The bioactive and anti-inflammatory role of human milk components has been recognized; active milk components include soluble forms of Toll-like receptors (TLRs). Preterm babies are more susceptible to infections and may succumb to necrotizing enterocolitis (NEC), a gastrointestinal disease which is exacerbated by an excessive inflammatory response after TLR activation. Here, we investigated the presence of Toll-like receptors TLR1/2/4/6 in colostrum and mature milk of women who delivered before (preterm) or after (term) 37 weeks of gestational age, integrating classical immune-related techniques with proteomic LC-MS/MS analysis. We have detected immunoreactivity for TLRs mostly in preterm samples, even for TLR1 and TLR6, until now not described in human milk. We demonstrated the presence of only TLR2 in the milk fat globule membrane, while the immunoreactivity of TLR1/4/6 was ascribed to crossreaction with some interesting milk proteins sharing leucine-rich repeat domains. These results will provide new insights into the definition of the role of TLRs in intestinal immune regulation of the newborns.
Milk is the first food of mammals, providing them nutrients but also protection via immunoglobulins and other immune-related molecules. Milk composition is extremely dynamic, changing its content in nutrient and bioactive factors through the lactation stages (colostrum-transition-mature milk) in order to fulfil the growth needs of the newborn [
Babies born before 37 weeks of gestation are defined as “preterm” and present a multifactorial syndrome due to their limited organ development at birth. Preterm babies are more susceptible to infections and may succumb to necrotizing enterocolitis (NEC), a gastrointestinal disease which is exacerbated by an excessive inflammatory response after TLR activation. In this study, we have investigated the presence of TLR1/2/4/6 in breast milk, using both immunodetection techniques and proteomic LC-MS/MS analysis. It has been reported that immunochemical methods applied to a complex substrate such as milk are of concern due to crossreactivity or nonspecific antibody recognition [
In this study, we investigated the presence of immune-related proteins such as TLRs in colostrum (0) and mature milk (2) of women who delivered before (preterm) or after (term) 37 weeks of gestation. Human milk samples used in this study were provided by Azienda Ospedaliera Nazionale SS. Antonio e Biagio e Cesare Arrigo, Alessandria, and the Milk Bank of Ospedale Sant’Andrea, Vercelli, Italy. The samples (5-10 mL) were not pooled, in order to keep a trace of biological variability, added with a protease inhibitor cocktail (cOmplete, Mini; Roche), aliquoted in 1.5 mL tubes, and stored at -80°C until use.
Milk aliquots were centrifuged at
Protein concentration of each fraction was quantified by the method of Bradford [
The fractions of MFGM and skimmed milk were analyzed by SDS-PAGE and immunoblotting against TLR1, TLR2, TLR4, and TLR6. The immunoreactivity to
Proteins from skimmed milk and MFGMs were analyzed by SDS-PAGE. The samples (20
After SDS-electrophoresis, proteins were transferred onto a polyvinylidene fluoride (PVDF) membrane in a Mini Trans-Blot cell (Bio-Rad, CA, USA) at a constant voltage of 100 V on ice for about 90 min. The membranes were incubated for one hour at room temperature with a blocking solution of PBS/BSA 5%+NaN3, then washed for 5 min in Tris-buffered saline (TBS: 150 mM NaCl, 10 mM Tris-HCl (pH 7.4)), to remove the exceeding BSA. Multiple replicates were run in order to probe different antibodies on the same sample, as reported in Table
List of antibodies used in this study.
Antibody | Source | Producer | Dilution |
---|---|---|---|
Anti-TLR1 | Polyclonal rabbit antibody, directed against a recombinant peptide from human TLR1 (STJ25862) | St John’s Laboratory | 1 : 500 |
Anti-TLR2 | Polyclonal rabbit antibody (H-175), directed against aa 180-354 of human TLR2 (sc-10739) | Santa Cruz Biotechnology, Inc. | 1 : 500 |
Anti-TLR4 | Polyclonal rabbit antibody (H-80), directed against aa 242-321 of human TLR4 (sc-10741) | Santa Cruz Biotechnology, Inc. | 1 : 500 |
Anti-TLR6 | Polyclonal rabbit antibody directed against a peptide of 13 aa near the central part of human TLR6 (PRS3653) | Sigma-Aldrich | 1 : 1000 |
Anti- |
Polyclonal rabbit antibody (20536-1-AP) | Proteintech | 1 : 2000 |
Anti-rabbit-AP | Polyclonal goat anti-rabbit antibody conjugated with alkaline phosphatase (A3687) | Sigma-Aldrich | 1 : 3000 |
In order to identify soluble components of skimmed milk bound to anti-TLR1/TLR2/TLR4/TLR6 Abs, colostral skimmed fractions from healthy preterm donors were analyzed with a modified ELISA as described in [
Before mass spectrometry analysis, protein bands separated on SDS-PAGE, immunoreactive bands detected on PVDF membranes, and eluted fractions from the modified ELISA were submitted to trypsin digestion. Proteins from SDS-gels were in-gel digested as described in [
Immunoreactive protein bands on PVDF membranes were excised, thoroughly washed with HPLC water, then washed once with 50 mM NH4HCO3, and submitted to reducing-alkylating steps as described for in-gel digestion. After two washes in 50 mM NH4HCO3, protein bands were covered with trypsin solution and incubated overnight at 37°C. Trypsin-digested peptides were collected as described above. In the case of fainter bands, 4-5 replicates of the same band were digested and pooled in order to increase the amount of protein to be analyzed.
