Purpose: Pregnancy increases breast cancer risk for all women for at least 5 years after parturition. During weaning and involution, the breast microenvironment becomes tumor promotional. Exosomes provide cell–cell communication during physiologic processes such as lactation, but also in breast cancer. We determined whether molecules in milk exosomes from healthy lactating women modulate the development and progression of breast cancer.

Experimental Design: Thirteen nursing women provided three (transitional, mature, and wean) milk samples. Exosomes were extracted and MCF7 and MCF10A breast cells labeled. The expression of six proteins linked to breast cancer was measured. On the basis of the findings, TGFβ2 concentration in exosome samples was measured, breast cells incubated with the exosomes and effect (epithelial–mesenchymal transition, EMT) on EMT-related proteins [E-cadherin, α-smooth muscle actin (α-SMA), filamentous (F)-actin and vimentin] measured.

Results: Human milk exosomes entered benign and malignant breast cells. The greatest change in wean milk protein was in TGFβ2 (P = 0.01). Exosomes with a high (but not low) level of TGFβ2 led to EMT in both cancer and benign cells, based on (i) change in cell morphology, actin cytoskeleton, and loss of cell–cell junction structure and (ii) increased α-SMA and vimentin and decreased E-cadherin.

Conclusions: TGFβ2 is significantly upregulated in breast milk exosomes during weaning/early involution. Breast milk exosomes containing high levels of TGFβ2 induce changes in both benign and malignant breast epithelial cells, consistent with the development and progression of breast cancer, suggesting a role for high TGFβ2-expressing breast milk exosomes in influencing breast cancer risk. Clin Cancer Res; 22(17); 4517–24. ©2016 AACR.

Translational Relevance

Pregnancy-associated breast cancer is an aggressive form of the disease that is difficult to both diagnose and treat for which there is no effective screening tool. Relatively little is known regarding the molecular drivers of the disease, especially in humans. We report that TGFβ2 levels in milk exosomes significantly increase during breast involution, a time when the breast microenvironment is tumor promotional. We further demonstrate that milk samples with high (but not low) levels of TGFβ2 appear to drive both breast cancer and benign breast cells toward an aggressive phenotype, more prone to tumor invasion and metastasis.

Pregnancy increases breast cancer risk (at least for the short term)

Pregnancy-associated breast cancers (PABCs) diagnosed after childbirth are generally aggressive, with a poor prognosis. Moreover, pregnancy has a lasting influence on breast cancer risk, beyond the time of PABCs, which is variably defined as lasting anywhere from 1 to up to 10 years after childbirth. Epidemiologic data suggest that pregnancy increases a woman's risk of developing breast cancer compared with nulliparous women over the short term, but that the length in years of this increased risk depends on the age at which the pregnancy occurs. Our understanding of the human biology of PABCs is very limited. Our ultimate goal is to identify biologic markers (biomarkers) that will help predict which parous women are most likely to develop breast cancer.

Immediately following parturition, there is an increased risk of breast cancer observed for all age groups (1). This is not surprising, as the growth factors required to allow breast glandular proliferation, as well as the extensive remodeling required during involution, may stimulate the growth of already present neoplastic breast epithelial cell(s). The stimulation of extant neoplastic cells versus their inhibition is a key concept in whether pregnancy and lactation decrease or increase breast cancer risk. Over the long term, parity is protective for women whose first full-term pregnancy (FFTP) was completed at a young age (variously defined as < 20 up to age 26), and increased in parous women whose FFTP occurred after 35 years of age (2). The time of weaning, a period of breast involution and remodeling, appears critical to future breast cancer risk (3).

There is good evidence that during involution, which is initiated by weaning, the breast microenvironment becomes tumor promotional (3). A variety of proteins are important to the involution process. Milk fat globule EGF factor 8 (MFG-E8) or lactadherin, which is secreted from exosomes (discussed below) fulfills a tethering function between apoptotic cells. MFG-E8 is involved in the recognition and clearance of apoptotic mammary epithelial cells during mammary involution (4). Matrix metalloproteinase (MMP)-dependent reorganization of the extracellular matrix during breast involution releases growth factors, supporting breast metastasis (5). TGFβ has both positive and negative effects on tumorigenesis (6). In mouse models, TGFβ is induced during mammary gland lactation and involution (7), and it mediates proapoptotic effects during involution (1). On the other hand, TGFβ promotes tumor growth through activation of the stroma and it stimulates proliferation of cells in the healing wound (8). TGFβ is linked to breast cancer in preclinical models (7) and to prognosis in human breast cancer (9). TGFβ-containing exosomes from mesothelioma cells and breast cancer cells converted fibroblasts to a myofibroblast phenotype (10).

