabolism, and altered three proteins enhancing the hypoxia inducible factor-1 (HIF-1)
AC CE P
pathway. In the preterm pig intestine, bLF at 10 g/L decreased villus height/crypt depth ratio and upregulated the Bax/Bcl-2 ratio and HIF-1α, indicating elevated intestinal apoptosis and inflammation. In conclusion, bLF dose-dependently affects IECs via metabolic, apoptotic and inflammatory pathways. It is important to select an appropriate dose when feeding neonates with bLF to avoid detrimental effects exerted by excessive doses.
Biological significance: The present work elucidates dose-dependent effects of bLF on the proteomic changes of IECs in vitro supplemented with data from a preterm pig study confirming detrimental effects of enteral feeding with the highest dose of bLF (10 g/L). The study contributes to further understanding on mechanisms that bLF, as an important milk protein, can regulate the homeostasis of the immature intestine. Results from this study urge neonatologists to carefully consider the dose of bLF to supplement into infant formula used for preterm neonates. Key words: apoptosis, inflammation, lactoferrin, necrotizing enterocolitis 2
ACCEPTED MANUSCRIPT 1. Introduction Lactoferrin (LF) is a multifunctional 80-kDa protein present in both human (1-7 g/L) and
PT
bovine (0.2-1.5 g/L) milk and colostrum [1,2]. Its bioactivity has been well documented, including its iron-binding, anti-microbial, immune-stimulating and anti-inflammatory effects [2–4]. LF has been
RI
suggested to protect preterm infants against the development of necrotizing enterocolitis (NEC), a
SC
devastating disease with high morbidity and mortality rates [3]. The high homology in protein
NU
sequence (77%) between human (hLF) and bovine lactoferrin (bLF) suggests that supplementation of bLF to infant formula may exert beneficial effects in a similar manner to that of hLF in human milk
MA
[5].
In preterm infants, bLF at a dose of 0.1 g/day protects against late-onset sepsis and acts in
D
synergy with probiotics to decrease NEC sensitivity [6]. In rodents, oral administration of 0.24-0.4 g/kg
TE
bLF decreased dextran sulfate sodium-induced colitis and inflammatory cytokine secretion [7]. A high
AC CE P
dose of oral bLF (20 g/L) protected germ-free newborn pigs against endotoxin-induced lethal shock [8], possibly due to the binding of bLF to lipopolysaccharide (LPS), which prevents LPS-induced inflammation [9,10]. However, we have previously shown in preterm pigs that bLF-enriched formula at a relatively high dose (10 g/L, 1.2 g/kg/day) did not protect against NEC, but tended to exacerbate the disease severity and increase intestinal permeability [11]. In porcine intestinal epithelial cells (IECs), low levels of bLF (0.1-1 g/L) stimulated cell proliferation whereas a high level of 10 g/L inhibited cell proliferation and induced inflammatory cytokine production via NF-κB signaling [11]. bLF may activate NF-κB following its binding to TNF receptors, as shown for monocytic cells [12], or to TLR4, as shown for macrophages [13]. Most IECs express the LF receptor (LfR) [14], and upon binding of bLF, the LfR undergoes internalization, which enables the intracellular uptake of bLF and causes the activation of ERK and cell proliferation [4]. Consequently, the bLF concentration and expression of
3
ACCEPTED MANUSCRIPT IEC receptors, such as LfR and TLR-4, may be crucial in determining which intracellular signaling pathways and regulatory effects are initiated by bLF stimulation.
PT
In this study, the porcine IEC cell line PsIc1, derived from not fully differentiated crypt cells of a 6-month-old pig, was used as a model of immature IECs. We hypothesized that bLF dose-
RI
dependently affects the fate and functions of IECs, as reflected by the differentiated cellular proteome
SC
following stimulation with bLF at 0, 0.1, 1 and 10 g/L. The chosen doses of 0.1 and 1 g/L were close to
NU
LF levels in bovine and human mature milk [1,2], whereas 10 g/L is a relatively high dose, slightly higher than the average level of LF in human colostrum. The proteins involved in key cellular
MA
pathways identified by proteomics were further analyzed in the small intestine of formula-fed preterm pigs with a supplementation of 10 g/L bLF. Our study aimed to help understand the mechanisms by
2.1. Cell culture
AC CE P
2. Materials and methods
TE
D
which different bLF doses may exert different, or even contrasting, effects on the immature intestine.
Porcine IECs (PsIc1, Bionutritech, Lunel, France) at passages 5-25 were cultured at 37 oC and 5% CO2 using an advanced DMEM medium supplemented with 2% heat-inactivated fetal bovine serum, 40 U/mL penicillin, 40 μg/mL streptomycin and 2 mM glutamax (Life Technologies, Nærum, Denmark). 2.2. Proteomics of IECs treated by bLF A high-throughput approach using LC-MS-based proteomics with iTRAQ labeling was applied [15]. PsIc1 cells were cultured until 90-95% confluency and maintained in serum-free medium for 24 h, when cells reached 100% of confluency, prior to stimulation with bLF (Morinaga Milk Industry, Japan) at different doses of 0, 0.1, 1 and 10 g/L (LF0, LF0.1, LF1 and LF10, respectively) in triplicates (4 treatments × 3 replicates = 12 samples) for another 24 h. bLF was 15% iron-saturated with 4
ACCEPTED MANUSCRIPT a low endotoxin content (1.6 EU/mg protein) detected by Limulus Amebocyte Lysate Chromogenic Endotoxin Quantitation Kit (Fisher Scientific, Slangerup, Denmark). Thereafter, the IECs were
PT
collected and lyzed by sonication in TES buffer (10 mM Tris pH 7.6, 1 mM EDTA and 0.25 M sucrose), and the protein concentration was determined using the BCA protein assay (Thermo
RI
Scientific, Denmark). The extracted proteins were precipitated in cold acetone and reduced with
SC
dithiothreitol, and free cysteines were blocked with methyl methanethiosulfonate before trypsin
NU
digestion overnight. The obtained peptides were labeled with 4-plex iTRAQ tags (Applied Biosystems, Forster, CA, USA), and a common reference sample was made by pooling equal amounts of protein
MA
from all 12 samples. This reference was then used in each of the 4-plexed samples, as specified in Supplementary Table S1 online. The procedure from lysis to labeling has previously been described in
TE
until LC-MS/MS analysis.
D
detail [16]. After labeling, the 4-plexed samples were dried at room temperature and stored at -80 oC
AC CE P
The LC-MS/MS analysis and raw-data file processing were performed as previously described [17]. In brief, the iTRAQ-labelled tryptic peptide samples were analyzed on a Dionex RSLC UPLC system with a nanopump module coupled with a Thermo-Electron Q-exactive Plus mass spectrometer (MS, Thermo Scientific, Waltham, USA). Duplicate runs of each 4-plexed sample were loaded onto a C18 reversed-phase column (Dionex; Acclaim PepMap100 C18, with a 5 μm pre-column and a 50 cm Acclaim Pepmap RSLC, 75 μm ID main column, Thermo Scientific) and eluted with a linear gradient from 96% formic acid (1%) and 4% acetonitrile to 65% formic acid and 35% acetonitrile. The 12 precursor-ions with the highest intensity were selected for higher energy collisional dissociation (HCD) fragmentation using m/z 100 as fixed lower set point. The resulting raw files were analyzed using Thermo Proteome Discoverer 1.4 (Thermo Scientific) and MaxQuant LFQ, as previously described [18]. The protein abundance in treatment groups was reported relative to that in the common reference sample. Data were searched against Uniprot Sus scrofa (UPID000008227) and Bos taurus 5
ACCEPTED MANUSCRIPT (UPID000005136) protein reference proteome databases and in-house compiled database optimized for a novel PeptideAtlas database for Sus scrofa. The MS proteomic data have been deposited to the
PT
ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD002755.
SC
protein loaded into the gel, as previously described [19].
RI
The intracellular uptake of bLF in the IECs was determined by SDS-PAGE with 15 µg of total
NU
2.3. Preterm pig study and gut morphology
Twenty-eight preterm pigs from four sows (Large White × Danish Landrace × Duroc) were
MA
delivered by caesarean section at 90-92% of gestational age (105-106 days). The pigs were housed, and administered parenteral and enteral nutrition for five days prior to euthanasia as previously described
D
[11]. The enteral nutrition was either infant formula (CON, n = 15) or infant formula enriched with 10
TE
g/L bLF (LF, n = 11) [11]. The diets were optimized for piglet’s needs with energy and macronutrient
AC CE P
composition according to Table 1, as previously reported [11]. Commercial ingredients used were: protein (Lacprodan DI-9224, Arla Foods Ingredients (AFI), Denmark), lactose (Variolac 960, AFI), maltodextrin (Ross Polycose, Abbott Nutrition, USA), lipids (Liquigen and Calogen, Nutricia), vitamins and minerals (SHS Seravit, Nutricia). The bLF used was from the same source as in the cell study. At euthanasia, the clinical sign of NEC was recorded, based on our established marcroscopic NEC scoring system from 1-6 [11]. NEC was regarded as positive when a score of 3 or higher was observed in any gastrointestinal regions (stomach, proximal, middle, distal intestine and colon). Proximal small intestinal sections were removed, fixed by formaldehyde and embedded in paraffin prior to staining with hematoxylin and eosin to measure villus height and crypt depth, as previously described [11]. For each individual section, the length of 10 random villi and crypts was measured and the average values for each pig were used for statistics. The protocol of the animal study was approved by the Danish National Committee on Animal Experimentation. 6
ACCEPTED MANUSCRIPT 2.4. Western blot for intestinal tissues Middle small intestinal tissues of the healthy CON and LF pigs (n = 5/group), that did not
PT
develop NEC, were homogenized in an extraction buffer containing 1% Triton X-100 and 1% of protease inhibitor cocktail [15]. Thereafter, 25 µg protein of all tissue samples was resolved on a 16%
RI
SDS-PAGE gel and transferred onto PVDF membranes (Life Technologies) [15]. Membranes were
SC
then probed with primary and secondary antibodies to detect bLF (Hycult Biotech, The Netherlands),
NU
proliferating cell nuclear antigen (PCNA), apoptosis regulator Bax, B cell lymphoma 2 (Bcl-2), caspase 3 (all from Santa Cruz Biotechnology, CA, USA), and heavy chain ferritin (H-ferritin, Abcam,
MA
Cambridge, UK). Band intensity was analyzed and extracted from ImageJ (National Institute of Health, USA).
D
2.5. Statistics
TE
The relative abundance of identified proteins from cellular proteomics was imported into R
AC CE P
integrated with R Studio for statistical analysis. Abundance of each protein was fitted to linear mixedeffect models with and without LF concentration as a fixed factor and iTRAQ plex as a random factor using the ‘lmer’ function [20]. The significance of bLF treatment was tested by comparing these two models using ‘anova’ function. The difference among the groups was further tested using with a “Tukey test” using the ‘glht’ function [21]. P values were extracted from specific tests. An identified protein with an overall P value < 0.05 of the treatment and at least one significance (P < 0.05) in the post hoc comparisons was regarded as a significantly affected protein by bLF treatment. Among these proteins, only those showing fold-changes greater than 1.2 or less than 0.85 in LF 0.1, 1 or 10 compared with LF0, were used for assignment of biological function. For the preterm pig study, the villus height, crypt depth, and ratio of villus height and crypt depth were analyzed using linear mixed-effect model with nutritional treatment and litter as fixed factors (JMP, SAS Institute, NC, USA). Band densities of selected proteins by Western blot were 7
ACCEPTED MANUSCRIPT analyzed using a linear mixed model with treatment as the fix factor and a following post hoc Tukey
PT
test to compare levels in CON and LF pigs (JMP).