Eluates from the modified ELISA were reduced, alkylated, covered with trypsin solution as described above, and incubated overnight at 37°C. The digested samples were finally dried in Concentrator Plus (Eppendorf, Germany) and stored at -20°C until mass spectrometry analyses.
The peptide digests were desalted on the Discovery® DSC-18 solid phase extraction (SPE) 96-well plate (25 mg/well) (Sigma-Aldrich Inc., St. Louis, MO, USA) prior to the mass spectrometry analysis. The LC-MS/MS analyses were performed with a micro-LC Eksigent Technologies (Dublin, USA) system that included a micro LC200 Eksigent pump with flow module 5-50
The LC system was interfaced with a 5600+ TripleTOF™ system (AB Sciex, Concord, Canada) equipped with DuoSpray™ Ion Source and CDS (Calibrant Delivery System). The mass spectrometer worked in a data-dependent acquisition mode (DDA). Peptide profiling was performed using a mass range of 100–1300 Da (TOF scan with an accumulation time of 100.0 ms), followed by a MS/MS product ion scan from 200 to 1250 Da (accumulation time of 5.0 ms) with the abundance threshold set at 30 cps (35 candidate ions can be monitored per cycle). The ion source parameters in an electrospray positive mode were set as follows: curtain gas (N2) at 25 psig, nebulizer gases GAS1 at 25 psig and GAS2 at 20 psig, ionspray floating voltage (ISFV) at 5000 V, source temperature at 450°C, and declustering potential at 25 V.
The DDA files were searched using Mascot v. 2.4 (Matrix Science Inc., Boston, USA). Trypsin as a digestion enzyme was specified with 2 missed cleavages. The instrument was set to ESI-QUAD-TOF, and the following modifications were specified for the search: carbamidomethyl cysteines as fixed modification and oxidized methionine as variable modification. A search tolerance of 50 ppm was specified for the peptide mass tolerance, and 0.1 Da for the MS/MS tolerance. The charges of the peptides to search for were set to 2+, 3+, and 4+, and the search was set on monoisotopic mass. The UniProt Swiss-Prot
Sequences of TLR1/2/4/6 mature forms were aligned with Clustal Omega [
Sequences of TLR1, TLR2, TLR4, and TLR6 mature forms, preceded by UniProt identifier, initial and final amino acid number, and estimated molecular weight.
The protein profile of MFGM and skimmed milk (SM) fractions after SDS-PAGE is shown in Figures
Electrophoretic profile of MFGM proteins in colostrum and mature milk. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample.
Electrophoretic profile of skimmed milk proteins in colostrum and mature milk. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample.
MFGM and skimmed milk fractions show a different and characteristic protein profile. Though protein profiles of samples from the same milk fraction are quite similar, it is possible to observe little individual variability, with protein bands of different intensities. The most notable difference is observed between samples of different gestational ages; the colostrum of donors who delivered at term shows the lack of a band related to casein, with regard to preterm samples. The absence of casein in the colostrum of early production is reported in literature [
Immunoblots of MFGMs obtained from colostrum and mature milk samples using anti-TLR1/TLR2/TLR4/TLR6 are shown in Figures
Immunoblot of TLR1/2/4/6 in MFGMs of colostrum. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample. Standard molecular weights are indicated on the left. Each box shows the blot reactivity to a specific anti-TLR antibody and, on the lower side, the same blot incubated with an anti-
Immunoblot of TLR1/2/4/6 in MFGMs of mature milk. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample. Standard molecular weights are indicated on the left. Each box shows the blot reactivity to a specific anti-TLR antibody and, on the lower side, the same blot incubated with an anti-
Immunoblot analysis of mature milk MFGM proteins confirmed anti-TLR1 reactivity in both term and preterm samples at 150 kDa, but if compared with colostrum, the band at 100 kDa disappeared while a new band more evident in preterm samples came out at 30 kDa. Two bands detectable in all samples appeared at 100 and 80 kDa after incubation with an anti-TLR2 antibody, while anti-TLR4 reactivity was observed at 75 kDa (see also Supplementary Figure
Semiquantitative analysis of immunoreactive bands in MFGMs (data not shown), obtained after actin standardization, confirmed the differences observed between term and preterm mainly in colostral samples. Immunoblotting results show changes in the amount of actin; anyway, we used actin normalization assuming that the presence of actin in milk originates from cell residues, since actin is not a typical milk secreted protein and could represent proportionally the amount of cells from which TLRs are derived.