Milk exosomes

Exosomes are cellular organelles that are shed into the surrounding extracellular environment (11). They contain membrane and cytoplasmic components of the cell. While some exosomes are quickly degraded, others are not and can travel by diffusion in body fluids such as breast milk. Exosomes are known to contain a wide variety of proteins, lipids, and RNA (12). Exosomes provide cell–cell communication during physiologic processes such as lactation, but also in breast cancer (12). Exosomes have been shown to be critical for directional migration of cancer cells (13). While exosome isolation and analysis from human blood has been widely reported, there are few reports on human breast milk exosomes (one focused on how to isolate the exosomes [14] and a second on exosome populations related to allergic sensitization [15]).

Epithelial–mesenchymal transition

During epithelial–mesenchymal transition (EMT), an epithelial cell loses polarity and cell–cell contacts, acquiring migratory and invasive properties (16). EMT occurs both in normal developmental processes including mesoderm formation and neural tube formation, in wound healing, and in cancer progression. TGFβ is known to induce EMT in cancer, can promote tumor invasion and is associated with breast cancer bone metastasis (17). The role of TGFβ in cancer development is more complex, as TGFβ can also lead to cellular apoptosis (6).

Epithelial cell–cell junctions (subapical tight junctions, adherens junctions and desmosomes at lateral surfaces, and scattered gap junctions at lateral surfaces) maintain cell–cell contact (18). EMT requires restructuring of these junctions and of the actin cytoskeleton. During the destabilization of adherens junctions, epithelial (E)-cadherin is cleaved and then degraded (19). There is also upregulation of α-smooth muscle actin (SMA) and reorganization of filamentous (F)-actin (20). This reorganization leads to increased migratory and invasive capabilities (20). Proteins such as vimentin, which promote cell migration and invasion, are also upregulated (20).

Summary

During breast involution, the tumor promotional microenvironment may drive epithelial cells in the breast toward cancer development or progression. We investigated whether breast milk exosomes contained proteins involved in breast involution, which of the detected proteins was increased at the time of involution, whether these protein(s) change benign cells toward a cancer phenotype, and/or make cancer cells more aggressive. We chose two immortalized breast cell lines for study, one derived from a patient with cancer and the second from a patient with benign breast disease. Both have epithelial characteristics. MCF7 cells are derived from pleural metastases of a lobular breast carcinoma. The cells require estrogen supplementation or genetic manipulation for tumor formation in immunodeficient mice (21). MCF10A cells are derived from a woman with benign fibrocystic changes. The cells are immortal, but nontumorigenic unless genetically modified (22). We observed that breast milk exosomes from healthy lactating women containing high levels of TGFβ2 induced changes in benign epithelial cells consistent with transformation toward neoplasia, and changes in breast cancer cells toward a more aggressive, invasive phenotype.

Preparation of milk samples

A total of 38 matched milk samples (one wean sample missing) were collected from 13 lactating women (transitional: within 10 days of lactation start; mature milk: collected 2 months after lactation start, and wean (involution) milk: collected once breastfeedings were significantly curtailed at the end of nursing) after Institutional Review Board informed consent was obtained. The women ranged in age from 25 to 37 years. The samples were separated by centrifugation into three components: fat, supernatant, and cell pellet, snap frozen, labeled, and placed in a −80°C freezer until analysis.

Isolation and resuspension of exosomes

The ExoQuick eExosome Isolation Kit (System Biosciences) was used as per the manufacturer's instructions. A total of 200 μL of milk supernatant, PBS, and exosome isolation reagent added, incubated at room temperature, centrifuged, and the excess supernatant removed. The samples were centrifuged a second time to remove additional supernatant. A total of 500 μL of PBS was then added to each pellet, and the tube vortexed overnight at 4°C to fully dissolve the pellet. After dissolution, the sample was centrifuged, the supernatant aliquoted, and the remaining pellet discarded. The remaining specimen contained exosomes for analysis.

Analysis of total milk protein and milk exosome protein

A Bicinchoninic Acid Kit (Life Technologies) was used to measure total milk protein and milk exosome protein. After adding standard and sample to each well, reagent was added, the plate mixed on a plate shaker, the plate incubated and absorbance read at 562 nm on a plate reader.