RI
3. Results 3.1. IEC proteomics
SC
In total, 1258 proteins were identified in at least three plexes of the proteomic study. Following
NU
statistics, bLF and 193 porcine proteins showed significant differences in abundance among the four treatment groups (P < 0.05). After applying a cut-off threshold of 1.2-fold for up-regulation and 0.85-
MA
fold for down-regulation, 122 porcine proteins and bLF were differentially expressed (63 up-regulated and 60 down-regulated proteins). Information of the identified proteins, including accession number, %
D
coverage, number of unique peptides, protein score and fold change are categorized into groups based
TE
on their major biological functions (Supplementary Table S2 online). The major protein functions
AC CE P
regulated by bLF were proliferation and apoptosis (14 proteins), energy metabolism (12 proteins), protein synthesis, processing and degradation (23 proteins), DNA and RNA binding and processing (15 proteins), cytoskeleton (13 proteins), protein transport (9 proteins), amino acid metabolism (7 proteins), and signal transduction (4 proteins).
Both SDS-PAGE and proteomics results showed that the intracellular bLF level increased after bLF stimulation, indicating cellular bLF uptake (Fig. 1A-B). The bLF uptake reached a plateau at 1 g/L. 3.2. bLF effects at high and low doses in vitro At low doses (0.1-1 g/L), bLF up-regulated 11 proteins involved in glycolysis and energy metabolism (pyruvate kinase, pyruvate dehydrogenase, pyruvate carboxylase and UDP-glucose 4-epimerase), and
8
ACCEPTED MANUSCRIPT protein synthesis and processing (Fig. 2). Negligible differences in these protein levels were found between LF0.1 and LF1.
PT
Only the high dose (10 g/L) of bLF, but not 0.1-1 g/L, altered eight proteins involved in proliferation and apoptosis (Fig. 3A). Three up-regulated proteins play a role in the early phase of
RI
apoptosis and cell death, including apoptosis inducing factor (AIF), annexin-1 and cyclophilin-40. The
SC
remaining five proteins, including two anti-apoptotic proteins (catalase and huntingtin-interacting
NU
protein 1, HIP1) and three proliferation-inducing proteins (CD63, granulins and 7-dehydrocholesterol reductase - 7DHCR) were down-regulated in LF10. bLF at 10 g/L also impaired energy metabolism by
MA
down-regulating 10 out of 11 proteins involved in the TCA cycle, electron transport and ATP synthesis (Fig. 3B).
D
bLF at 10 g/L differentially expressed five proteins involved in the iron-dependent and HIF-1
TE
pathways (Fig. 3C). HIF-1 pathway-related proteins were ubiquitin carboxyl-terminal hydrolase
AC CE P
(UCH), DNA-(apurinic or apyrimidinic site) lyase (Apex/Ref-1) and heterogeneous nuclear ribonucleoprotein a2/b1 (hnRNP a2/b1). UCH and Apex/Ref-1 were up-regulated by all bLF doses, with higher levels in LF10 than in LF0.1 and LF1. In contrast, hnRNP a2/b1 and two other proteins involved in iron binding and transport (transferrin receptor and vesicle trafficking protein 22b) were down-regulated in
LF10, relative to
other groups (Fig. 3C).
With the exception of
adenosylhomocysteinase being up-regulated by all bLF doses, the other five proteins involved in amino acid metabolism were down-regulated in LF10 relative to other groups (Fig. 3D). 3.3. Clinical evaluation and gut morphology in CON and LF pigs At euthanasia on day 5, the incidence of NEC did not differ between CON (9/15 pigs, 60%) and LF pigs (8/13 pigs, 62%), but LF pigs had more severe colonic lesions than CON pigs (higher NEC score, P < 0.05), as previously reported [11]. In the proximal small intestine, villus height in LF pigs tended to be lower than
that in CON pigs (P = 0.09, Fig. 4A), whereas crypt depth was similar between the two groups (Fig. 9
ACCEPTED MANUSCRIPT 4B). Of note, the ratio of villus height and crypt depth was significantly lower in LF vs. CON pigs (P < 0.05, Fig. 4C), suggesting a decreased intestinal epithelial cell proliferation or an increased cell death in
PT
LF pigs. Representative cross sections of the proximal small intestine reflected the shorter villi in LF vs. CON pigs (Fig. 4D-E).
RI
3.4. bLF effects on the immature pig intestine
SC
bLF was detected in the small intestine of LF pigs, but absent in CON pigs (P < 0.001, Fig. 5A),
NU
indicating bLF uptake into the intestine of LF pigs. In line with this uptake, the intestine of LF pigs had higher HIF-1α levels than those of CON pigs (P < 0.05, Fig. 5B). H-ferritin (an iron storing protein)
MA
and PCNA (a protein involved in DNA damage) did not differ between the two treatments (Fig. 5C-D). Proteins involved in apoptosis, as shown by Bax/Bcl-2 ratio, were elevated in LF vs. CON pigs (P <
4. Discussion
TE
AC CE P
the two groups (Fig. 5F).
D
0.05, Fig. 5E). In contrast, the levels of pro-caspase 3 and activated caspase 3 did not differ between
LF is an important protein in both human and bovine milk with multiple functions [2–4]. It has a great potential as a supplement for preterm infant formulas [22] to exert protection against infection in early life [6]. However, previous studies have not elucidated the mechanisms by which LF exerts its protective effects. In the present study, we examined the effects of bLF at doses that were comparable to the levels of LF in bovine and human milk using both IECs in vitro and immature pig intestine in vivo. We demonstrated the beneficial effects of bLF at low doses (0.1-1 g/L, close to LF levels in bovine and human mature milk) and detrimental effects at a high dose (10 g/L, close to LF levels in human colostrum). In the cultured IECs, bLF at this high dose mainly targeted pathways related to energy metabolism, cell survival/apoptosis, as well as the HIF-1 pathway, which is highly related to iron homeostasis and cell death. The detrimental effect of bLF at this high dose was also observed in 10
ACCEPTED MANUSCRIPT the immature small intestine of piglets receiving supplemented bLF, shown by greater colonic lesions, shorter intestinal villus height/crypt depth ratio, and an elevated Bax/Bcl-2 ratio and HIF-1α levels.
PT
LF binds to its receptor on IECs and this facilitates receptor internalization and LF uptake [4,14]. In the current study, we confirmed the uptake of bLF in both cultured IECs and in the immature
RI
pig small intestine. Compared with 1 g/L, bLF at 10 g/L did not further increase the bLF uptake,
SC
probably due to the saturation of available cell surface receptors by 1 g/L of bLF. Any unbound bLF
NU
may thereby bind to alternative membrane receptors such as TLR4 or TNF receptors [12,13,23] and trigger inflammation. For instance, we have previously shown that bLF at 10 g/L initiated NF-κB
MA
signaling leading to an increased IL-8 production in cultured porcine IECs [11]. Furthermore, as a transferrin, bLF strongly binds iron in the intestinal lumen to facilitate iron absorption via LF
D
internalization [24]. Iron is then released from bLF and transferred to mitochondria or stored in ferritin
TE
[24], causing ferritin up-regulation [25]. In the current study in preterm pigs, intestinal H-ferritin was
AC CE P
unchanged in LF pigs relative to CON pigs (Fig. 5C), suggesting an impaired iron storage following the accumulation of iron-bound LF (Fig. 5A). Because H-Ferritin is anti-apoptotic [26], the impaired up-regulation of this protein after LF challenge may contribute to the increased apoptosis in the intestine of LF pigs.
bLF at low doses (0.1-1 g/L), similar to the LF levels in bovine and human milk, up-regulated proteins involved in glycolysis, energy metabolism and protein synthesis including the important enzymes pyruvate kinase, pyruvate carboxylase and pyruvate dehydrogenase. Pyruvate kinase catalyzes the final step of the glycolytic pathway to form pyruvate, while pyruvate can be converted into oxaloacetate, an intermediate in the TCA cycle, by pyruvate carboxylase [27]. Pyruvate dehydrogenase catalyzes the reaction from pyruvate to acetyl-CoA, a key molecule in many metabolic processes [27]. The up-regulation of these enzymes suggests that low bLF doses facilitate energy production and contribute to IEC proliferation [4,11]. Among the identified proteins involved in 11
ACCEPTED MANUSCRIPT protein synthesis, three aminoacyl-tRNA ligases (glycine, isoleucine and tryptophan-tRNA ligases) and DNA lyase were also up-regulated by bLF at 0.1 and 1 g/L. These ligases belong to the tRNA
PT
synthetase family, catalyzing the ligation between amino acids and their cognate tRNA during the initial step of protein synthesis [28]. DNA lyase is a protein with dual functions: stimulation of DNA
RI
repair and facilitation of DNA binding activity of numerous transcription factors, thereby playing
SC
important roles in protein synthesis [29]. The up-regulation of these proteins may reflect the elevated
NU
protein synthesis or turnover, which is required for IEC proliferation [30]. bLF at low doses did not affect proteins associated with cell apoptosis and inflammation, agreeing with our previous study
MA
showing consistent levels of HIF-1α and IL-8 secretion following IEC stimulation with bLF at these doses [11].
