Semaphorin 7A Promotes Macrophage-Mediated Lymphatic Remodeling During Postpartum Mammary Gland Involution and in Breast Cancer Alan M

Published OnlineFirst September 25, 2018; DOI: 10.1158/0008-5472.CAN-18-1642

Cancer Tumor Biology and Immunology Research

Semaphorin 7A Promotes Macrophage-Mediated Lymphatic Remodeling during Postpartum Mammary Gland Involution and in Breast Cancer Alan M. Elder1,2, Beth A.J. Tamburini3, Lyndsey S. Crump1,2, Sarah A. Black1,2, Veronica M. Wessells1,2, Pepper J. Schedin2,4, Virginia F. Borges1,2, and Traci R. Lyons1,2

Abstract

Postpartum mammary gland involution is a tissue remo- mammary tissue from SEMA7A-knockout mice exhibited deling event that occurs in all mammals in the absence of decreased myeloid-derived PDPN-expressing cells, PDPN- nursing or after weaning to return the gland to the pre-preg- expressing endothelial cells, and lymphatic vessel density. nant state. The tissue microenvironment created by involution Furthermore, coexpression of SEMA7A, PDPN, and macro- has proven to be tumor promotional. Here we report that the phage marker CD68 predicted for decreased distant metastasis- GPI-linked protein semaphorin 7A (SEMA7A) is expressed free survival in a cohort of over 600 cases of breast cancer as on mammary epithelial cells during involution and use pre- well as in ovarian, lung, and gastric cancers. Together, our clinical models to demonstrate that tumors induced during results indicate that SEMA7A may orchestrate macrophage- involution express high levels of SEMA7A. Overexpression of mediated lymphatic vessel remodeling, which in turn drives SEMA7A promoted the presence of myeloid-derived podopla- metastasis in breast cancer. nin (PDPN)-expressing cells in the tumor microenvironment and during involution. SEMA7A drove the expression of PDPN Signﬁcance: SEMA7A, which is expressed on mammary in macrophages, which led to integrin- and PDPN-dependent cells during glandular involution, alters macrophage biology motility and adherence to lymphatic endothelial cells to and lymphangiogenesis to drive breast cancer metastasis. promote lymphangiogenesis. In support of this mechanism, Cancer Res; 78(22); 6473–85. 2018 AACR.

Introduction SEMA7A has been described as tumor promotional. Specifically, SEMA7A drives tumor progression via promotion of tumor Semaphorin 7A (SEMA7A), or CDw108, is an 80-kDa glyco- growth, invasiveness, adhesion to tumor-associated extracellular sylphosphatidylinositol (GPI)-anchored membrane glycoprotein matrices, epithelial to mesenchymal transition, and metastasis in originally identified as preferentially expressed on activated T models of melanoma, glioblastoma, and oral and breast carci- lymphocytes, erythrocytes, and in some leukemia cell lines (1). noma (6–9). In addition, tumor-derived SEMA7A can alter the In addition, SEMA7A mRNA expression is observed in some adult breast tumor microenvironment by increasing the presence of tissues such as placenta, testis, lung, brain, thymus, and spleen (1). macrophages with pro-angiogenic phenotypes and by increasing Roles for SEMA7A have been identified in neuronal guidance and tumor-associated lymphatics (7, 8). Overall, data to date suggest immune modulation (2–4). The SEMA7A protein can exist as a that SEMA7A expression is sufficient to alter both tumor cell membrane bound or shed form after cleavage of the GPI mem- biology and the tumor microenvironment. brane anchor; once shed SEMA7A can act in both cell autonomous SEMA7A also has defined roles in tissue remodeling during and nonautonomous manners. As such, SEMA7A has been shown fibrosis (4); in the mammary gland, postpartum breast/mammary to promote motility of melanocytes, neuronal, and immune cells gland involution is a fibrosis-like remodeling event that is essen- through its signaling via b1 integrin (2, 3, 5). More recently, tial for returning the mammary tissue to a pre-pregnant-like state after pregnancy and/or lactation in mammals. In the absence of proper coordination of postpartum involution, multiple rounds 1Department of Medicine, Division of Medical Oncology, University of Colorado of lactation could not occur in mammals. Although the majority Anschutz Medical Campus, Aurora, Colorado. 2University of Colorado Cancer of studies investigating postpartum mammary gland involution Center Young Women's Breast Cancer Translational Program, Aurora, Colorado. have occurred in rodents, recent confirmation in human tissues 3Division of Gastroenterology, University of Colorado School of Medicine, 4 suggest that mechanisms of involution appear to be conserved Aurora, Colorado. Department of Cell, Development and Cancer Biology, amongst species (10–12). Postpartum mammary involution is Oregon Health Sciences University, Oregon. characterized by epithelial cell apoptosis, immune cell infiltra- Note: Supplementary data for this article are available at Cancer Research tion, extracellular matrix (ECM), and vascular remodeling, and Online (http://cancerres.aacrjournals.org/). alterations to immune cell infiltrates into the tissue microenvi- Corresponding Author: Traci R. Lyons, University of Colorado Anschutz Medical ronment (11, 13–20). Macrophages play an essential part in the Campus, MS 8117, 12801 East 17th Avenue, Aurora, CO 80045. Phone: 303-724- involution program as demonstrated by data showing that mac- 3885; Fax: 303-724-3889; E-mail: [email protected] rophage ablation blocks the execution of weaning-induced mam- doi: 10.1158/0008-5472.CAN-18-1642 mary epithelial cell death in mice (11). We have identified that 2018 American Association for Cancer Research. programs of lymphangiogenesis are activated during involution