Colostrum of preterm donors shows higher immunoreactivity to anti-TLR1 Ab than that of term women, with two intense bands at 150 and 100 kDa. The band at 100 kDa disappeared in mature milk, while a new band at 30 kDa appeared, with higher intensity in preterm samples. Concerning colostrum, we could observe anti-TLR2 reactivity only in preterm samples, while two reactive bands at 100 and 80 kDa were detectable in each sample of mature milk tested. Interestingly, we could not observe any anti-TLR4 reactivity in colostral samples, while a band appeared in each sample of mature milk. Preterm colostral samples showed higher reactivity, at 75 and 50 kDa, to anti-TLR6 Ab too. The intensity values of these bands declined in mature milk.
Immunoblots of skimmed fractions from colostrum and mature milk samples using the anti-TLR1/TLR2/TLR4/TLR6 antibodies are shown in Figures
Immunoblot of TLR1/4/6 in colostral skimmed milk. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample. Standard molecular weights are indicated on the left. Each box shows the blot reactivity to a specific anti-TLR antibody and, on the lower side, the same blot incubated with anti-
Immunoblot of TLR1/6 in mature skimmed milk. Gestational age is indicated as “T” (term) or “PT” (preterm). Each number refers to a different milk sample. Standard molecular weights are indicated on the left. Each box shows the blot reactivity to a specific anti-TLR antibody and, on the lower side, the same blot incubated with anti-
Concerning colostrum skimmed fraction, two bands around 75 kDa appeared after anti-TLR1 Ab incubation, but if compared with MFGM fraction, no bands at 150-100 kDa were observed. Anti-TLR2 and anti-TLR4 Abs determined the appearance of two bands at 75 and 50 kDa (Supplementary Figure
A 150 kDa band appeared in all samples except in 7 T after anti-TLR1 Ab incubation. As observed in colostrum, we did not detect signals relative to TLR2 and TLR4 but only aspecific bands at 75 and 50 kDa (Supplementary Figure
Immunoreactive bands against anti-TLR1 and anti-TLR4 Abs showed higher intensity in preterm than in term samples of colostrum. This trend is no more evident in mature samples, where these bands disappeared and a clear band at 150 kDa appeared only after anti-TLR1 Ab incubation. Concerning anti-TLR6 immunoreactivity, bands of higher intensity were observed in colostrum than in mature skimmed milk. Again, colostrum preterm samples revealed higher intensity bands at 75-100 kDa. This trend is no more evident in lower MW bands of both colostrum and mature skimmed milk.
The observed differences between preterm and term samples are supported by evidences reported in literature. The concentration of total protein is higher in preterm milk [
Concerning TLRs, the presence of bands with apparent molecular weight lower than that of TLR mature forms (which is estimated between 87 kDa and 93 kDa considering the amino acidic sequence only) could be due to posttranslational modifications, with the production of truncated forms. In literature, there is some evidence of TLR2 and TLR4 soluble forms (sTLRs) revealed by Western blot analysis of biological samples such as saliva, amniotic fluid, plasma, and milk [
Mass spectrometry analysis of immunoreactive bands detected in MFGM and skimmed milk fractions (Figure
Representative image of proteins from MFGM (a) and skimmed milk (b) fractions of colostrum (0) and/or mature milk (2) after SDS-PAGE separation. The name of bands analyzed by mass spectrometry is reported beside each lane. Bands labelled with “B” were digested from PVDF blots, while “G” bands were digested from polyacrylamide gels. S1 was digested from polyacrylamide gels, and S2, S3, S4, S5, S6, S7, S8, S9, and S10 were digested from PVDF blots.
TLR2 was detected by mass spectrometry in the MFGM fraction only, with peptides of six different sequences. In order to assess if TLR2 found in the MFGM fraction corresponds to the whole receptor or parts of it, the sequences of the identified peptides were overlapped to the complete sequence of TLR2. As shown in Figure
Protein sequence of TLR2. The signal peptide is highlighted in green, while identified peptides after mass spectrometry are highlighted in yellow.