D
bLF at 10 g/L altered 26 proteins towards the directions of apoptosis, impaired energy and
TE
amino acid metabolism, and activation of the HIF-1 pathway in vitro. In addition, the lower ratio of
AC CE P
villus height/crypt depth and higher colonic lesions in LF pigs imply a reduced proliferation or elevated cell death of the IECs in the small intestine (Fig. 4C). Tissue analysis further supports the suggestion of elevated bLF-induced apoptosis with a greater Bax/Bcl-2 ratio (a key regulator of apoptotic pathways) and increased HIF-1α levels (regulator of HIF-1 pathway) in LF pigs. Five apoptosis-related proteins were up-regulated by bLF at 10 g/L in vitro including annexin1, AIF, cyclophilin 40, catalase and huntingtin-interacting protein 1 (HIP1). Annexin-1 is involved in pro-apoptotic pathways [31], and it is also released from apoptotic cells and triggers cell engulfment by macrophages [32]. AIF is a protein translocated into the nucleus following apoptotic stimuli to initiate caspase-independent apoptosis [33]. Cyclophilin-40 overexpression causes mitochondrial swelling and cell death and increases cell susceptibility to oxidative stress [34,35]. Catalase is an enzyme that is upregulated when protecting cells against apoptosis after exposure to reactive oxygen species [36,37]. HIP1 maintains cellular survival, and its deficiency leads to apoptosis [38,39]. Elevated annexin-1, AIF 12
ACCEPTED MANUSCRIPT and cyclophilin 40, and decreased catalase and HIP1 levels elicited only by a bLF dose of 10 g/L reflect that this high dose may reduce cell survival and trigger apoptosis. Besides, proliferation-
PT
inducing proteins, including 7-dehydrocholesterol reductase, CD63 and granulins were down-regulated by 10 g/L of bLF. 7-dehydrocholesterol reductase is involved in cell proliferation in rats [40]. CD63 is
RI
the cell receptor of TIMP-1 required for proliferation [41]. Granulins play a role in the mediation of
SC
cell proliferation [42] and ERK activation [43]. These trends are consistent with our previous study
NU
showing a marked reduction of IEC proliferation and impaired ERK signaling by 10 g/L bLF [11]. The Bax/Bcl-2 ratio can determine whether cells undergo apoptosis (increased ratio) or
MA
survival (reduced ratio), and their role in apoptosis is partly triggered by hypoxia and impaired energy metabolism [44]. Hypoxia, which is associated with the activation of the HIF-1 pathway and NEC [45],
D
can impair energy metabolism, inhibit electron transport and reduce ATP synthesis. These events
TE
activate Bax leading to cytochome C release from the mitochondria and apoptotic signaling cascades
AC CE P
[44]. In addition, Bcl-2 can inhibit Bax activation [44]. Importantly, bLF can act as a mimetic of hypoxia to up-regulate HIF-1α and the HIF-1 pathway [46] and bLF at 10 g/L elevated HIF-1α in cultured IECs [11]. In this current study, the activation of the HIF-1 pathway induced by 10 g/L of bLF in IECs was supported by proteomic data via altered levels of ubiquitin carboxyl-terminal hydrolase, hnRNPa2/b1 and Apex/Ref-1, which all lead to increased HIF-1α stability and HIF-1 transcriptional activity [47–50]. We also showed in the present study that 10 g/L bLF down-regulated 10 proteins involved in energy metabolism in IECs, and increased colonic lesions, intestinal HIF-1α levels and Bax/Bcl-2 ratio in preterm pigs, indicating a tendency to trigger intestinal apoptosis. However, caspase 3 did not show a clear activation in LF pigs, implying that IEC apoptosis may occur via caspase 3independent pathways. We speculate that bLF supplemented into enteral formula at high doses may act to mimic hypoxic conditions and elevate the HIF-1 pathway, leading to impaired energy metabolism
13
ACCEPTED MANUSCRIPT and activation of caspase 3-independent apopotic pathways involving Bax/Bcl-2 in the immature
PT
intestine.
RI
5. Conclusions
We conclude that exogenous bLF modulates intestinal proteins in preterm pigs and regulates
SC
the IEC proteome via survival, apoptotic and inflammatory pathways. Relatively low doses (close to
NU
levels present in bovine and human mature milk) may be beneficial to stimulate intestinal maturation via induction of IEC proliferation. A higher dose, such as 10 g/L, may exert adverse and detrimental
MA
effects by triggering apoptosis and inflammation. In this study, proteomics was used to elucidate the cellular pathways affected by bLF in a dose-dependent manner. The pig experiment using only the high
D
dose was used to confirm the effects of this dose of bLF in inducing apoptotic conditions in the
TE
immature small intestine in vivo. A comprehensive preterm pig study investigating the dose-dependent
AC CE P
effects of bLF could be important to determine the optimal dose of bLF for formula supplementation in preterm infants, thereby supporting intestinal maturation and preventing inflammation.
Conflict of interest statement
All authors declare no competing financial interest.
Acknowledgements The present study was supported by a grant from the Danish Dairy Research Foundation. Dorte Thomassen and Yanqi Li are thanked for technical assistance with sample preparation and animal study. The Obelske Family Foundation, Svend Andersen Foundation and SparNord Foundation are acknowledged for their support with the analytical platform.
Author contribution 14
ACCEPTED MANUSCRIPT D.N.N, P.T.S and D.E.W.C designed the study. D.N.N., P.J., A.S., and E.B. conducted the experiments. D.N.N. and P.J. analyzed data and performed statistics. D.N.N. wrote the manuscript
PT
together with the other authors. D.E.W.C. had primary responsibility for the final content. All authors have read and approved the final manuscript. Supplementary information to this article can be found
SC
RI
online.
NU
References
[1] K. Marshall, Therapeutic applications of whey protein, Altern Med Rev. 9 (2004) 136–156.
MA
[2] E.N. Baker, H.M. Baker, A structural framework for understanding the multifunctional character of lactoferrin, Biochimie. 91 (2009) 3–10.
D
[3] D.H. Adamkin, Mother’s milk, feeding strategies, and lactoferrin to prevent necrotizing
TE
enterocolitis, J. Parenter. Enteral. Nutr. 36 (2012) 25S–9S.
AC CE P
[4] R. Jiang, V. Lopez, S.L. Kelleher, B. Lönnerdal, Apo- and holo-lactoferrin are both internalized by lactoferrin receptor via clathrin-mediated endocytosis but differentially affect ERK-signaling and cell proliferation in caco-2 cells, J. Cell. Physiol. 226 (2011) 3022–3031. [5] P. Manzoni, L. Decembrino, I. Stolfi, L. Pugni, M. Rinaldi, S. Cattani, et al., Lactoferrin and prevention of late-onset sepsis in the pre-term neonates, Early Hum. Dev. 86 (2010) 59–61. [6] P. Manzoni, M. Rinaldi, S. Cattani, L. Pugni, M.G. Romeo, H. Messner, et al., Bovine lactoferrin supplementation for prevention of late-onset sepsis in very low-birth-weight neonates: a randomized trial, JAMA. 302 (2009) 1421–1428. [7] J.-I. Togawa, H. Nagase, K. Tanaka, M. Inamori, A. Nakajima, N. Ueno, et al., Oral administration of lactoferrin reduces colitis in rats via modulation of the immune system and correction of cytokine imbalance, J. Gastroenterol. Hepatol. 17 (2002) 1291–1298.
15
ACCEPTED MANUSCRIPT [8] W.J. Lee, J.L. Farmer, M. Hilty, Y.B. Kim, The protective effects of lactoferrin feeding against endotoxin lethal shock in germfree piglets, Infect. Immun. 66 (1998) 1421–1426.
PT
[9] N. Orsi, The antimicrobial activity of lactoferrin: Current status and perspectives, Biometals. 17 (2004) 189–196.
RI
[10] B.J. Appelmelk, Y.Q. An, M. Geerts, B.G. Thijs, H.A. de Boer, D.M. MacLaren, et al.,
SC
Lactoferrin is a lipid A-binding protein, Infect. Immun. 62 (1994) 2628–2632.
NU
[11] D.N. Nguyen, Y. Li, P.T. Sangild, S.B. Bering, D.E.W. Chatterton, Effects of bovine lactoferrin on the immature porcine intestine, Br. J. Nutr. 111 (2014) 321–331.
MA
[12] S.-M. Oh, S.-H. Lee, B.-J. Lee, C.-W. Pyo, N.-K. Yoo, S.Y. Lee, et al., A distinct role of neutrophil lactoferrin in RelA/p65 phosphorylation on Ser536 by recruiting TNF receptor-
D
associated factors to IkappaB kinase signaling complex, J. Immunol. 179 (2007) 5686–5692.
TE
[13] C.S. Curran, K.P. Demick, J.M. Mansfield, Lactoferrin activates macrophages via TLR4-
AC CE P
dependent and -independent signaling pathways, Cell. Immunol. 242 (2006) 23–30. [14] B. Lönnerdal, R. Jiang, X. Du, Bovine Lactoferrin Can Be Taken Up By The Human Intestinal Lactoferrin Receptor And Exert Bioactivities, J. Pediatr. Gastroenterol. Nutr. (2011). [15] D.N. Nguyen, P. Jiang, S. Jacobsen, P.T. Sangild, E. Bendixen, D.E.W. Chatterton, Protective Effects of Transforming Growth Factor β2 in Intestinal Epithelial Cells by Regulation of Proteins Associated with Stress and Endotoxin Responses, PLoS ONE. 10 (2015) e0117608. [16] M. Danielsen, H. Hornshøj, R.H. Siggers, B.B. Jensen, A.G. van Kessel, E. Bendixen, Effects of bacterial colonization on the porcine intestinal proteome, J. Proteome Res. 6 (2007) 2596–2604. [17] T. Bennike, U. Ayturk, C.M. Haslauer, J.W. Froehlich, B.L. Proffen, O. Barnaby, et al., A normative study of the synovial fluid proteome from healthy porcine knee joints, J. Proteome Res. (2014).
16
ACCEPTED MANUSCRIPT [18] J. Cox, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, MaxLFQ allows accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio
PT
extraction, Mol. Cell. Proteomics. (2014). [19] D.E.W. Chatterton, J.T. Rasmussen, C.W. Heegaard, E.S. Sørensen, T.E. Petersen, In vitro
SC
functions, Trends Food Sci. Technol. 15 (2004) 373–383.
RI
digestion of novel milk protein ingredients for use in infant formulas: Research on biological
NU
[20] D. Bates, M. Mächler, B. Bolker, S. Walker, Fitting Linear Mixed-Effects Models using lme4, arXiv:1406.5823 [stat]. (2014). http://arxiv.org/abs/1406.5823 (accessed September 1, 2014).
MA
[21] T. Hothorn, F. Bretz, P. Westfall, Simultaneous inference in general parametric models, Biom J. 50 (2008) 346–363.
D
[22] D. Sharma, S. Shastri, Lactoferrin and neonatology - role in neonatal sepsis and necrotizing
TE
enterocolitis: present, past and future, J. Matern. Fetal. Neonatal. Med. (2015) 1–8.
AC CE P
[23] K. Ando, K. Hasegawa, K.-I. Shindo, T. Furusawa, T. Fujino, K. Kikugawa, et al., Human lactoferrin activates NF-kappaB through the Toll-like receptor 4 pathway while it interferes with the lipopolysaccharide-stimulated TLR4 signaling, FEBS J. 277 (2010) 2051–2066. [24] H.S. Birgens, L.Ø. Kristensen, N. Borregaard, H. Karle, N.E. Hansen, Lactoferrin-mediated transfer of iron to intracellular ferritin in human monocytes, Eur. J. Haematol. 41 (1988) 52–57. [25] S. Shimmura, M. Shimoyama, M. Hojo, K. Urayama, K. Tsubota, Reoxygenation injury in a cultured corneal epithelial cell line protected by the uptake of lactoferrin., Invest. Ophthalmol. Vis. Sci. 39 (1998) 1346–1351. [26] A. Cozzi, S. Levi, B. Corsi, P. Santambrogio, A. Campanella, G. Gerardi, et al., Role of iron and ferritin in TNFα-induced apoptosis in HeLa cells, FEBS Letters. 537 (2003) 187–192. [27] S.Y. Lunt, M.G. Vander Heiden, Aerobic Glycolysis: Meeting the Metabolic Requirements of Cell Proliferation, Annu Rev Cell Dev Biol. 27 (2011) 441–464. 17
ACCEPTED MANUSCRIPT [28] K. Wakasugi, B.M. Slike, J. Hood, A. Otani, K.L. Ewalt, M. Friedlander, et al., A human aminoacyl-tRNA synthetase as a regulator of angiogenesis, PNAS. 99 (2002) 173–177.