www.aacrjournals.org 6473

Elder et al.

and that peak macrophage infiltration (involution day 3) pre- and bred as previously described (10). Briefly, Sema7atm1Alk/J (a cedes peak lymph vessel density (LVD; involution day 6) in mouse generous gift from Alex Kolodkin at John's Hopkins University) mammary tissues during involution (11, 21). Roles for macro- and C57Bl/6 female mice (age 6–8 weeks) were bred with BALB/c phages in lymphangiogenesis have been described previously males and BALB/c female mice were bred with C57Bl/6 male mice. (22–30). In addition, SEMA7A stimulates monocyte, neutrophil, Age-matched nulliparous animals were set aside as controls. and macrophage chemotaxis (7, 31, 32) as well as tumor asso- Mammary gland involution was initiated in the bred females by ciated lymphangiogenesis (8). Here, we have sought to investigate force-weaning pups at 10 to 14 days post-parturition. For normal whether macrophages and SEMA7A cooperate to promote mam- involution studies, tissues were obtained from nulliparous and mary lymphangiogenesis during involution as well as lymphan- involution (Inv 1, 2, 3, 4, 6, 8, and 28) mice. Rat tissues analyzed giogenesis in mouse mammary tumor models. were from a previously described study (11). For the 66cl4 tumor Our results show that SEMA7A is expressed in the mammary studies, 200,000 66cl4-DDK or 66cl4-DDK SEMA7A-overexpres- epithelium during postpartum involution and suggest that sing cells were injected into the number 4 left and right mammary SEMA7A may be involved in macrophage-mediated lymphangio- glands of BALB/c dams 1 day after forced weaning (involution genesis, or lymphatic mimicry by macrophages, during normal group) or into age-matched nulliparous control females. For the development and in breast cancer. Using preclinical models, we E0771 tumor studies, 250,000 E0771-DDK or E0771-DDK show that subsets of monocytes and macrophages express the SEMA7A-overexpressing cells were injected into the number 4 lymphatic marker podoplanin (PDPN) in a SEMA7A-dependent left and right mammary glands of C57Bl/6 dams 1 day after forced manner, which SEMA7A promotes macrophage motility, adher- weaning (involution group) or into age-matched nulliparous ence to, and incorporation into lymphatic-like structures in vitro, control females. Tumor study mice were sacrificed based on and that PDPN-expressing macrophages isolated from mouse primary tumor cell growth or ulceration at 3 to 4 weeks postin- mammary glands during involution are sufficient to drive lym- jection (66cl4) or 2 to 3 weeks postinjection (E0771). In vivo phangiogenesis. We also show that SEMA7A is significantly asso- studies were performed in triplicate with representative or pooled ciated with decreased survival in patients with lymph node- data shown. positive breast cancer. In addition, we show that a gene signature composed of SEMA7A, the macrophage marker CD68, and PDPN Matrigel plug assay is associated with poor prognosis in breast, ovarian, lung, and To generate macrophages for the Matrigel plug assay, BALB/c gastric cancer—indicating that this triad may predict for aggressive mice were bred (described above) and mammary tissues har- tumors in multiple tumor types. Together our results suggest that vested and digested to single cells (described below); macro- SEMA7A-dependent macrophage-driven lymphangiogenesis may phages were isolated by F4/80 magnetic bead selection. Equal be an important driver of breast tumor metastasis via the lymph numbers of mammary macrophages harvested from nullipa- vasculature. Furthermore, we identify a novel, developmentally- rous and involution groups were mixed with phenol red free regulated mechanism of inflammation-induced lymphangiogen- Matrigel (Corning; #356231) and injected 1:1 with 66cl4 esis that may inform studies beyond those of mammary devel- tumor cells into intact mammary fat pads of nulliparous opment and breast cancer. BALB/c mice. Eight days later, the mammary gland with Matri- gel plug intact was harvested and fixed in 10% NBF. Sections were obtained and analyzed for F480þLYVE-1þ structures by Materials and Methods immunofluorescence. Cell culture Authenticated 66cl4-luciferase cells were acquired from H. Ford Tumor cell clonogenic assay (University of Colorado Anschutz Medical Campus, Denver, CO). Lung and axillary lymph node metastases were analyzed by Authenticated E0771 cells were acquired from CH3 BioSystems. clonogenic assays as previously described (10). Briefly, axillary Cells were cultured as previously described. 66cl4-luciferase and lymph nodes and lungs were harvested from 66cl4 tumor- E0771 cells were transfected with plasmids containing DDK-tagged containing mice (from above). Excised tissues were minced and Sema7a or a DDK empty vector control using X-tremeGENE trans- digested with a final concentration of 500 units/mL collagenase fection reagent (Sigma Aldrich; Catalog No. #XTG9-RO). Trans- type II and IV and 20 mg/mL DNase (Worthington Biochemical fected cells were selected using G418 (neomycin and concentration Corporation). Tissues were digested at 37C for 90 minutes, is 0.2 mg/ml). Human dermal lymphatic endothelial cells (HDLEC) washed, plated in complete media plus 6-thioguanine, and left were purchased from PromoCell and cultured on plates coated with to grow for 2 weeks when the plates were washed, stained with 10 mg/mL Matrigel. All other HDLEC culturing was performed crystal violet, and colonies quantitated. according to manufacturer's instructions. J774 and RAW264.7 macrophages were obtained from the laboratory of Raphael Immunohistochemistry Nemenoff and cultured as previously described (11). All cells were IHC, immunofluorescence, and quantitation for SEMA7A, kept at low passage, tested for mycoplasma following thawing from CD68, PDPN, LYVE-1, and F4/80 was performed as previously liquid nitrogen using the Lonza MycoAlert Detection Kit, and described (8, 10, 11, 33). allowed to grow for up to ten passages before new cells were thawed. Flow cytometry Animal studies Cell populations from mammary glands and tumors were All animal procedures were approved by the University of assessed using flow cytometry. Tissues and tumors were digested Colorado Anschutz Medical Campus Institutional Animal Care using 500 units/mL collagenase type II and IV and 20 mg/mL and Use Committee. Sema7atm1Alk/J mice (Jackson Laboratories), DNase and incubated at 37C for 90 minutes prior to filtering C57Bl/6, and BALB/c (Charles River Laboratories) were housed through a 100 mm filter. For the endothelial/epithelial panel, cells