Exosome labeling; coculture of breast milk exosomes and breast cells

Exosomes isolated from milk supernatant were labeled using an Exo-Green Exosome Labeling Kit as per the manufacturer's instructions (System Biosciences). MCF7 and MCF10A cells we separately cocultured with breast milk exosomes containing high (1.5 ng/mL) or low (0.06 ng/mL) levels of TGFβ2 for 48 hours. Cells were cultured in exosome-depleted specific FBS medium (System Biosciences) as recommended. The cells were imaged using a confocal microscope to confirm the presence of exosomes within the cells.

ELISA

The expression of 6 proteins (MMP2, MMP3, MMP9, TGFβ1, TGFβ2, and MFG-E8) in breast milk exosomes was assessed in duplicate by ELISA. A TGFβ2 ELISA Kit (R&D Systems) was used, following the manufacturer's instructions. An MMP9 ELISA Kit was purchased from Abcam. MMP2, MMP3, TGFβ1, and MFG-E8 ELISA kits were purchased from R&D Systems. The milk exosome pellet was resuspended in buffer, an assay diluent added, then standard, control, or activated sample added to each well. After incubation, the cells were washed, conjugate added, followed by washing and the addition of substrate. A second incubation for 20 minutes was followed by adding stop solution. Absorbance was assessed with a plate reader at 450 nm.

EMT

MCF7 and MCF10A cells were obtained from the ATCC. The MCF7 line was derived from a woman with breast cancer, the MCF10A line from a woman with benign breast disease. The cells were washed twice with FBS, cell number and viability assessed using trypan blue, and the cells resuspended to a final concentration of 105 cells/mL in FBS-containing culture media. 1.5 × 105 cells were suspended into the wells of a 6-well tissue culture plate, incubated for 72h-MCF7 or 96h-MCF10A (to obtain similar plate confluence) then milk exosomes containing a high (1.5 ng/mL) or low (0.06 ng/mL) concentration of TGFβ2 added, with or without 1.5 μg TGFβ2-blocking antibody (R&D Systems). Cell morphology was then visualized by inverted light microscopy.

Immufluorescence was performed on MCF10A cells before and after treatment. Both negative (PBS-treated cells) and positive (pharmacologic treatment of cells with 10 ng/mL TGFβ2) controls were included for each protein evaluated. Cells were treated with 200 μg breast milk exosomes containing low (0.06 ng/mL) or high (1.5 ng/mL) levels of TGFβ2. The following EMT-related proteins were evaluated: E-cadherin, α-SMA, F-actin, and vimentin. Cells cultured on glass coverslips were fixed in 4% formaldehyde, washed twice with PBS, permeabilized with 0.1% Triton X-100, washed again and blocked with 1% BSA. Cells were then incubated at 4°C overnight with the indicated primary antibodies (E-cadherin: Cell Signaling Technology; α-SMA: R&D Systems; vimentin: Sigma) and then incubated for 1 hour with fluorescent secondary antibody (Invitrogen), Alexa 647 phalloidin for F-actin (Life Technologies) and 4′,6-diamidino-2-phenylindole (DAPI) for nuclear staining (Sigma). Fluorescence images were viewed with a Leica TCS SP8 laser-scanning confocal microscope (Leica Microsystems).

Quantitative real-time (RT) PCR analysis was performed in duplicate to measure the expression of E-cadherin after treatment of MCF10A cells with negative and positive controls, as well as low or high levels of TGFβ2. Briefly, total RNA was extracted from MCF10A cells with an RNeasy Plus Micro Kit (Qiagen). Total RNA (800 ng) was reverse transcribed to cDNA with a QuantiTect Reverse Transcription Kit (Qiagen). RT-PCR reactions were performed in duplicate using an IQ SYBR Green Supermix (Bio-Rad) according to the manufacturer's instructions using a MyiQ RT-PCR detection system (Bio-Rad). Relative changes in gene expression were calculated using the 2−ΔΔCt formula. GAPDH mRNA levels served as the internal control. The E-cadherin and GAPDH primers were purchased from Qiagen.

Statistical analysis

Expression levels of the six proteins (Table 1) were summarized using their mean and median level at each of the time points. Post-baseline levels were compared with baseline using a paired t test for statistical significance. As the sample size was small, no test for normality was performed. All analyses were performed using statistical software R version 3.0.2.