PT
[29] Y. Cun, N. Dai, M. Li, C. Xiong, Q. Zhang, J. Sui, et al., APE1/Ref-1 enhances DNA binding activity of mutant p53 in a redox-dependent manner, Oncol. Rep. 31 (2014) 901–909.
RI
[30] S.G. Park, P. Schimmel, S. Kim, Aminoacyl tRNA synthetases and their connections to disease,
SC
PNAS. 105 (2008) 11043–11049.
NU
[31] L.H.K. Lim, S. Pervaiz, Annexin 1: the new face of an old molecule, FASEB J. 21 (2007) 968– 975.
MA
[32] M. Scannell, M.B. Flanagan, A. deStefani, K.J. Wynne, G. Cagney, C. Godson, et al., Annexin-1 and peptide derivatives are released by apoptotic cells and stimulate phagocytosis of apoptotic
D
neutrophils by macrophages, J. Immunol. 178 (2007) 4595–4605.
TE
[33] N. Bidère, H.K. Lorenzo, S. Carmona, M. Laforge, F. Harper, C. Dumont, et al., Cathepsin D
AC CE P
Triggers Bax Activation, Resulting in Selective Apoptosis-inducing Factor (AIF) Relocation in T Lymphocytes Entering the Early Commitment Phase to Apoptosis, J. Biol. Chem. 278 (2003) 31401–31411.
[34] C.P. Baines, R.A. Kaiser, N.H. Purcell, N.S. Blair, H. Osinska, M.A. Hambleton, et al., Loss of cyclophilin D reveals a critical role for mitochondrial permeability transition in cell death, Nature. 434 (2005) 658–662. [35] Y. Li, N. Johnson, M. Capano, M. Edwards, M. Crompton, Cyclophilin-D promotes the mitochondrial permeability transition but has opposite effects on apoptosis and necrosis, Biochem. J. 383 (2004) 101. [36] K. Hasegawa, S. Wakino, K. Yoshioka, S. Tatematsu, Y. Hara, H. Minakuchi, et al., Sirt1 protects against oxidative stress-induced renal tubular cell apoptosis by the bidirectional regulation of catalase expression, Biochem. Biophys. Res. Commun. 372 (2008) 51–56. 18
ACCEPTED MANUSCRIPT [37] H.-U. Simon, A. Haj-Yehia, F. Levi-Schaffer, Role of reactive oxygen species (ROS) in apoptosis induction, Apoptosis. 5 (2000) 415–418.
PT
[38] D.S. Rao, T.S. Hyun, P.D. Kumar, I.F. Mizukami, M.A. Rubin, P.C. Lucas, et al., Huntingtininteracting protein 1 is overexpressed in prostate and colon cancer and is critical for cellular
RI
survival, J. Clin. Investig. 110 (2002) 351–360. doi:10.1172/JCI15529.
SC
[39] I. Dragatsis, M.S. Levine, S. Zeitlin, Inactivation of Hdh in the brain and testis results in
NU
progressive neurodegeneration and sterility in mice, Nat. Genet. 26 (2000) 300–306. [40] S. Wu, F.D. Luca, Role of Cholesterol in the Regulation of Growth Plate Chondrogenesis and
MA
Longitudinal Bone Growth, J. Biol. Chem. 279 (2004) 4642–4647. [41] Y. Xia, N. Yeddula, M. Leblanc, E. Ke, Y. Zhang, E. Oldfield, et al., Reduced cell proliferation
D
by IKK2 depletion in a mouse lung-cancer model, Nat. Cell. Biol. 14 (2012) 257–265.
TE
[42] P.C. Hanington, D.R. Barreda, M. Belosevic, A Novel Hematopoietic Granulin Induces
9970.
AC CE P
Proliferation of Goldfish (Carassius auratus L.) Macrophages, J. Biol. Chem. 281 (2006) 9963–
[43] J.Q. Feng, F.-J. Guo, B.-C. Jiang, Y. Zhang, S. Frenkel, D.-W. Wang, et al., Granulin epithelin precursor: a bone morphogenic protein 2-inducible growth factor that activates Erk1/2 signaling and JunB transcription factor in chondrogenesis, FASEB J. 24 (2010) 1879–1892. [44] A.E. Greijer, E. van der Wall, The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis, J Clin Pathol. 57 (2004) 1009–1014. [45] D.J. Hackam, A. Afrazi, M. Good, C.P. Sodhi, Innate immune signaling in the pathogenesis of necrotizing enterocolitis, Clin. Dev. Immunol. 2013 (2013) 475415. [46] E.T. Zakharova, V.A. Kostevich, A.V. Sokolov, V.B. Vasilyev, Human apo-lactoferrin as a physiological mimetic of hypoxia stabilizes hypoxia-inducible factor-1 alpha, Biometals. 25 (2012) 1247–1259. 19
ACCEPTED MANUSCRIPT [47] G.L. Semenza, Targeting HIF-1 for cancer therapy, Nat. Rev Cancer. 3 (2003) 721–732. [48] B. Seliger, A. Fedorushchenko, W. Brenner, A. Ackermann, D. Atkins, S. Hanash, et al.,
PT
Ubiquitin COOH-Terminal Hydrolase 1: A biomarker of renal cell carcinoma associated with enhanced tumor cell proliferation and migration, Clin. Cancer Res. 13 (2007) 27–37.
RI
[49] M. Fähling, R. Mrowka, A. Steege, P. Martinka, P.B. Persson, B.J. Thiele, Heterogeneous
NU
Stability, J. Biol. Chem. 281 (2006) 9279–9286.
SC
Nuclear Ribonucleoprotein-A2/B1 Modulate Collagen Prolyl 4-Hydroxylase, α (I) mRNA
[50] G. Tell, F. Quadrifoglio, C. Tiribelli, M.R. Kelley, The Many Functions of APE1/Ref-1: Not Only
AC CE P
TE
D
MA
a DNA Repair Enzyme, Antioxid. Redox. Signal. 11 (2008) 601–619.
20
ACCEPTED MANUSCRIPT Figure legends Fig. 1. Lactoferrin (LF) detected in porcine IECs by SDS-PAGE (A) and by quantification from
PT
iTRAQ-LCMS-based proteomics following treatment with LF 0, 0.1, 1 and 10 g/L for 24 h (B). * and
RI
#: P < 0.05, compared with LF0 and LF 0.1, respectively.
SC
Fig. 2. bLF at the low doses (0.1-1 g/L) up-regulated proteins involved in glycolysis, metabolism, protein synthesis and processing. * and #: P < 0.05, relative to LF0 and LF0.1, respectively .
NU
Fig. 3. Differentially expressed proteins involved in cell proliferation and cell death (A), energy
MA
metabolism (B), the HIF-1 pathway (C), and amino acid metabolism (D) induced by bLF. *, #, and §: P < 0.05, relative to LF0, LF0.1 and LF1, respectively. AIF: apoptosis inducing factor; HIP1: Huntingtin
D
interacting protein 1; 7DHCR:7-dehydrocholesterol reductase; UCH: ubiquitin carboxyl-terminal
TE
hydrolase; hnRNP a2/b1: heterogeneous nuclear ribonucleoprotein a2/b1.
AC CE P
Fig. 4. Intestinal mucosal structure in CON pigs (formula-fed, n = 13) and LF pigs (10 g/L bLFenriched formula-fed, n = 11). (A) Villus height, (B) Crypt depth, (C) Ratio of villus height/crypt depth, (D and E) Representative hematoxylin and eosin stained cross sections of the proximal intestine in CON (D) and LF pigs (E). Values are means ± SEM. * P < 0.05. Fig. 5. Western blot analysis of bLF (A) and proteins associated with apoptosis and iron-dependent pathways, including HIF-1α (B), H-ferritin (C), PCNA (D), Bax/Bcl-2 (E) and caspase 3 (F), in the middle intestinal tissues of preterm pigs fed control formula or bLF-enriched formula at 10 g/L. Values are means ± SEM. * and ** P < 0.05 and 0.01, respectively.
21
ACCEPTED MANUSCRIPT
4.51
4.51
Protein, g/L
79
79
Whey protein, g/L
79
69
Supplemented bLF, g/L
0
10
Carbohydrates, g/L
62.5
62.5
Maltodextrin, g/L
47.5
47.5
Lactose, g/L
15
Fat, g/L
57
RI
Energy, MJ/L
SC
LF
NU
CON
PT
Table 1. Macronutrient compositions of formula diets for CON and LF pigs1
15
MA
57
1
AC CE P
TE
D
Energy and macronutrient composition was calculated according to the ingredient specifications provided by the suppliers (Eurofins Steins Laboratorium, and Morinaga Industry)
22
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
Fig. 1
23
Fig. 2
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
24
Fig. 3
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
25
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Fig. 4
26
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Fig. 5
27
ACCEPTED MANUSCRIPT Biological significance: The present work elucidates dose-dependent effects of bLF on the proteomic changes of IECs in vitro supplemented with data from a preterm pig study confirming detrimental
PT
effects of enteral feeding with the highest dose of bLF (10 g/L). The study contributes to further understanding on mechanisms that bLF, as an important milk protein, can regulate the homeostasis of
RI
the immature intestine. Results from this study urge neonatologists to carefully consider the dose of
AC CE P
TE
D
MA
NU
SC
bLF to supplement into infant formula used for preterm neonates.
28
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
Graphical abstract
29
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Highlights: Lactoferrin (LF) is present in bovine and human colostrum and milk (0.1-10 g/L). Bovine LF (bLF) dose-dependently affected the intestinal cell proteome. bLF at low doses (0.1-1 g/L) regulated proteins supporting cell homeostasis. bLF at 10 g/L regulated proteins involved in apoptosis and inflammation. bLF at 10 g/L induced apoptosis and inflammation in the immature pig intestine.
30
S1874-3919(16)30081-1 doi: 10.1016/j.jprot.2016.03.020 JPROT 2459
To appear in:
Journal of Proteomics
Received date: Revised date: Accepted date:
17 December 2015 14 February 2016 11 March 2016
Please cite this article as: Nguyen Duc Ninh, Jiang Pingping, Stensballe Allan, Bendixen Emøke, Sangild Per T., Chatterton Dereck E.W., Bovine lactoferrin regulates cell survival, apoptosis and inflammation in intestinal epithelial cells and preterm pig intestine, Journal of Proteomics (2016), doi: 10.1016/j.jprot.2016.03.020
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Bovine lactoferrin regulates cell survival, apoptosis and inflammation in
PT
intestinal epithelial cells and preterm pig intestine Duc Ninh Nguyena,b, Pingping Jianga, Allan Stensballec, Emøke Bendixend, Per T. Sangilda, Dereck
Comparative Pediatrics and Nutrition, Department of Veterinary Clinical and Animal Sciences, and b
SC
a
RI
E.W. Chattertona,b*
NU
Department of Food Science, University of Copenhagen, DK-1958, Denmark Department of Health Science and Technology, Aalborg University, DK-9220, Denmark
d
Department of Molecular Biology and Genetics, Aarhus University, DK-8000, Denmark
MA
c
TE
D
* Corresponding author: Dereck E.W. Chatterton, Department of Food Science, and Department of Veterinary Clinical and Animal Sciences, University of Copenhagen, Rolighedsvej 30, Frederiksberg
AC CE P
DK-1958, Denmark. Tel: +45. 35 33 35 90; Fax: +45. 35 33 31 90; Email: [email protected]
Running title: LF regulates the fate of IECs Abbreviations: iTRAQ, isobaric tag for relative and absolute quantitation; LF, lactoferrin; NEC, necrotizing enterocolitis.