6474 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

were stained with CD45 (BV510; Biolegend, #103137), CD31 software, and adherent cell area was calculated using ImageJ (Pacific Blue; Biolegend, #102422), PDPN (APC; Biolegend, software. Experiments were replicated in triplicate with represen- #127410), CD49f (PerCP; Biolegend, #313617), EpCAM (APC- tative data shown. Cy7; Biolegend, #118218), and SEMA7A (FITC; Abcam, #26012). For the macrophage panel, cells were stained with CD45 (BV510), Macrophage expression of podoplanin CD11b (PerCP; Biolegend, #101229), F4/80 (APC-Cy7; Biole- Macrophages were plated in six-well plates with complete gend, #123117), CD64 (Pacific Blue; Biolegend, #139309), media 5 mg/mL SEMA7A and incubated at 37C for 24, 48, or MerTK (FITC; Biolegend, #151503), and PDPN (APC). Cells were 72 hours. Cells were harvested and RNA was isolated using Zymo analyzed on a Beckman Coulter CyAn ADP using Summit soft- Quick RNA Microprep Kit (Zymo Research; Catalog No. #R1050). ware in the UCCC Flow Cytometry Core or the Tamburini cDNA was synthesized from 1,000 ng RNA using the Bio-Rad Laboratory at the CU Anschutz Medical Campus. Data were cDNA Synthesis Kit and cDNA was amplified using primers for analyzed using FlowJo analysis software. reference gene b-actin (Actb; forward: TTGCTGACAGGATGCA- GAAGGAGA, reverse: ACTCCTGCTTGCTGATCCACATCT) and Lymphangiogenesis Pdpn (forward:GGGACAGGATAGGGCAATAAG, reverse: GGA- In a 96-well plate, 4 mg/mL Matrigel pads, containing 5 mg/mL GAGATGGTTCAGTGGTTAG). qPCR was performed using Bio- recombinant SEMA7A (purified by the UCCC Tissue Culture Rad SYBR Green Supermix with the following conditions: 95C Core) and/or 0.6 mg/mL anti-b1-integrin (BD Biosciences, clone: for 3 minutes, 40 cycles of 95C for 15 seconds, 60C for 30 9EG7, #553715), pMAB (ThermoFischer #MA5-16270), and IgG seconds, and 72C for 30 seconds. (BD Biosciences, clone: R35-95, #553927) were allowed to solid- ify for 2 hours at 37C prior to cell seeding. After 2 hours, 15,000 Human tissue acquisition HDLECs and 15,000 macrophages [stained green with CellTracker Research using de-identified human breast tissue was con- (Invitrogen, #C2925)] were seeded on the tops of the Matrigel ducted under a protocol deemed exempt from subject consent pads in 50% endothelial cell media and 50% experimental as approved by the Colorado Multiple Institution Review Board medias (containing identical compounds and concentrations as (COMIRB) and tissues were acquired by Virginia Borges as pre- above). Macrophages used for experiments in Matrigel plug assays viously reported (46). Dr. Borges obtained written informed were isolated from nulliparous and involution mammary glands consent from the patients, the studies were conducted in accor- using F4/80 magnetic bead separation. J774 and RAW264.7 dance with recognized ethical guidelines (e.g., Declaration of macrophages used in other experiments were pretreated overnight Helsinki, CIOMS, Belmont Report, U.S. Common Rule), and the with either 5 mg/mL SEMA7A, 0.6 mg/mL anti-b1-integrin, 0.6 mg/mL studies were approved by an institutional review board. pMAB, or 0.6 mg/mL IgG. After cells were seeded on Matrigel