Table 1.

Protein expression in breast milk exosomes

Median (Mean) milk concentration (pg/mg)PP
ProteinTransitionalWholeWeanTrans. vs. WholeTrans. vs. Whole
TGFβ1 80.2 (76.4) 47.6 (48.8) 83.0 (86.0) 0.01 0.54 
TGFβ2 503.5 (474.2) 187.0 (295.8) 1,096.0 (1,883.0) 0.17 0.01 
MMP2 0.48 (0.49) 0.34 (0.40) 0.27 (0.29) 0.16 <0.01 
MMP3 0.31 (0.33) 0.12 (0.13) 0.14 (0.18) <0.01 0.05 
MMP9 65.0 (167.7) 28.3 (143.5) 44.4 (92.3) 0.77 0.24 
MFG-E8 1,288.0 (1,537.0) 1,114.0 (1,028.0) 1,001.0 (1,147.0) 0.05 0.15 
Median (Mean) milk concentration (pg/mg)PP
ProteinTransitionalWholeWeanTrans. vs. WholeTrans. vs. Whole
TGFβ1 80.2 (76.4) 47.6 (48.8) 83.0 (86.0) 0.01 0.54 
TGFβ2 503.5 (474.2) 187.0 (295.8) 1,096.0 (1,883.0) 0.17 0.01 
MMP2 0.48 (0.49) 0.34 (0.40) 0.27 (0.29) 0.16 <0.01 
MMP3 0.31 (0.33) 0.12 (0.13) 0.14 (0.18) <0.01 0.05 
MMP9 65.0 (167.7) 28.3 (143.5) 44.4 (92.3) 0.77 0.24 
MFG-E8 1,288.0 (1,537.0) 1,114.0 (1,028.0) 1,001.0 (1,147.0) 0.05 0.15 

NOTE: Comparisons with a P value < or = 0.05 are in bold.

Abbreviations: MMP, matrix metalloproteinase; MFG, milk fat globulin.

Milk exosomes from healthy lactating women enter breast cancer and benign breast cells

We used confocal microscopy to confirm that, when cocultured, human milk exosomes get inside of both MCF7 and MCF10A breast cells (Fig. 1). Confocal imaging demonstrated GFP binding to CD63, a protein commonly used to identify exosomes. Proliferation was measured in MCF7 cells and noted to be significantly increased (P < 0.01) after treatment with milk exosomes containing a high level of TGFβ2 versus control.

Figure 1.

Confocal image of MCF7 (A–C) and MCF10a (D–F) cells cocultured with milk exosomes demonstrating GFP binding to CD63, a protein used to identify exosomes. A and D, without GFP; B and E, GFP with CD63 staining; C and F, merged image. Arrow indicates internalized exosome. Scale bar, 25 μm.

Figure 1.

Confocal image of MCF7 (A–C) and MCF10a (D–F) cells cocultured with milk exosomes demonstrating GFP binding to CD63, a protein used to identify exosomes. A and D, without GFP; B and E, GFP with CD63 staining; C and F, merged image. Arrow indicates internalized exosome. Scale bar, 25 μm.

Close modal

TGFβ2 is significantly upregulated in wean milk exosomes

We were able to detect each of the six proteins in the exosomes of all milk samples. For each protein, we compared expression in transitional milk to milk collected later (in mature milk or wean milk). TGFβ2 expression in the 38 samples ranged from 0.55 to 53.2 ng/mL, with a mean of 9.4 ng/mL. We considered high expression to be above the mean, and low expression below the mean. In all cases, median and mean protein expression decreased in mature milk. This decrease was statistically significant for TGFβ1, MMP3, and MFG-E8 (Table 1; Fig. 2). Protein expression in wean milk significantly changed in three of the proteins, increasing for TGFβ2 (P = 0.01) and decreasing for MMP2 and MMP3. The median (mean) increase in TGFβ2 from transitional to wean milk was 2.2-fold (4-fold), and from mature to wean milk 5.9-fold (6.4-fold). There was no correlation (P = 0.46) between nursing length and TGFβ2 expression.

Figure 2.

Expression of TGFβ1, TGFβ2, MMP2, MMP3, MMP9, and MFG-E8 in breast milk exosomes. Expression was measured in matched transitional, mature and wean milk samples from nursing mothers. The box plots span the first to the third quartile, with the bold line in the middle representing the median. The extended lines beyond the box show the 95% confidence region and points beyond that the outlier results. Asterisks indicate a significant difference (P < 0.05) between expression in whole or wean milk versus transitional milk.