1
ACCEPTED MANUSCRIPT
Abstract
PT
Bovine lactoferrin (bLF) may modulate neonatal intestinal inflammation. Previous studies in intestinal epithelial cells (IECs) indicated that moderate bLF doses enhance proliferation whereas high doses
RI
trigger inflammation. To further elucidate cellular mechanisms, we profiled the porcine IEC proteome
SC
after stimulation with bLF at 0, 0.1, 1 and 10 g/L by LC-MS-based proteomics. Key pathways were
NU
analyzed in the intestine of formula-fed preterm pigs with and without supplementation of 10 g/L bLF. Levels of 123 IEC proteins were altered by bLF. Low bLF doses (0.1-1 g/L) up-regulated 11 proteins
MA
associated with glycolysis, energy metabolism and protein synthesis, indicating support of cell survival. In contrast, a high bLF dose (10 g/L) up-regulated three apoptosis-inducing proteins, down-
D
regulated five anti-apoptotic and proliferation-inducing proteins and 15 proteins related to energy and
TE
amino acid metabolism, and altered three proteins enhancing the hypoxia inducible factor-1 (HIF-1)
AC CE P
pathway. In the preterm pig intestine, bLF at 10 g/L decreased villus height/crypt depth ratio and upregulated the Bax/Bcl-2 ratio and HIF-1α, indicating elevated intestinal apoptosis and inflammation. In conclusion, bLF dose-dependently affects IECs via metabolic, apoptotic and inflammatory pathways. It is important to select an appropriate dose when feeding neonates with bLF to avoid detrimental effects exerted by excessive doses.
Biological significance: The present work elucidates dose-dependent effects of bLF on the proteomic changes of IECs in vitro supplemented with data from a preterm pig study confirming detrimental effects of enteral feeding with the highest dose of bLF (10 g/L). The study contributes to further understanding on mechanisms that bLF, as an important milk protein, can regulate the homeostasis of the immature intestine. Results from this study urge neonatologists to carefully consider the dose of bLF to supplement into infant formula used for preterm neonates. Key words: apoptosis, inflammation, lactoferrin, necrotizing enterocolitis 2
ACCEPTED MANUSCRIPT 1. Introduction Lactoferrin (LF) is a multifunctional 80-kDa protein present in both human (1-7 g/L) and
PT
bovine (0.2-1.5 g/L) milk and colostrum [1,2]. Its bioactivity has been well documented, including its iron-binding, anti-microbial, immune-stimulating and anti-inflammatory effects [2–4]. LF has been
RI
suggested to protect preterm infants against the development of necrotizing enterocolitis (NEC), a
SC
devastating disease with high morbidity and mortality rates [3]. The high homology in protein
NU
sequence (77%) between human (hLF) and bovine lactoferrin (bLF) suggests that supplementation of bLF to infant formula may exert beneficial effects in a similar manner to that of hLF in human milk
MA
[5].
In preterm infants, bLF at a dose of 0.1 g/day protects against late-onset sepsis and acts in
D
synergy with probiotics to decrease NEC sensitivity [6]. In rodents, oral administration of 0.24-0.4 g/kg
TE
bLF decreased dextran sulfate sodium-induced colitis and inflammatory cytokine secretion [7]. A high
AC CE P
dose of oral bLF (20 g/L) protected germ-free newborn pigs against endotoxin-induced lethal shock [8], possibly due to the binding of bLF to lipopolysaccharide (LPS), which prevents LPS-induced inflammation [9,10]. However, we have previously shown in preterm pigs that bLF-enriched formula at a relatively high dose (10 g/L, 1.2 g/kg/day) did not protect against NEC, but tended to exacerbate the disease severity and increase intestinal permeability [11]. In porcine intestinal epithelial cells (IECs), low levels of bLF (0.1-1 g/L) stimulated cell proliferation whereas a high level of 10 g/L inhibited cell proliferation and induced inflammatory cytokine production via NF-κB signaling [11]. bLF may activate NF-κB following its binding to TNF receptors, as shown for monocytic cells [12], or to TLR4, as shown for macrophages [13]. Most IECs express the LF receptor (LfR) [14], and upon binding of bLF, the LfR undergoes internalization, which enables the intracellular uptake of bLF and causes the activation of ERK and cell proliferation [4]. Consequently, the bLF concentration and expression of
3
ACCEPTED MANUSCRIPT IEC receptors, such as LfR and TLR-4, may be crucial in determining which intracellular signaling pathways and regulatory effects are initiated by bLF stimulation.
PT
In this study, the porcine IEC cell line PsIc1, derived from not fully differentiated crypt cells of a 6-month-old pig, was used as a model of immature IECs. We hypothesized that bLF dose-
RI
dependently affects the fate and functions of IECs, as reflected by the differentiated cellular proteome
SC
following stimulation with bLF at 0, 0.1, 1 and 10 g/L. The chosen doses of 0.1 and 1 g/L were close to
NU
LF levels in bovine and human mature milk [1,2], whereas 10 g/L is a relatively high dose, slightly higher than the average level of LF in human colostrum. The proteins involved in key cellular
MA
pathways identified by proteomics were further analyzed in the small intestine of formula-fed preterm pigs with a supplementation of 10 g/L bLF. Our study aimed to help understand the mechanisms by
2.1. Cell culture
AC CE P
2. Materials and methods
TE
D
which different bLF doses may exert different, or even contrasting, effects on the immature intestine.
Porcine IECs (PsIc1, Bionutritech, Lunel, France) at passages 5-25 were cultured at 37 oC and 5% CO2 using an advanced DMEM medium supplemented with 2% heat-inactivated fetal bovine serum, 40 U/mL penicillin, 40 μg/mL streptomycin and 2 mM glutamax (Life Technologies, Nærum, Denmark). 2.2. Proteomics of IECs treated by bLF A high-throughput approach using LC-MS-based proteomics with iTRAQ labeling was applied [15]. PsIc1 cells were cultured until 90-95% confluency and maintained in serum-free medium for 24 h, when cells reached 100% of confluency, prior to stimulation with bLF (Morinaga Milk Industry, Japan) at different doses of 0, 0.1, 1 and 10 g/L (LF0, LF0.1, LF1 and LF10, respectively) in triplicates (4 treatments × 3 replicates = 12 samples) for another 24 h. bLF was 15% iron-saturated with 4
ACCEPTED MANUSCRIPT a low endotoxin content (1.6 EU/mg protein) detected by Limulus Amebocyte Lysate Chromogenic Endotoxin Quantitation Kit (Fisher Scientific, Slangerup, Denmark). Thereafter, the IECs were
PT
collected and lyzed by sonication in TES buffer (10 mM Tris pH 7.6, 1 mM EDTA and 0.25 M sucrose), and the protein concentration was determined using the BCA protein assay (Thermo
RI
Scientific, Denmark). The extracted proteins were precipitated in cold acetone and reduced with
SC
dithiothreitol, and free cysteines were blocked with methyl methanethiosulfonate before trypsin
NU
digestion overnight. The obtained peptides were labeled with 4-plex iTRAQ tags (Applied Biosystems, Forster, CA, USA), and a common reference sample was made by pooling equal amounts of protein
MA
from all 12 samples. This reference was then used in each of the 4-plexed samples, as specified in Supplementary Table S1 online. The procedure from lysis to labeling has previously been described in
TE
until LC-MS/MS analysis.
D
detail [16]. After labeling, the 4-plexed samples were dried at room temperature and stored at -80 oC
AC CE P
The LC-MS/MS analysis and raw-data file processing were performed as previously described [17]. In brief, the iTRAQ-labelled tryptic peptide samples were analyzed on a Dionex RSLC UPLC system with a nanopump module coupled with a Thermo-Electron Q-exactive Plus mass spectrometer (MS, Thermo Scientific, Waltham, USA). Duplicate runs of each 4-plexed sample were loaded onto a C18 reversed-phase column (Dionex; Acclaim PepMap100 C18, with a 5 μm pre-column and a 50 cm Acclaim Pepmap RSLC, 75 μm ID main column, Thermo Scientific) and eluted with a linear gradient from 96% formic acid (1%) and 4% acetonitrile to 65% formic acid and 35% acetonitrile. The 12 precursor-ions with the highest intensity were selected for higher energy collisional dissociation (HCD) fragmentation using m/z 100 as fixed lower set point. The resulting raw files were analyzed using Thermo Proteome Discoverer 1.4 (Thermo Scientific) and MaxQuant LFQ, as previously described [18]. The protein abundance in treatment groups was reported relative to that in the common reference sample. Data were searched against Uniprot Sus scrofa (UPID000008227) and Bos taurus 5
ACCEPTED MANUSCRIPT (UPID000005136) protein reference proteome databases and in-house compiled database optimized for a novel PeptideAtlas database for Sus scrofa. The MS proteomic data have been deposited to the
PT
ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD002755.
SC
protein loaded into the gel, as previously described [19].
RI
The intracellular uptake of bLF in the IECs was determined by SDS-PAGE with 15 µg of total
NU
2.3. Preterm pig study and gut morphology
Twenty-eight preterm pigs from four sows (Large White × Danish Landrace × Duroc) were
MA
delivered by caesarean section at 90-92% of gestational age (105-106 days). The pigs were housed, and administered parenteral and enteral nutrition for five days prior to euthanasia as previously described
D
[11]. The enteral nutrition was either infant formula (CON, n = 15) or infant formula enriched with 10
TE
g/L bLF (LF, n = 11) [11]. The diets were optimized for piglet’s needs with energy and macronutrient
AC CE P
composition according to Table 1, as previously reported [11]. Commercial ingredients used were: protein (Lacprodan DI-9224, Arla Foods Ingredients (AFI), Denmark), lactose (Variolac 960, AFI), maltodextrin (Ross Polycose, Abbott Nutrition, USA), lipids (Liquigen and Calogen, Nutricia), vitamins and minerals (SHS Seravit, Nutricia). The bLF used was from the same source as in the cell study. At euthanasia, the clinical sign of NEC was recorded, based on our established marcroscopic NEC scoring system from 1-6 [11]. NEC was regarded as positive when a score of 3 or higher was observed in any gastrointestinal regions (stomach, proximal, middle, distal intestine and colon). Proximal small intestinal sections were removed, fixed by formaldehyde and embedded in paraffin prior to staining with hematoxylin and eosin to measure villus height and crypt depth, as previously described [11]. For each individual section, the length of 10 random villi and crypts was measured and the average values for each pig were used for statistics. The protocol of the animal study was approved by the Danish National Committee on Animal Experimentation. 6
ACCEPTED MANUSCRIPT 2.4. Western blot for intestinal tissues Middle small intestinal tissues of the healthy CON and LF pigs (n = 5/group), that did not
PT
develop NEC, were homogenized in an extraction buffer containing 1% Triton X-100 and 1% of protease inhibitor cocktail [15]. Thereafter, 25 µg protein of all tissue samples was resolved on a 16%
RI
SDS-PAGE gel and transferred onto PVDF membranes (Life Technologies) [15]. Membranes were
SC
then probed with primary and secondary antibodies to detect bLF (Hycult Biotech, The Netherlands),
NU
proliferating cell nuclear antigen (PCNA), apoptosis regulator Bax, B cell lymphoma 2 (Bcl-2), caspase 3 (all from Santa Cruz Biotechnology, CA, USA), and heavy chain ferritin (H-ferritin, Abcam,
MA
Cambridge, UK). Band intensity was analyzed and extracted from ImageJ (National Institute of Health, USA).