Our MS results are supported by studies reporting the finding of TLR2 in MFGMs but not in serum of milk [
It is interesting to note that our MS analysis of immunoreactive bands detected some TLR-related proteins such as CD14 and CD36 or LRR-containing proteins. Cluster of differentiation 14 (CD14) is a pattern recognition receptor with a bent solenoid structure typical of leucine-rich repeat proteins [
Identified proteins in skimmed milk include zinc
Leucine-rich repeats and immunoglobulin-like domains protein 1 (LRIG1) is another LRR-containing protein found in both MFGM and skimmed fractions of milk. LRIG1 is a membrane protein with a series of LRRs, three immunoglobulin-like (Ig-like) domains, one transmembrane domain, and a cytosolic tail. It is a negative regulator of signaling by receptor tyrosine kinases. LRIG1 interacts with EGFR/ERBB1, ERBB2, ERBB3, and ERBB4 family receptors through the LRR and Ig-like extracellular domains [
Finally, tenascin, a glycoprotein of the extracellular matrix implicated in neuronal and axonal migration during development, was identified in the MFGM and skimmed fractions. It contains EGF-like and fibronectin type III domains and can exist in homoexameric or homotrimeric forms. Tenascin is involved in various physiological processes such as cell adhesion, cellular response to vitamin D and retinoic acid, extracellular matrix organization, and positive regulation of cell proliferation.
Its expression is induced by TGF-
Since TLR2 was detected only in MFGM, a further test on skimmed milk fractions was performed with the modified ELISA, in order to selectively detect proteins interacting with the anti-TLRs used in this study. We limited the assay to colostrum of preterm women, since after Western blotting analyses, we observed more intense immunoreactive profiles in these samples. After anti-TLR1 Ab incubation, we detected lactotransferrin and phosphatidylinositol 4,5-bisphosphate 5-phosphatase A, a protein with two isoforms of 100 and 70 kDa (in the WB, we observed a 75 kDa band). Anti-TLR2 incubation revealed again the presence of lactotransferrin and Rho GTPase-activating protein 7 (in the WB, a 75 kDa band was observed). Anti-TLR4 interacted with zinc
MFGM fraction from preterm colostrum showed higher reactivity to anti-TLR1, anti-TLR2, and anti-TLR6 antibodies, compared to that from term samples. In the skimmed fraction, no immunoreactivity to anti-TLR2 was observed, whereas bands at low molecular weight appeared after incubation with anti-TLR4 and anti-TLR6 antibodies. LC-MS/MS analysis of immunoreactive bands confirmed the presence of the entire membrane receptor TLR2 in the MFGM of human milk, but not in the skimmed milk. With regard to TLR1, TLR4, and TLR6, despite the observation of immunoreactive bands, their presence was not confirmed after MS analysis. However, it is interesting to note that the MS analysis of these bands allowed the identification of proteins related to TLRs, because of the presence of LRR domains in their sequence or their direct interaction with TLRs. MS analysis of the ELISA fractions did not reveal any TLR but confirmed proteins identified in immunoreactive bands. The presence of soluble forms of TLRs in human milk remains an open question. It is possible that the soluble forms of TLRs in milk reported in literature, which were observed only by immunoblot assays, are related to crossreactivity events with these proteins. Anyway, the lack of MS identification for TLR1, TLR4, and TLR6 cannot assure their total absence in milk, where their concentration may be inferior to the limit of detection of the instrument and the DDA method used in this study could have excluded them. It is clear that TLR2 is surely present in milk, and we suggest that soluble forms may be ascribed to other milk proteins sharing conformational domains capable of reacting with and modulating TLR signaling.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that they have no competing interest.
This research was supported by the University of Piemonte Orientale (bando ateneo ricerca).
Table S1: list of proteins identified after mass spectrometry such as TLRs or TLR-related proteins. Figure S1: complete Western blot with anti-TLR2 for MFGM colostrum. Figure S2: complete Western blot with anti-TLR2 for MFGM mature milk. Figure S3: complete Western blot with anti-TLR4 for MFGM colostrum. Figure S4: complete Western blot with anti-TLR4 for MFGM mature milk. Figure S5: complete Western blot with anti-TLR2 for skimmed colostrum. Figure S6: complete Western blot with anti-TLR2 for skimmed mature milk. Figure S7: complete Western blot with anti-TLR4 for skimmed colostrum. Figure S8: complete Western blot with anti-TLR4 for skimmed mature milk.