pads, images were taken 4 hours later using Zeiss Zen software, Human tissue data and the surface areas of tubule formations were calculated using Analysis of SEMA7A expression in histologically normal tissues ImageJ software. Experiments were replicated in triplicate with from breast biopsy or surgical specimens was performed by representative or pooled data shown. quantitative IHC, described below. All normal tissues examined were greater than 1 mm from adjacent tumor and determined to Macrophage migration assay be histologically normal. Macrophage migration was assessed using a transwell-filter migration assay. RAW264.7 macrophages were pretreated over- Aperio software analysis for quantification of IHC night with either 5 mg/mL SEMA7A, 0.6 mg/mL anti-b1-integrin, Entire histologic sections were imaged with Aperio ScanCope 0.6 mg/mL pMAB, or 0.6 mg/mL IgG. The following day, treated T3 scanner at 0.47 mm/pixel. The images were then down-sampled macrophages were seeded into a collagen-coated transwell filter to a resolution of 1.5 mm per pixel to facilitate subsequent image in serum-free media placed into a well containing serum-free manipulation. Areas for quantification were annotated using media þ 5 mg/mL SEMA7A. After 4 hours, the media was Aperio analysis tools and percent weak, medium, and strong was removed and filters were fixed using 10% neutral-buffered determined using the color-deconvolution tool. Percent positive formalin. Following fixation, filters were stained with crystal was calculated as total medium þ strong IHC positive signal violet, images were taken using Zeiss Zen software, and stained divided by total stained area multiplied by 100. Lymphatic vessel surface area of migrating cells was calculated using ImageJ density was quantitated as previously described (10). software. Experiments were replicated in triplicate with representative data shown. Kaplan–Meier survival statistical analysis Survival statistics were performed using KmPlot analysis on Macrophage endothelial cell adhesion assay cohorts of tissues. The Kaplan–Meier plotter is capable of asses- Ability of macrophages to adhere to endothelial cells was sing the effect of 54,675 genes on survival using 10,461 cancer assessed using an endothelial cell adhesion assay. HDLECs were samples. These include 5,143 breast, 1,816 ovarian, 2,437 lung, plated in a 24-well plate coated with 10 mg/mL Matrigel and and 1,065 patients with gastric cancer with a mean follow-up of allowed to form a monolayer overnight. At the same time, 69/40/49/33 months. RAW264.7 macrophages were pretreated overnight with either 5 mg/mL SEMA7A, 0.6 mg/mL anti-b1-integrin, 0.6 mg/mL pMAB, Statistical analysis or 0.6 mg/mL IgG. The following day, treated macrophages were Unpaired t-test and one-way ANOVA were performed in harvested and stained using CellTracker and seeded on top of the GraphPad Prism, assuming independent samples and normal HDLEC monolayer. After 4 hours, the media was removed, wells distributions. Only P values of less than 0.05 were considered were washed 3 with 1 PBS, images were taken using Zeiss Zen significant.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6475

Elder et al.