D
2.5. Statistics
TE
The relative abundance of identified proteins from cellular proteomics was imported into R
AC CE P
integrated with R Studio for statistical analysis. Abundance of each protein was fitted to linear mixedeffect models with and without LF concentration as a fixed factor and iTRAQ plex as a random factor using the ‘lmer’ function [20]. The significance of bLF treatment was tested by comparing these two models using ‘anova’ function. The difference among the groups was further tested using with a “Tukey test” using the ‘glht’ function [21]. P values were extracted from specific tests. An identified protein with an overall P value < 0.05 of the treatment and at least one significance (P < 0.05) in the post hoc comparisons was regarded as a significantly affected protein by bLF treatment. Among these proteins, only those showing fold-changes greater than 1.2 or less than 0.85 in LF 0.1, 1 or 10 compared with LF0, were used for assignment of biological function. For the preterm pig study, the villus height, crypt depth, and ratio of villus height and crypt depth were analyzed using linear mixed-effect model with nutritional treatment and litter as fixed factors (JMP, SAS Institute, NC, USA). Band densities of selected proteins by Western blot were 7
ACCEPTED MANUSCRIPT analyzed using a linear mixed model with treatment as the fix factor and a following post hoc Tukey
PT
test to compare levels in CON and LF pigs (JMP).
RI
3. Results 3.1. IEC proteomics
SC
In total, 1258 proteins were identified in at least three plexes of the proteomic study. Following
NU
statistics, bLF and 193 porcine proteins showed significant differences in abundance among the four treatment groups (P < 0.05). After applying a cut-off threshold of 1.2-fold for up-regulation and 0.85-
MA
fold for down-regulation, 122 porcine proteins and bLF were differentially expressed (63 up-regulated and 60 down-regulated proteins). Information of the identified proteins, including accession number, %
D
coverage, number of unique peptides, protein score and fold change are categorized into groups based
TE
on their major biological functions (Supplementary Table S2 online). The major protein functions
AC CE P
regulated by bLF were proliferation and apoptosis (14 proteins), energy metabolism (12 proteins), protein synthesis, processing and degradation (23 proteins), DNA and RNA binding and processing (15 proteins), cytoskeleton (13 proteins), protein transport (9 proteins), amino acid metabolism (7 proteins), and signal transduction (4 proteins).
Both SDS-PAGE and proteomics results showed that the intracellular bLF level increased after bLF stimulation, indicating cellular bLF uptake (Fig. 1A-B). The bLF uptake reached a plateau at 1 g/L. 3.2. bLF effects at high and low doses in vitro At low doses (0.1-1 g/L), bLF up-regulated 11 proteins involved in glycolysis and energy metabolism (pyruvate kinase, pyruvate dehydrogenase, pyruvate carboxylase and UDP-glucose 4-epimerase), and
8
ACCEPTED MANUSCRIPT protein synthesis and processing (Fig. 2). Negligible differences in these protein levels were found between LF0.1 and LF1.
PT
Only the high dose (10 g/L) of bLF, but not 0.1-1 g/L, altered eight proteins involved in proliferation and apoptosis (Fig. 3A). Three up-regulated proteins play a role in the early phase of
RI
apoptosis and cell death, including apoptosis inducing factor (AIF), annexin-1 and cyclophilin-40. The
SC
remaining five proteins, including two anti-apoptotic proteins (catalase and huntingtin-interacting
NU
protein 1, HIP1) and three proliferation-inducing proteins (CD63, granulins and 7-dehydrocholesterol reductase - 7DHCR) were down-regulated in LF10. bLF at 10 g/L also impaired energy metabolism by
MA
down-regulating 10 out of 11 proteins involved in the TCA cycle, electron transport and ATP synthesis (Fig. 3B).
D
bLF at 10 g/L differentially expressed five proteins involved in the iron-dependent and HIF-1
TE
pathways (Fig. 3C). HIF-1 pathway-related proteins were ubiquitin carboxyl-terminal hydrolase
AC CE P
(UCH), DNA-(apurinic or apyrimidinic site) lyase (Apex/Ref-1) and heterogeneous nuclear ribonucleoprotein a2/b1 (hnRNP a2/b1). UCH and Apex/Ref-1 were up-regulated by all bLF doses, with higher levels in LF10 than in LF0.1 and LF1. In contrast, hnRNP a2/b1 and two other proteins involved in iron binding and transport (transferrin receptor and vesicle trafficking protein 22b) were down-regulated in
LF10, relative to
other groups (Fig. 3C).
With the exception of
adenosylhomocysteinase being up-regulated by all bLF doses, the other five proteins involved in amino acid metabolism were down-regulated in LF10 relative to other groups (Fig. 3D). 3.3. Clinical evaluation and gut morphology in CON and LF pigs At euthanasia on day 5, the incidence of NEC did not differ between CON (9/15 pigs, 60%) and LF pigs (8/13 pigs, 62%), but LF pigs had more severe colonic lesions than CON pigs (higher NEC score, P < 0.05), as previously reported [11]. In the proximal small intestine, villus height in LF pigs tended to be lower than
that in CON pigs (P = 0.09, Fig. 4A), whereas crypt depth was similar between the two groups (Fig. 9
ACCEPTED MANUSCRIPT 4B). Of note, the ratio of villus height and crypt depth was significantly lower in LF vs. CON pigs (P < 0.05, Fig. 4C), suggesting a decreased intestinal epithelial cell proliferation or an increased cell death in
PT
LF pigs. Representative cross sections of the proximal small intestine reflected the shorter villi in LF vs. CON pigs (Fig. 4D-E).
RI
3.4. bLF effects on the immature pig intestine
SC
bLF was detected in the small intestine of LF pigs, but absent in CON pigs (P < 0.001, Fig. 5A),
NU
indicating bLF uptake into the intestine of LF pigs. In line with this uptake, the intestine of LF pigs had higher HIF-1α levels than those of CON pigs (P < 0.05, Fig. 5B). H-ferritin (an iron storing protein)
MA
and PCNA (a protein involved in DNA damage) did not differ between the two treatments (Fig. 5C-D). Proteins involved in apoptosis, as shown by Bax/Bcl-2 ratio, were elevated in LF vs. CON pigs (P <
4. Discussion
TE
AC CE P
the two groups (Fig. 5F).
D
0.05, Fig. 5E). In contrast, the levels of pro-caspase 3 and activated caspase 3 did not differ between
LF is an important protein in both human and bovine milk with multiple functions [2–4]. It has a great potential as a supplement for preterm infant formulas [22] to exert protection against infection in early life [6]. However, previous studies have not elucidated the mechanisms by which LF exerts its protective effects. In the present study, we examined the effects of bLF at doses that were comparable to the levels of LF in bovine and human milk using both IECs in vitro and immature pig intestine in vivo. We demonstrated the beneficial effects of bLF at low doses (0.1-1 g/L, close to LF levels in bovine and human mature milk) and detrimental effects at a high dose (10 g/L, close to LF levels in human colostrum). In the cultured IECs, bLF at this high dose mainly targeted pathways related to energy metabolism, cell survival/apoptosis, as well as the HIF-1 pathway, which is highly related to iron homeostasis and cell death. The detrimental effect of bLF at this high dose was also observed in 10
ACCEPTED MANUSCRIPT the immature small intestine of piglets receiving supplemented bLF, shown by greater colonic lesions, shorter intestinal villus height/crypt depth ratio, and an elevated Bax/Bcl-2 ratio and HIF-1α levels.
PT
LF binds to its receptor on IECs and this facilitates receptor internalization and LF uptake [4,14]. In the current study, we confirmed the uptake of bLF in both cultured IECs and in the immature
RI
pig small intestine. Compared with 1 g/L, bLF at 10 g/L did not further increase the bLF uptake,
SC
probably due to the saturation of available cell surface receptors by 1 g/L of bLF. Any unbound bLF
NU
may thereby bind to alternative membrane receptors such as TLR4 or TNF receptors [12,13,23] and trigger inflammation. For instance, we have previously shown that bLF at 10 g/L initiated NF-κB
MA
signaling leading to an increased IL-8 production in cultured porcine IECs [11]. Furthermore, as a transferrin, bLF strongly binds iron in the intestinal lumen to facilitate iron absorption via LF
D
internalization [24]. Iron is then released from bLF and transferred to mitochondria or stored in ferritin
TE
[24], causing ferritin up-regulation [25]. In the current study in preterm pigs, intestinal H-ferritin was
AC CE P
unchanged in LF pigs relative to CON pigs (Fig. 5C), suggesting an impaired iron storage following the accumulation of iron-bound LF (Fig. 5A). Because H-Ferritin is anti-apoptotic [26], the impaired up-regulation of this protein after LF challenge may contribute to the increased apoptosis in the intestine of LF pigs.
bLF at low doses (0.1-1 g/L), similar to the LF levels in bovine and human milk, up-regulated proteins involved in glycolysis, energy metabolism and protein synthesis including the important enzymes pyruvate kinase, pyruvate carboxylase and pyruvate dehydrogenase. Pyruvate kinase catalyzes the final step of the glycolytic pathway to form pyruvate, while pyruvate can be converted into oxaloacetate, an intermediate in the TCA cycle, by pyruvate carboxylase [27]. Pyruvate dehydrogenase catalyzes the reaction from pyruvate to acetyl-CoA, a key molecule in many metabolic processes [27]. The up-regulation of these enzymes suggests that low bLF doses facilitate energy production and contribute to IEC proliferation [4,11]. Among the identified proteins involved in 11
ACCEPTED MANUSCRIPT protein synthesis, three aminoacyl-tRNA ligases (glycine, isoleucine and tryptophan-tRNA ligases) and DNA lyase were also up-regulated by bLF at 0.1 and 1 g/L. These ligases belong to the tRNA
PT
synthetase family, catalyzing the ligation between amino acids and their cognate tRNA during the initial step of protein synthesis [28]. DNA lyase is a protein with dual functions: stimulation of DNA
RI
repair and facilitation of DNA binding activity of numerous transcription factors, thereby playing
SC
important roles in protein synthesis [29]. The up-regulation of these proteins may reflect the elevated
NU
protein synthesis or turnover, which is required for IEC proliferation [30]. bLF at low doses did not affect proteins associated with cell apoptosis and inflammation, agreeing with our previous study
MA
showing consistent levels of HIF-1α and IL-8 secretion following IEC stimulation with bLF at these doses [11].