Figure 2.

Expression of TGFβ1, TGFβ2, MMP2, MMP3, MMP9, and MFG-E8 in breast milk exosomes. Expression was measured in matched transitional, mature and wean milk samples from nursing mothers. The box plots span the first to the third quartile, with the bold line in the middle representing the median. The extended lines beyond the box show the 95% confidence region and points beyond that the outlier results. Asterisks indicate a significant difference (P < 0.05) between expression in whole or wean milk versus transitional milk.

Close modal

Milk exosomes promote breast cancer through EMT

We determined the dose-dependent role of TGFβ2 containing milk exosomes on EMT. Milk exosomes containing high or low concentrations of TGFβ2 or PBS control were added to MCF7 (Fig. 3) or MCF10A (Fig. 4) cells. Exosomes with high TGFβ2 (but not low TGFβ2) demonstrated morphologic changes consistent with EMT, including disruption of cell–cell junctions with cellular extensions (, Fig. 3A and B, right). A TGFβ2-blocking antibody reversed the EMT (Fig. 3B, left). Pharmacologic TGFβ2 (10 ng/mL) also induced EMT (Fig. 4).

Figure 3.

EMT and filopodia formation in MCF7 cells after 72-hour coincubation with milk exosomes expressing high (H) but not low (L) TGFβ2 or PBS (control; top row). Reversal of EMT and filopodia formation following treatment with a TGFβ2 neutralizing antibody (Nab; bottom, left two panels). Magnification 200× (bottom, right panel). Close-up of filopodia formation (arrows). Magnification 200×.

Figure 3.

EMT and filopodia formation in MCF7 cells after 72-hour coincubation with milk exosomes expressing high (H) but not low (L) TGFβ2 or PBS (control; top row). Reversal of EMT and filopodia formation following treatment with a TGFβ2 neutralizing antibody (Nab; bottom, left two panels). Magnification 200× (bottom, right panel). Close-up of filopodia formation (arrows). Magnification 200×.

Close modal
Figure 4.

MCF10A cells were coincubated for 96 hours with PBS (control), 10 ng/mL TGFβ2 or milk exosomes with low or high TGFβ2 levels. Magnification 200×.

Figure 4.

MCF10A cells were coincubated for 96 hours with PBS (control), 10 ng/mL TGFβ2 or milk exosomes with low or high TGFβ2 levels. Magnification 200×.

Close modal

Effect of TGFβ2-containing milk exosomes on EMT-related proteins

The actin cytoskeletal structure dramatically changes during EMT, which is correlated with a migration phenotype of the transformed cells. During this transition, the cells lose cell–cell junctions. MCF10A cells show a nonmigratory phenotype that is accompanied with the expression of E-cadherin, a key adherens junction component, at the cell–cell interface and low expression of α-SMA and vimentin (Neg controls, Fig. 5A and B). TGFβ2 stimulation markedly changed the actin cytoskeletal structure showing stress fibers and cell–cell junction structures with the loss of E-cadherin (Pos control, Fig. 5A and B). During this transition, there is a notable increase in α-SMA and vimentin expression. A similar actin cytoskeletal change and loss of E-cadherin expression were evident after high (TGFβH) but not low (TGFβL) TGFβ2 exosome treatment.

Figure 5.

EMT of MCF10A cells after the addition of milk exosomes. MCF10A cells were stimulated with PBS (Neg control), pharmacologic TGFβ2 (Pos control), or milk exosomes containing low (L) or high (H) levels of TGFβ2. During EMT, E-cadherin demonstrates decreased expression, α-SMA and vimentin were upregulated and F-actin indicated more stress fibers and an asymmetric cell structure. Merge figures demonstrate nuclear DAPI staining with membranous E-cadherin (A) or vimentin (B). Scale bar, 25 μm.

Figure 5.

EMT of MCF10A cells after the addition of milk exosomes. MCF10A cells were stimulated with PBS (Neg control), pharmacologic TGFβ2 (Pos control), or milk exosomes containing low (L) or high (H) levels of TGFβ2. During EMT, E-cadherin demonstrates decreased expression, α-SMA and vimentin were upregulated and F-actin indicated more stress fibers and an asymmetric cell structure. Merge figures demonstrate nuclear DAPI staining with membranous E-cadherin (A) or vimentin (B). Scale bar, 25 μm.