Results during involution, prior to the increase in LVD, in both rat and Mammary macrophages and LEC marker expression mouse tissues (Supplementary Fig. S1A–S1D). In addition, using Lymphatic vessel endothelial hyaluronan receptor (LYVE-1) is a multicolor IHC stain for lymphatics (podoplanin, PDPN) and highly expressed on LEC and has been utilized as a marker for macrophages (CD68) in human breast tissues, we observed þ þ newly formed lymphatic vessels in mouse tissues (34, 35). Acti- structures composed of CD68 and PDPN cells from women vated macrophages can express LYVE-1 in certain contexts (35, 36) biopsied during involution (Fig. 1D). and tumor-associated macrophages (TAM) express LYVE-1 (37, To further describe mammary macrophage-lymphatic charac- 38). To identify whether mammary macrophages express LYVE-1 teristics during involution, we performed flow cytometry analysis and whether LYVE-1-expressing macrophages are associated with on mouse mammary tissues harvested from nulliparous mice and the lymphatic vasculature during involution, we performed dual mice at involution day 6. We used the general immune cell marker immunofluorescence for LYVE-1 and F4/80—a marker of murine CD45, endothelial cell adhesion molecule (PECAM-1, or CD31), þ þ macrophages. We observe that LYVE-1 F4/80 single cells are epithelial cell adhesion molecule (EpCAM), and PDPN. Consis- present in nulliparous tissues and that lymphatic-like structures tent with our previous results, we observed increased LECs þ þ þ þ express both LYVE-1 F4/80 in mammary tissues at involution (CD45 EpCAM CD31 PDPN ) at involution day 6 in compar- day 6 (Fig. 1A), which is the peak of lymphatic vessel density ison to nulliparous (Fig. 2A). Also consistent with a previous þ þ (LVD) during involution (10). High resolution analysis and 3D report, we observed an increase in CD45 CD11b leukocytes re-construction of these structures revealed the presence of both during involution (Fig. 2B; ref. 17). We also observed a þ þ F4/80 and LYVE-1 expressing cells within the structure (Fig. 1B). CD45 PDPN population of leukocytes during involution þ þ Quantification of these chimeric structures reveal significantly (Fig. 2C). To further characterize the CD45 PDPN cells, we more of them in mouse mammary tissues during involution, with examined F4/80, a widely used marker of murine macrophages; peak density at involution day 6, when compared with tissues CD64, constitutively found on monocytes and macrophages; from virgin or nulliparous mice (Fig. 1C). We also provide IHC and MerTK, a macrophage apoptotic cell receptor found evidence that single LYVE-1 positive cells are observed early primarily on phagocytic macrophages (Fig. 2D; gating strategy

Figure 1. LYVE-1 expressing macrophages during involution. A, Vessel-like structures that stain positive for LYVE-1 (green) and F4/80 (red) are abundant during involution. B, Merged image and 3D reconstruction of a LYVE-1þF4/80þ vessel at involution day 6. C, Quantiﬁcation of LYVE-1þF4/80þ vessel density in mouse mammary glands from virgin mice and mice at various stages of involution. D, Macrophage (CD68, pink) and lymphatic (PDPN, green) interaction in breast tissue biopsied from a women less than 1 year after pregnancy. Unpaired t test: , P < 0.01; , P < 0.001; , P < 0.0001. Arrows, vessels that were quatntitated. Scale bars, 50 mm.

6476 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

Figure 2. Podoplanin expression on macrophages drives lymphangiogenesis. Flow cytometry analysis of cells isolated from nulliparous or involution day 6 mouse mammary tissues reveals increased lymphatic endothelial cells (A), PDPNþ leukocytes (B), PDPNþ myeloid cells (C), monocytes and macrophages (D–F), as well as PDPN-expressing F4/80þ macrophages (G and H). I and J, F4/80þ macrophages isolated from nulliparous and involution day 6 mouse mammary glands enhance in vitro lymphangiogenesis. Unpaired t test: , P < 0.05; , P < 0.01; , P < 0.001.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6477

Elder et al.