D
bLF at 10 g/L altered 26 proteins towards the directions of apoptosis, impaired energy and
TE
amino acid metabolism, and activation of the HIF-1 pathway in vitro. In addition, the lower ratio of
AC CE P
villus height/crypt depth and higher colonic lesions in LF pigs imply a reduced proliferation or elevated cell death of the IECs in the small intestine (Fig. 4C). Tissue analysis further supports the suggestion of elevated bLF-induced apoptosis with a greater Bax/Bcl-2 ratio (a key regulator of apoptotic pathways) and increased HIF-1α levels (regulator of HIF-1 pathway) in LF pigs. Five apoptosis-related proteins were up-regulated by bLF at 10 g/L in vitro including annexin1, AIF, cyclophilin 40, catalase and huntingtin-interacting protein 1 (HIP1). Annexin-1 is involved in pro-apoptotic pathways [31], and it is also released from apoptotic cells and triggers cell engulfment by macrophages [32]. AIF is a protein translocated into the nucleus following apoptotic stimuli to initiate caspase-independent apoptosis [33]. Cyclophilin-40 overexpression causes mitochondrial swelling and cell death and increases cell susceptibility to oxidative stress [34,35]. Catalase is an enzyme that is upregulated when protecting cells against apoptosis after exposure to reactive oxygen species [36,37]. HIP1 maintains cellular survival, and its deficiency leads to apoptosis [38,39]. Elevated annexin-1, AIF 12
ACCEPTED MANUSCRIPT and cyclophilin 40, and decreased catalase and HIP1 levels elicited only by a bLF dose of 10 g/L reflect that this high dose may reduce cell survival and trigger apoptosis. Besides, proliferation-
PT
inducing proteins, including 7-dehydrocholesterol reductase, CD63 and granulins were down-regulated by 10 g/L of bLF. 7-dehydrocholesterol reductase is involved in cell proliferation in rats [40]. CD63 is
RI
the cell receptor of TIMP-1 required for proliferation [41]. Granulins play a role in the mediation of
SC
cell proliferation [42] and ERK activation [43]. These trends are consistent with our previous study
NU
showing a marked reduction of IEC proliferation and impaired ERK signaling by 10 g/L bLF [11]. The Bax/Bcl-2 ratio can determine whether cells undergo apoptosis (increased ratio) or
MA
survival (reduced ratio), and their role in apoptosis is partly triggered by hypoxia and impaired energy metabolism [44]. Hypoxia, which is associated with the activation of the HIF-1 pathway and NEC [45],
D
can impair energy metabolism, inhibit electron transport and reduce ATP synthesis. These events
TE
activate Bax leading to cytochome C release from the mitochondria and apoptotic signaling cascades
AC CE P
[44]. In addition, Bcl-2 can inhibit Bax activation [44]. Importantly, bLF can act as a mimetic of hypoxia to up-regulate HIF-1α and the HIF-1 pathway [46] and bLF at 10 g/L elevated HIF-1α in cultured IECs [11]. In this current study, the activation of the HIF-1 pathway induced by 10 g/L of bLF in IECs was supported by proteomic data via altered levels of ubiquitin carboxyl-terminal hydrolase, hnRNPa2/b1 and Apex/Ref-1, which all lead to increased HIF-1α stability and HIF-1 transcriptional activity [47–50]. We also showed in the present study that 10 g/L bLF down-regulated 10 proteins involved in energy metabolism in IECs, and increased colonic lesions, intestinal HIF-1α levels and Bax/Bcl-2 ratio in preterm pigs, indicating a tendency to trigger intestinal apoptosis. However, caspase 3 did not show a clear activation in LF pigs, implying that IEC apoptosis may occur via caspase 3independent pathways. We speculate that bLF supplemented into enteral formula at high doses may act to mimic hypoxic conditions and elevate the HIF-1 pathway, leading to impaired energy metabolism
13
ACCEPTED MANUSCRIPT and activation of caspase 3-independent apopotic pathways involving Bax/Bcl-2 in the immature
PT
intestine.
RI
5. Conclusions
We conclude that exogenous bLF modulates intestinal proteins in preterm pigs and regulates
SC
the IEC proteome via survival, apoptotic and inflammatory pathways. Relatively low doses (close to
NU
levels present in bovine and human mature milk) may be beneficial to stimulate intestinal maturation via induction of IEC proliferation. A higher dose, such as 10 g/L, may exert adverse and detrimental
MA
effects by triggering apoptosis and inflammation. In this study, proteomics was used to elucidate the cellular pathways affected by bLF in a dose-dependent manner. The pig experiment using only the high
D
dose was used to confirm the effects of this dose of bLF in inducing apoptotic conditions in the
TE
immature small intestine in vivo. A comprehensive preterm pig study investigating the dose-dependent
AC CE P
effects of bLF could be important to determine the optimal dose of bLF for formula supplementation in preterm infants, thereby supporting intestinal maturation and preventing inflammation.
Conflict of interest statement
All authors declare no competing financial interest.
Acknowledgements The present study was supported by a grant from the Danish Dairy Research Foundation. Dorte Thomassen and Yanqi Li are thanked for technical assistance with sample preparation and animal study. The Obelske Family Foundation, Svend Andersen Foundation and SparNord Foundation are acknowledged for their support with the analytical platform.
Author contribution 14
ACCEPTED MANUSCRIPT D.N.N, P.T.S and D.E.W.C designed the study. D.N.N., P.J., A.S., and E.B. conducted the experiments. D.N.N. and P.J. analyzed data and performed statistics. D.N.N. wrote the manuscript
PT
together with the other authors. D.E.W.C. had primary responsibility for the final content. All authors have read and approved the final manuscript. Supplementary information to this article can be found
SC
RI
online.
NU
References
[1] K. Marshall, Therapeutic applications of whey protein, Altern Med Rev. 9 (2004) 136–156.
MA
[2] E.N. Baker, H.M. Baker, A structural framework for understanding the multifunctional character of lactoferrin, Biochimie. 91 (2009) 3–10.
D
[3] D.H. Adamkin, Mother’s milk, feeding strategies, and lactoferrin to prevent necrotizing
TE
enterocolitis, J. Parenter. Enteral. Nutr. 36 (2012) 25S–9S.
AC CE P
[4] R. Jiang, V. Lopez, S.L. Kelleher, B. Lönnerdal, Apo- and holo-lactoferrin are both internalized by lactoferrin receptor via clathrin-mediated endocytosis but differentially affect ERK-signaling and cell proliferation in caco-2 cells, J. Cell. Physiol. 226 (2011) 3022–3031. [5] P. Manzoni, L. Decembrino, I. Stolfi, L. Pugni, M. Rinaldi, S. Cattani, et al., Lactoferrin and prevention of late-onset sepsis in the pre-term neonates, Early Hum. Dev. 86 (2010) 59–61. [6] P. Manzoni, M. Rinaldi, S. Cattani, L. Pugni, M.G. Romeo, H. Messner, et al., Bovine lactoferrin supplementation for prevention of late-onset sepsis in very low-birth-weight neonates: a randomized trial, JAMA. 302 (2009) 1421–1428. [7] J.-I. Togawa, H. Nagase, K. Tanaka, M. Inamori, A. Nakajima, N. Ueno, et al., Oral administration of lactoferrin reduces colitis in rats via modulation of the immune system and correction of cytokine imbalance, J. Gastroenterol. Hepatol. 17 (2002) 1291–1298.
15
ACCEPTED MANUSCRIPT [8] W.J. Lee, J.L. Farmer, M. Hilty, Y.B. Kim, The protective effects of lactoferrin feeding against endotoxin lethal shock in germfree piglets, Infect. Immun. 66 (1998) 1421–1426.
PT
[9] N. Orsi, The antimicrobial activity of lactoferrin: Current status and perspectives, Biometals. 17 (2004) 189–196.
RI
[10] B.J. Appelmelk, Y.Q. An, M. Geerts, B.G. Thijs, H.A. de Boer, D.M. MacLaren, et al.,
SC
Lactoferrin is a lipid A-binding protein, Infect. Immun. 62 (1994) 2628–2632.
NU
[11] D.N. Nguyen, Y. Li, P.T. Sangild, S.B. Bering, D.E.W. Chatterton, Effects of bovine lactoferrin on the immature porcine intestine, Br. J. Nutr. 111 (2014) 321–331.
MA
[12] S.-M. Oh, S.-H. Lee, B.-J. Lee, C.-W. Pyo, N.-K. Yoo, S.Y. Lee, et al., A distinct role of neutrophil lactoferrin in RelA/p65 phosphorylation on Ser536 by recruiting TNF receptor-
D
associated factors to IkappaB kinase signaling complex, J. Immunol. 179 (2007) 5686–5692.
TE
[13] C.S. Curran, K.P. Demick, J.M. Mansfield, Lactoferrin activates macrophages via TLR4-
AC CE P
dependent and -independent signaling pathways, Cell. Immunol. 242 (2006) 23–30. [14] B. Lönnerdal, R. Jiang, X. Du, Bovine Lactoferrin Can Be Taken Up By The Human Intestinal Lactoferrin Receptor And Exert Bioactivities, J. Pediatr. Gastroenterol. Nutr. (2011). [15] D.N. Nguyen, P. Jiang, S. Jacobsen, P.T. Sangild, E. Bendixen, D.E.W. Chatterton, Protective Effects of Transforming Growth Factor β2 in Intestinal Epithelial Cells by Regulation of Proteins Associated with Stress and Endotoxin Responses, PLoS ONE. 10 (2015) e0117608. [16] M. Danielsen, H. Hornshøj, R.H. Siggers, B.B. Jensen, A.G. van Kessel, E. Bendixen, Effects of bacterial colonization on the porcine intestinal proteome, J. Proteome Res. 6 (2007) 2596–2604. [17] T. Bennike, U. Ayturk, C.M. Haslauer, J.W. Froehlich, B.L. Proffen, O. Barnaby, et al., A normative study of the synovial fluid proteome from healthy porcine knee joints, J. Proteome Res. (2014).
16
ACCEPTED MANUSCRIPT [18] J. Cox, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, MaxLFQ allows accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio
PT
extraction, Mol. Cell. Proteomics. (2014). [19] D.E.W. Chatterton, J.T. Rasmussen, C.W. Heegaard, E.S. Sørensen, T.E. Petersen, In vitro
SC
functions, Trends Food Sci. Technol. 15 (2004) 373–383.
RI
digestion of novel milk protein ingredients for use in infant formulas: Research on biological
NU
[20] D. Bates, M. Mächler, B. Bolker, S. Walker, Fitting Linear Mixed-Effects Models using lme4, arXiv:1406.5823 [stat]. (2014). http://arxiv.org/abs/1406.5823 (accessed September 1, 2014).
MA
[21] T. Hothorn, F. Bretz, P. Westfall, Simultaneous inference in general parametric models, Biom J. 50 (2008) 346–363.
D
[22] D. Sharma, S. Shastri, Lactoferrin and neonatology - role in neonatal sepsis and necrotizing
TE
enterocolitis: present, past and future, J. Matern. Fetal. Neonatal. Med. (2015) 1–8.