Close modal

We quantified the average change in E-cadherin after each treatment. Compared with control, there was a > 60% decrease in expression in the positive control and in the TGFβ2H-treated group, whereas the decrease after treatment with TGFβ2L was 21% (Supplementary Fig. S1).

Pregnancy is known to influence future breast cancer risk, and PABC generally portends a poor prognosis. Current assessment of how pregnancy influences future breast cancer risk is focused on clinical factors during the period between menarch and FFTP, including length in years, diet, adiposity, physical activity, and alcohol intake (23). While these factors provide insight into risk, they are insufficiently informative at the individual patient risk level to clearly guide a pregnant or lactating woman's cancer prevention efforts.

In preclinical studies, Schedin and colleagues reported that tumors developing in an involuting mammary gland were larger, greater in number and had a higher proliferation index than tumors that developed in a nulliparous mammary gland (24, 25). Our prior work demonstrated that of 16 proteins linked to breast cancer which we analyzed in lactating women, the greatest changes, regardless of age at FFTP, were with kallikreins (KLK) 6, 8, and TGFβ2 (26). Among KLKs and TGFβ isoforms, the greatest increase in wean milk supernatant collected at the time of breast involution was in TGFβ2. Among women whose breast cancer was detected while they were nursing, we found that TGFβ2 expression was significantly higher in breast milk collected from the cancer containing compared with the matched clinically normal breast (27).

In the current study, we demonstrate that milk exosomes from healthy lactating women can be reliably isolated, that proteins involved in breast involution can be measured in exosomes, and that the exosomes enter human breast cancer and benign breast cells. The decrease in exosomal protein expression in mature (compared with transitional) milk is consistent with our prior reports in breast milk supernatant, even after controlling for total milk protein (28). Among the six proteins analyzed in the current study, only one, TGFβ2, significantly increased in wean compared with transitional milk. Given our observations on TGFβ2 expression, that the protein increases both in milk supernatant and in milk exosomes at the time of breast involution, as well as the fact that expression is significantly higher in the milk of cancer containing than matched clinically normal breasts, we elected to focus our mechanistic studies on this protein.

We observed that TGFβ2-containing exosomes increase the proliferation of breast cancer cells as well as dose-dependent EMT in both breast cancer cells and cells derived from a woman with benign disease, with a marked change in cytoskeletal reorganization. Exogenously added chemical sources of TGFβ2 have previously been shown to induce EMT, and we confirmed this change in MCF10A breast cells; however, little is known regarding if TGFβ2 from human body fluids, and specifically from breast milk exosomes, drives EMT (29).

The ability of TGFβ isoforms to induce EMT appears cell line–dependent. While some cell lines such as MCF10A undergo EMT when TGFβ is used alone (30), other cell lines such as MCF7 require additional reagents (30, 31). Consistent with these reports, we observed that pharmacologic TGFβ2 induced EMT in MCF10A cells (Fig. 3), whereas it did not in MCF7 cells (data not shown). Moreover, TGFβ2 induced changes in EMT-related proteins in MCF10A-treated cells, whereas it did not in MCF7 cells (data not shown).

Both pharmacologic TGFβ2 and high-dose TGFβ2 decreased E-cadherin (Fig. 5A) and increased α-SMA, actin stress fibers and vimentin (Fig. 5B), the latter three of which are increased with EMT. The dose of TGFβ2 in milk exosomes (1.5 ng/mL) was less than pharmacologic TGFβ2 (10 ng/mL). There is evidence of a dose–reponse, perhaps best observed comparing pharmacologic treatment to high-dose TGFβ2 milk exosome treatment for cell morphology (Fig. 4) and F-actin structure (Fig. 5A and B). This is consistent with the observation that the induction of EMT, and change in protein expression with EMT, is both cell line- and dose-dependent.

Our findings suggest that breast milk exosomes containing high levels of TGFβ2 may promote EMT. Our ultimate goal is to identify one or more biomarkers of breast cancer risk in parous women by noninvasively analyzing a readily available body fluid, breast milk.

No potential conflicts of interest were disclosed.