in Supplementary Fig. S2A). In the involution group, we detected robustly than nulliparous-derived macrophages, which do not þ þ þ þ a significant increase in CD45 CD11b F4/80loCD64 cells express PDPN. To test our hypothesis, F4/80 macrophages (Fig. 2E), which we characterize as monocytes. We then identified harvested from involution day 6 mouse mammary glands, or þ macrophages through their coexpression of CD11b F4/ nulliparous controls, were mixed 1:1 with 66cl4 mouse mammary þ 80hiCD64 MerTKhi to show that macrophages also increased in tumor cells, incorporated into a Matrigel plug, and injected number during involution (Fig. 2F). Importantly, the MerTKhi orthotopically into BALB/c mice. Plugs were harvested and dual þ þ cells were also F4/80hi and we show that these cells exhibit IF revealed a significant increase in F4/80 LYVE staining asso- increased expression of PDPN during involution (Fig. 2G and ciated with the Matrigel plug that contained the involution- H). Together, our results suggest that monocytes may be recruited derived macrophages (Fig. 3A). To determine whether SEMA7A into the mammary gland during involution and differentiated contributes to TAM-mediated lymphangiogenesis, we engineered into PDPN-expressing macrophages. Consistent with this hypoth- 66cl4 mouse mammary tumor cells to overexpress SEMA7A esis, analysis of microarray data from whole mouse mammary (Supplementary Fig. S3A) and orthotopically injected them, glands harvested during the involution cycle reveals expression of along with empty vector controls, into nulliparous BALB/c mice. monocyte chemoattractant protein-1 (MCP-1 or CCL2)—which We also injected the empty vector control cell line into mice on is involved in monocyte infiltration into tissues during inflam- involution day 1 to determine whether the involution microen- mation—and its receptor (CCR2) during involution (Supplemen- vironment would be sufficient to support increased tumor- þ þ tary Fig. S2B and S2C; refs. 14, 19). Finally, to determine whether associated F4/80 PDPN cells. Consistent with our previously mammary macrophages are primed for promotion of lymphan- published results, SEMA7A and involution group tumors grew þ giogenesis, we isolated F4/80 mammary macrophages from faster than the controls in nulliparous hosts (Fig. 3B; refs. 8, 10, nulliparous and involution day 6 mouse mammary glands, which 33). Interestingly, vector control cells injected at involution day 1 were then cocultured for 24 hours with human dermal HDLECs in acquired SEMA7A-expression levels similar to those observed in tube formation assays. Increased tubule area was observed when the cells engineered to overexpress SEMA7A (Fig. 3C). Flow the LECs were cultured with involution-derived macrophages in cytometry analysis revealed that SEMA7A-overexpressing tumors þ þ comparison to the LECs cultured with nulliparous-derived macro- and involution group tumors had increased CD45 F4/80 and þ þ þ phages (Fig. 2I and J). CD45 F4/80 PDPN cells (Fig. 3D). We validated our flow þ analysis by showing increased F480 cells in the tumor and þ SEMA7A promotes prolymphatic macrophages in the tumor LYVE-1 vessels in the tumor microenvironment by IHC (Fig. microenvironment 3E and F). In an additional model, E0771 and E0771 SEMA7A We have shown that implantation of mouse mammary tumor cells were injected into C57BL/6 hosts. We utilized the E0771 cells cells during involution results in tumors with increased LVD in the for these additional studies to not only validate our results from tumor microenvironment. Therefore, we hypothesized that invo- the 66cl4 model, but to validate our results in a cell line that does lution-derived macrophages, which express more PDPN, may not express detectable levels of SEMA7A by immunoblot, whereas integrate into or recruit tumor-associated lymphatic vessels more the 66cl4 cells do (Supplementary Fig. S3A). Consistent with this

Figure 3. SEMA7A promotes TAMs and lymphangiogenesis. A, In an in vivo Matrigel plug assay, 66cl4 tumor cells were mixed with macrophages isolated from nulliparous or recently weaned mice (at involution day 6). The resulting plugs were harvested and F4/80/LYVE-1 costaining quantitated. B, Growth of 66cl4 cells engineered to overexpress SEMA7A compared with empty vector controls injected into nulliparous and involution day 1 mouse mammary tissues. C, SEMA7A expression by IHC in tumors from B. D, CD45þCD31-F4/80þPDPNþ cells in tumors from B. E and F, Percentage of F4/80þ cells and number of LYVE-1þ vessel density (F) in tumors from B. G, F4/80þPDPNþ cells in E0771 SEMA7A tumors in nulliparous mice. H, LYVE-1þ density in tumors from G, with the addition of tumors from SEMA7A/ hosts injected with E0771 cells. Unpaired t test: , P < 0.05; , P < 0.01; , P < 0.001.