AC CE P
[23] K. Ando, K. Hasegawa, K.-I. Shindo, T. Furusawa, T. Fujino, K. Kikugawa, et al., Human lactoferrin activates NF-kappaB through the Toll-like receptor 4 pathway while it interferes with the lipopolysaccharide-stimulated TLR4 signaling, FEBS J. 277 (2010) 2051–2066. [24] H.S. Birgens, L.Ø. Kristensen, N. Borregaard, H. Karle, N.E. Hansen, Lactoferrin-mediated transfer of iron to intracellular ferritin in human monocytes, Eur. J. Haematol. 41 (1988) 52–57. [25] S. Shimmura, M. Shimoyama, M. Hojo, K. Urayama, K. Tsubota, Reoxygenation injury in a cultured corneal epithelial cell line protected by the uptake of lactoferrin., Invest. Ophthalmol. Vis. Sci. 39 (1998) 1346–1351. [26] A. Cozzi, S. Levi, B. Corsi, P. Santambrogio, A. Campanella, G. Gerardi, et al., Role of iron and ferritin in TNFα-induced apoptosis in HeLa cells, FEBS Letters. 537 (2003) 187–192. [27] S.Y. Lunt, M.G. Vander Heiden, Aerobic Glycolysis: Meeting the Metabolic Requirements of Cell Proliferation, Annu Rev Cell Dev Biol. 27 (2011) 441–464. 17
ACCEPTED MANUSCRIPT [28] K. Wakasugi, B.M. Slike, J. Hood, A. Otani, K.L. Ewalt, M. Friedlander, et al., A human aminoacyl-tRNA synthetase as a regulator of angiogenesis, PNAS. 99 (2002) 173–177.
PT
[29] Y. Cun, N. Dai, M. Li, C. Xiong, Q. Zhang, J. Sui, et al., APE1/Ref-1 enhances DNA binding activity of mutant p53 in a redox-dependent manner, Oncol. Rep. 31 (2014) 901–909.
RI
[30] S.G. Park, P. Schimmel, S. Kim, Aminoacyl tRNA synthetases and their connections to disease,
SC
PNAS. 105 (2008) 11043–11049.
NU
[31] L.H.K. Lim, S. Pervaiz, Annexin 1: the new face of an old molecule, FASEB J. 21 (2007) 968– 975.
MA
[32] M. Scannell, M.B. Flanagan, A. deStefani, K.J. Wynne, G. Cagney, C. Godson, et al., Annexin-1 and peptide derivatives are released by apoptotic cells and stimulate phagocytosis of apoptotic
D
neutrophils by macrophages, J. Immunol. 178 (2007) 4595–4605.
TE
[33] N. Bidère, H.K. Lorenzo, S. Carmona, M. Laforge, F. Harper, C. Dumont, et al., Cathepsin D
AC CE P
Triggers Bax Activation, Resulting in Selective Apoptosis-inducing Factor (AIF) Relocation in T Lymphocytes Entering the Early Commitment Phase to Apoptosis, J. Biol. Chem. 278 (2003) 31401–31411.
[34] C.P. Baines, R.A. Kaiser, N.H. Purcell, N.S. Blair, H. Osinska, M.A. Hambleton, et al., Loss of cyclophilin D reveals a critical role for mitochondrial permeability transition in cell death, Nature. 434 (2005) 658–662. [35] Y. Li, N. Johnson, M. Capano, M. Edwards, M. Crompton, Cyclophilin-D promotes the mitochondrial permeability transition but has opposite effects on apoptosis and necrosis, Biochem. J. 383 (2004) 101. [36] K. Hasegawa, S. Wakino, K. Yoshioka, S. Tatematsu, Y. Hara, H. Minakuchi, et al., Sirt1 protects against oxidative stress-induced renal tubular cell apoptosis by the bidirectional regulation of catalase expression, Biochem. Biophys. Res. Commun. 372 (2008) 51–56. 18
ACCEPTED MANUSCRIPT [37] H.-U. Simon, A. Haj-Yehia, F. Levi-Schaffer, Role of reactive oxygen species (ROS) in apoptosis induction, Apoptosis. 5 (2000) 415–418.
PT
[38] D.S. Rao, T.S. Hyun, P.D. Kumar, I.F. Mizukami, M.A. Rubin, P.C. Lucas, et al., Huntingtininteracting protein 1 is overexpressed in prostate and colon cancer and is critical for cellular
RI
survival, J. Clin. Investig. 110 (2002) 351–360. doi:10.1172/JCI15529.
SC
[39] I. Dragatsis, M.S. Levine, S. Zeitlin, Inactivation of Hdh in the brain and testis results in
NU
progressive neurodegeneration and sterility in mice, Nat. Genet. 26 (2000) 300–306. [40] S. Wu, F.D. Luca, Role of Cholesterol in the Regulation of Growth Plate Chondrogenesis and
MA
Longitudinal Bone Growth, J. Biol. Chem. 279 (2004) 4642–4647. [41] Y. Xia, N. Yeddula, M. Leblanc, E. Ke, Y. Zhang, E. Oldfield, et al., Reduced cell proliferation
D
by IKK2 depletion in a mouse lung-cancer model, Nat. Cell. Biol. 14 (2012) 257–265.
TE
[42] P.C. Hanington, D.R. Barreda, M. Belosevic, A Novel Hematopoietic Granulin Induces
9970.
AC CE P
Proliferation of Goldfish (Carassius auratus L.) Macrophages, J. Biol. Chem. 281 (2006) 9963–
[43] J.Q. Feng, F.-J. Guo, B.-C. Jiang, Y. Zhang, S. Frenkel, D.-W. Wang, et al., Granulin epithelin precursor: a bone morphogenic protein 2-inducible growth factor that activates Erk1/2 signaling and JunB transcription factor in chondrogenesis, FASEB J. 24 (2010) 1879–1892. [44] A.E. Greijer, E. van der Wall, The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis, J Clin Pathol. 57 (2004) 1009–1014. [45] D.J. Hackam, A. Afrazi, M. Good, C.P. Sodhi, Innate immune signaling in the pathogenesis of necrotizing enterocolitis, Clin. Dev. Immunol. 2013 (2013) 475415. [46] E.T. Zakharova, V.A. Kostevich, A.V. Sokolov, V.B. Vasilyev, Human apo-lactoferrin as a physiological mimetic of hypoxia stabilizes hypoxia-inducible factor-1 alpha, Biometals. 25 (2012) 1247–1259. 19
ACCEPTED MANUSCRIPT [47] G.L. Semenza, Targeting HIF-1 for cancer therapy, Nat. Rev Cancer. 3 (2003) 721–732. [48] B. Seliger, A. Fedorushchenko, W. Brenner, A. Ackermann, D. Atkins, S. Hanash, et al.,
PT
Ubiquitin COOH-Terminal Hydrolase 1: A biomarker of renal cell carcinoma associated with enhanced tumor cell proliferation and migration, Clin. Cancer Res. 13 (2007) 27–37.
RI
[49] M. Fähling, R. Mrowka, A. Steege, P. Martinka, P.B. Persson, B.J. Thiele, Heterogeneous
NU
Stability, J. Biol. Chem. 281 (2006) 9279–9286.
SC
Nuclear Ribonucleoprotein-A2/B1 Modulate Collagen Prolyl 4-Hydroxylase, α (I) mRNA
[50] G. Tell, F. Quadrifoglio, C. Tiribelli, M.R. Kelley, The Many Functions of APE1/Ref-1: Not Only
AC CE P
TE
D
MA
a DNA Repair Enzyme, Antioxid. Redox. Signal. 11 (2008) 601–619.
20
ACCEPTED MANUSCRIPT Figure legends Fig. 1. Lactoferrin (LF) detected in porcine IECs by SDS-PAGE (A) and by quantification from
PT
iTRAQ-LCMS-based proteomics following treatment with LF 0, 0.1, 1 and 10 g/L for 24 h (B). * and
RI
#: P < 0.05, compared with LF0 and LF 0.1, respectively.
SC
Fig. 2. bLF at the low doses (0.1-1 g/L) up-regulated proteins involved in glycolysis, metabolism, protein synthesis and processing. * and #: P < 0.05, relative to LF0 and LF0.1, respectively .
NU
Fig. 3. Differentially expressed proteins involved in cell proliferation and cell death (A), energy
MA
metabolism (B), the HIF-1 pathway (C), and amino acid metabolism (D) induced by bLF. *, #, and §: P < 0.05, relative to LF0, LF0.1 and LF1, respectively. AIF: apoptosis inducing factor; HIP1: Huntingtin
D
interacting protein 1; 7DHCR:7-dehydrocholesterol reductase; UCH: ubiquitin carboxyl-terminal
TE
hydrolase; hnRNP a2/b1: heterogeneous nuclear ribonucleoprotein a2/b1.
AC CE P
Fig. 4. Intestinal mucosal structure in CON pigs (formula-fed, n = 13) and LF pigs (10 g/L bLFenriched formula-fed, n = 11). (A) Villus height, (B) Crypt depth, (C) Ratio of villus height/crypt depth, (D and E) Representative hematoxylin and eosin stained cross sections of the proximal intestine in CON (D) and LF pigs (E). Values are means ± SEM. * P < 0.05. Fig. 5. Western blot analysis of bLF (A) and proteins associated with apoptosis and iron-dependent pathways, including HIF-1α (B), H-ferritin (C), PCNA (D), Bax/Bcl-2 (E) and caspase 3 (F), in the middle intestinal tissues of preterm pigs fed control formula or bLF-enriched formula at 10 g/L. Values are means ± SEM. * and ** P < 0.05 and 0.01, respectively.
21
ACCEPTED MANUSCRIPT
4.51
4.51
Protein, g/L
79
79
Whey protein, g/L
79
69
Supplemented bLF, g/L
0
10
Carbohydrates, g/L
62.5
62.5
Maltodextrin, g/L
47.5
47.5
Lactose, g/L
15
Fat, g/L
57
RI
Energy, MJ/L
SC
LF
NU
CON
PT
Table 1. Macronutrient compositions of formula diets for CON and LF pigs1
15
MA
57
1
AC CE P
TE
D
Energy and macronutrient composition was calculated according to the ingredient specifications provided by the suppliers (Eurofins Steins Laboratorium, and Morinaga Industry)
22
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
Fig. 1
23
Fig. 2
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
24
Fig. 3
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
25
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Fig. 4
26
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Fig. 5
27
ACCEPTED MANUSCRIPT Biological significance: The present work elucidates dose-dependent effects of bLF on the proteomic changes of IECs in vitro supplemented with data from a preterm pig study confirming detrimental
PT
effects of enteral feeding with the highest dose of bLF (10 g/L). The study contributes to further understanding on mechanisms that bLF, as an important milk protein, can regulate the homeostasis of
RI
the immature intestine. Results from this study urge neonatologists to carefully consider the dose of
AC CE P
TE
D
MA
NU
SC
bLF to supplement into infant formula used for preterm neonates.
28
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
Graphical abstract
29
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Highlights: Lactoferrin (LF) is present in bovine and human colostrum and milk (0.1-10 g/L). Bovine LF (bLF) dose-dependently affected the intestinal cell proteome. bLF at low doses (0.1-1 g/L) regulated proteins supporting cell homeostasis. bLF at 10 g/L regulated proteins involved in apoptosis and inflammation. bLF at 10 g/L induced apoptosis and inflammation in the immature pig intestine.
30