Conception and design: S. Dasgupta, M. Ikebe, E.R. Sauter

Development of methodology: W. Qin, S. Dasgupta, N. Mukhopadhyay, E.R. Sauter

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Tsukasaki, S. Dasgupta, M. Ikebe, E.R. Sauter

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Y. Tsukasaki, S. Dasgupta, N. Mukhopadhyay

Writing, review, and/or revision of the manuscript: S. Dasgupta, N. Mukhopadhyay, M. Ikebe, E.R. Sauter

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): W. Qin, S. Dasgupta

Study supervision: S. Dasgupta, M. Ikebe, E.R. Sauter

This project was funded by Avon Foundation for Women (grant no. 02-2012-090).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Flanders
KC
,
Wakefield
LM
. 
Transforming growth factor-(beta)s and mammary gland involution; functional roles and implications for cancer progression
.
J Mammary Gland Biol Neoplasia
2009
;
14
:
131
44
.
2.
Medina
D
. 
Breast cancer: the protective effect of pregnancy
.
Clin Cancer Res
2004
;
10
:
380S
4S
.
3.
Schedin
P
,
O'Brien
J
,
Rudolph
M
,
Stein
T
,
Borges
V
. 
Microenvironment of the involuting mammary gland mediates mammary cancer progression
.
J Mammary Gland Biol Neoplasia
2007
;
12
:
71
82
.
4.
Nakatani
H
,
Aoki
N
,
Nakagawa
Y
,
Jin-No
S
,
Aoyama
K
,
Oshima
K
, et al
Weaning-induced expression of a milk-fat globule protein, MFG-E8, in mouse mammary glands, as demonstrated by the analyses of its mRNA, protein and phosphatidylserine-binding activity
.
Biochem J
2006
;
395
:
21
30
.
5.
Schedin
P
. 
Pregnancy-associated breast cancer and metastasis
.
Nat Rev Cancer
2006
;
6
:
281
91
.
6.
Wakefield
LM
,
Roberts
AB
. 
TGF-beta signaling: positive and negative effects on tumorigenesis
.
Curr Opin Genet Dev
2002
;
12
:
22
9
.
7.
Bierie
B
,
Gorska
AE
,
Stover
DG
,
Moses
HL
. 
TGF-beta promotes cell death and suppresses lactation during the second stage of mammary involution
.
J Cell Physiol
2009
;
219
:
57
68
.
8.
Lyons
TR
,
Schedin
PJ
,
Borges
VF
. 
Pregnancy and breast cancer: when they collide
.
J Mammary Gland Biol Neoplasia
2009
;
14
:
87
98
.
9.
Ghellal
A
,
Li
C
,
Hayes
M
,
Byrne
G
,
Bundred
N
,
Kumar
S
. 
Prognostic significance of TGF beta 1 and TGF beta 3 in human breast carcinoma
.
Anticancer Res
2000
;
20
:
4413
8
.
10.
Webber
J
,
Steadman
R
,
Mason
MD
,
Tabi
Z
,
Clayton
A
. 
Cancer exosomes trigger fibroblast to myofibroblast differentiation
.
Cancer Res
2010
;
70
:
9621
30
.
11.
Mathivanan
S
,
Ji
H
,
Simpson
RJ
. 
Exosomes: extracellular organelles important in intercellular communication
.
J Proteomics
2010
;
73
:
1907
20
.
12.
Hendrix
A
,
Hume
AN
. 
Exosome signaling in mammary gland development and cancer
.
The Int J Dev Biol
2011
;
55
:
879
87
.
13.
Sung
BH
,
Ketova
T
,
Hoshino
D
,
Zijlstra
A
,
Weaver
AM
. 
Directional cell movement through tissues is controlled by exosome secretion
.
Nat Commun
2015
;
6
:
7164
.
14.
Zonneveld
MI
,
Brisson
AR
,
van Herwijnen
MJ
,
Tan
S
,
van de Lest
CH
,
Redegeld
FA
, et al
Recovery of extracellular vesicles from human breast milk is influenced by sample collection and vesicle isolation procedures
.
J Extracell Vesicles
2014
Aug 14;
3
. doi: 10.3402/jev.v3.24215.
15.
Torregrosa
P
,
Gutzeit
C
,
Johansson
S
,
Admyre
C
,
Stenius
F
,
Alm
J
, et al
Differences in exosome populations in human breast milk in relation to allergic sensitization and lifestyle