6478 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

þ þ þ þ þ þ observation, baseline levels of both CD45 F480 PDPN cells CD45 F4/80 PDPN cells and lymphatic vessels in the tumor and LVD (LYVE-1/mm2) are significantly reduced in the control microenvironment. E0771 cells compared with the 66cl4 controls (Supplementary þ Fig. S3B and S3C). In the E0771 cell line, overexpression of SEMA7A and PDPN cell populations during postpartum SEMA7A was sufficient to drive increased tumor-associated mammary involution þ þ þ CD45 F4/80 PDPN cells (Fig. 3G), which also express MerTK To determine whether SEMA7A may play a role in recruiting (Supplementary Fig. S3D). In addition, we observed increased and/or differentiating macrophages during involution, we first LVD with OE of SEMA7A, which was lost in SEMA7A knockout examined whether SEMA7A is expressed in the mouse mammary (SEMA7A / ) mice, suggesting cooperation between host and epithelium. We observed that SEMA7A expressing cells are rare in tumor is necessary for the increased tumor-associated LVD by nulliparous mammary glands, but that nearly 8% to 15% of the þ SEMA7A (Fig. 3H). Together, our results suggest that tumor- epithelial (EpCAM ) cells at involution day 6 exhibit expression derived and host-derived SEMA7A can promote tumor-associated of SEMA7A (Fig. 4A). To determine whether SEMA7A directly

Figure 4. SEMA7A promotes macrophage-mediated lymphatic mimicry. A, SEMA7A is expressed on EpCAMþ cells at involution day 6. In mammary glands from SEMA7A/ mice, we observed decreased LECs (B), PDPNþ immune cells (C), monocytes (D), and monocytes and macrophages expressing PDPN (E and F). G, LYVE-1 (green) and F4/80 (brown) IHC on involution day 6 mouse mammary glands from WT (top) and SEMA7A/ (KO; bottom). a–c and e–g are mammary gland and d and h are lymph node. Unpaired t test: , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001. Scale bars, 50 mm.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6479

Elder et al.

þ supports the presence of PDPN cell populations during mouse frequently found near or associated with the lymphatics (B and C) mammary gland involution, we examined cell populations iso- and a lymphatic/macrophage network was evident in the lymph lated from involution day 6 mouse mammary tissues from node. In contrast, lymphatic vessels were far less abundant in the tm1Alk transgenic mice that lack SEMA7A expression (Sema7a /J; KO and frequently found near the blood vessels, not the mam- ref. 2). We observed a significant reduction in LECs at involution mary epithelium (E). Furthermore, the lymphatic vessels that day 6 in SEMA7A KO mice compared with wild-type C57BL/6 were present displayed altered morphology and there were fewer þ controls (Fig. 4B). We also observed decreased PDPN leuko- macrophages associated with them (F and G). Finally, the lym- cytes, monocytes, and macrophages (Fig. 4C–F). These results phatic/macrophage organization observed in the lymph node suggest a role for SEMA7A in the influx and/or promotion of appears to be altered in the KO (H). Thus, we suggest a mechanism PDPN expression on monocytes and macrophages during by which SEMA7A mediates macrophage-LEC and LEC-LEC con- involution. Interestingly, we also observed the presence of nection to form lymphatic vessels during mammary gland invo- þ þ þ CD45 CD31 PDPN cells at involution day 6, which was not lution and in the TME. influenced by SEMA7A status (Supplementary Fig. S4A). Using IHC to identify lymphatics by LYVE-1 and macrophages by F4/80, SEMA7A promotes macrophage PDPN expression, motility, we examined the relationship between lymphatics and macro- and adhesion to lymphatics phages in WT and KO mice at involution day 6 (Fig. 4G). As To determine whether SEMA7A directly promotes PDPN expected, and consistent with our previous results, the lymphatics expression in macrophages, we treated cultured murine macro- were tightly associated with the mammary epithelium and abun- phages with recombinant SEMA7A and observed increased dant in the WT mice (A). In addition, the macrophages were expression of Pdpn mRNA after 72 hours (Fig. 5A). Conversely,

Figure 5. SEMA7A drives PDPN expression, macrophage motility, and adherence to endothelial cell monolayers. SEMA7A promotes macrophage expression of Pdpn (A), migration (B), and adherence to endothelial cell monolayers in two-dimensional (C), and inclusion in vessel-like structures in three dimensional (D and E). Induction of motility (F) and adherence to LEC monolyaers in two-dimensional (G) is inhibited by blocking b1 integrin (motility and adherence) and PDPN (adherence). H, Macrophage incorporation into three-dimensional LEC vessel-like structures is inhibited by function blocking antibodies against b1 integrin (9EG7) and PDPN (pMAB). Unpaired t test: , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001.