.
Allergy
2014
;
69
:
463
71
.
16.
Beach
JR
,
Hussey
GS
,
Miller
TE
,
Chaudhury
A
,
Patel
P
,
Monslow
J
, et al
Myosin II isoform switching mediates invasiveness after TGF-beta-induced epithelial-mesenchymal transition
.
Proc Natl Acad Sci U S A
2011
;
108
:
17991
6
.
17.
Kang
Y
,
He
W
,
Tulley
S
,
Gupta
GP
,
Serganova
I
,
Chen
CR
, et al
Breast cancer bone metastasis mediated by the Smad tumor suppressor pathway
.
Proc Natl Acad Sci U S A
2005
;
102
:
13909
14
.
18.
Huang
RY
,
Guilford
P
,
Thiery
JP
. 
Early events in cell adhesion and polarity during epithelial-mesenchymal transition
.
J Cell Sci
2012
;
125
:
4417
22
.
19.
Yilmaz
M
,
Christofori
G
. 
EMT, the cytoskeleton, and cancer cell invasion
.
Cancer Metastasis Rev
2009
;
28
:
15
33
.
20.
Haynes
J
,
Srivastava
J
,
Madson
N
,
Wittman
T
,
Barber
DL
. 
Dynamic actin remodeling during epithelial-mesenchymal transition depends on increased moesin expression
.
Mol Biol Cell
2011
;
22
:
4750
66
.
21.
Sommers
CL
,
Papageorge
A
,
Wilding
G
,
Gelmann
EP
. 
Growth properties and tumorigenesis of MCF-7 cells transfected with isogenic mutants of rasH
.
Cancer Res
1990
;
50
:
67
71
.
22.
Zientek-Targosz
H
,
Kunnev
D
,
Hawthorn
L
,
Venkov
M
,
Matsui
S
,
Cheney
RT
, et al
Transformation of MCF-10A cells by random mutagenesis with frameshift mutagen ICR191: a model for identifying candidate breast-tumor suppressors
.
Mol Cancer
2008
;
Jun 5
;
7
:
51
.
23.
Colditz
GA
,
Bohlke
K
,
Berkey
CS
. 
Breast cancer risk accumulation starts early: prevention must also
.
Breast Cancer Res Treat
2014
;
145
:
567
79
.
24.
Lyons
TR
,
O'Brien
J
,
Borges
VF
,
Conklin
MW
,
Keely
PJ
,
Eliceiri
KW
, et al
Postpartum mammary gland involution drives progression of ductal carcinoma in situ through collagen and COX-2
.
Nat Med
2011
;
17
:
1109
15
.
25.
O'Brien
J
,
Lyons
T
,
Monks
J
,
Lucia
MS
,
Wilson
RS
,
Hines
L
, et al
Alternatively activated macrophages and collagen remodeling characterize the postpartum involuting mammary gland across species
.
Am J Pathol
2010
;
176
:
1241
55
.
26.
Qin
W
,
Zhang
K
,
Kliethermes
B
,
Ruhlen
RL
,
Browne
EP
,
Arcaro
KF
, et al
Differential expression of cancer associated proteins in breast milk based on age at first full term pregnancy
.
BMC Cancer
2012
;
12
:
100
.
27.
Arcaro
KF
,
Browne
EP
,
Qin
W
,
Zhang
K
,
Anderton
DL
,
Sauter
ER
. 
Differential expression of cancer-related proteins in paired breast milk samples from women with breast cancer
.
J Hum Lact
2012
;
28
:
543
6
.
28.
Qin
W
,
Zhang
K
,
Kliethermes
B
,
Amjad
R
,
Clarke
K
,
Sauter
ER
. 
Differential expression of cancer-associated proteins in breastmilk
.
Breastfeed Med
2013
;
8
:
120
6
.
29.
Vervoort
SJ
,
Lourenco
AR
,
van Boxtel
R
,
Coffer
PJ
. 
SOX4 mediates TGF-beta-induced expression of mesenchymal markers during mammary cell epithelial to mesenchymal transition
.
PLoS One
2013
;
8
:
e53238
.
30.
Tang
Y
,
Herr
G
,
Johnson
W
,
Resnik
E
,
Aho
J
. 
Induction and analysis of epithelial to mesenchymal transition
.
J Vis Exp
2013
Aug 27;(
78
). doi: 10.3791/50478.
31.
Walsh
LA
,
Damjanovski
S
. 
IGF-1 increases invasive potential of MCF7 breast cancer cells and induces activation of latent TGF-B1 resulting in epithelial to mesenchymal transition
.
Cell Commun Sig
2011
;
9
:
11
.