6480 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

we show that SEMA7A is not a driver of CD68 in cultured HDLECs never been pregnant (Fig. 6D). We have also provided histologic (Supplementary Fig. S4B). To determine whether SEMA7A can evidence that PDPN and CD68 co-occur in a patient that was also promote motility of macrophages and incorporation into the diagnosed within 1-year postpartum, who was also positive for lymphatic endothelium, we examined macrophage motility by lymphatic vessel invasion (LVI) and LN involvement. We observe transwell filter assay and adhesion to HDLEC monolayers with clusters of SEMA7Aþ tumor cell emboli within lymphatics in the and without purified SEMA7A. We observed increased motility same patient, providing indirect evidence that macrophages, and adherence in the SEMA7A-treated conditions (Fig. 5B and C). lymphatics, and SEMA7A may contribute to dissemination of Next, we tested whether SEMA7A could promote 3D in vitro tumor cells in postpartum women (Fig. 6E). lymphangiogenesis with 1:1 LEC-macrophage cocultures. Similar We have previously shown that SEMA7A expression is associ- to our results with involution macrophages (Fig. 2I), we observed ated with a significant decrease in OS for patients with breast that the lymphatic structures formed in vitro are larger and that cancer using the METABRIC dataset. A prediction of the results macrophages associate/incorporate with the lymphatic structures presented herein is that SEMA7A drives metastasis via the lym- more frequently with SEMA7A (Fig. 5D and E; Supplementary phatic system and thus, would be associated with decreased OS in Fig. S4C). We did not see the same extensive vessel networks with patients who are lymph node (LN) positive. In confirmation of the same number of LECs þSEMA7A as we did with the coculture, our previous results, analysis of the KmPlot dataset revealed a suggesting that macrophages are doing more than just increasing significant decrease in OS from breast cancer with expression of the cell number in the assay (Supplementary Fig. S4D). Because SEMA7A (n ¼ 664; HR ¼ 1.7; 95% CI, 1.22–2.36; P ¼ 0.0014), SEMA7A promotes motility of other cell types via b1 integrin, and which was greater when the analysis was restricted to LN-positive PDPN can also promote cell motility, we tested whether SEMA7A patients (n ¼ 177; HR ¼ 2.19; 95% CI, 1.28–3.75; P ¼ 0.0032) and promotes macrophage motility in an integrin- and/or PDPN- lost when only LN-negative patients were examined (n ¼ 122; HR dependent manner. We utilized function blocking antibodies ¼ 0.65; 95% CI, 0.26–1.66; P ¼ 0.37; n ¼ 122; Fig. 6 F–H). against b1 integrin and PDPN, 9EG7, or pMAB, respectively, in Interestingly, our data also predicted that SEMA7A expression our transwell migration assay, adhesion assay, and 3D-lymphan- would associate with increased metastasis, yet SEMA7A alone did giogenesis assay. Macrophage motility was most significantly not associate with decreased distant metastasis-free survival blocked by 9EG7, and, only a slight decrease in motility was (DMFS) in the KmPlot analysis (Supplementary Fig. S4). To observed with pMAB (Fig. 5F). In contrast, we also observed the determine whether coexpression of SEMA7A, PDPN, and CD68 most significantly disrupted adhesion with pMAB (Fig. 5G) and increased the probability of distant metastasis we examined overall incorporation into 3D structures was significantly dis- DMFS in KmPlot; we observe that high levels of SEMA7A, PDPN, rupted in the presence of 9EG7 and/or pMAB (Fig. 5H). Together, and CD68 together, but not any one alone (Supplementary our results suggest that SEMA7A may stimulate mammary mac- Fig. S5), result in a significant reduction in 10-year distant rophage expression of PDPN, migration, and adhesion to the metastasis-free survival rates (HR ¼ 1.74; 95% CI, 1.212.49; lymphatic endothelium to aid in neo-lymphatic formation during P ¼ 0.0025; Fig. 6I). The results of our analysis suggest SEMA7A is postpartum mammary gland involution and tumorigenesis. a potent driver of LN-positive disease and that coexpression of SEMA7A, PDPN, and CD68 predicts for metastasis in patients with Involution lymphatics promote breast cancer metastasis breast cancer. In addition, we looked for co-occurrence of We have proposed that lymphatics observed in the mammary SEMA7A, PDPN, and CD68 in The Cancer Genome Atlas tissue during postpartum involution could facilitate early tumor (TCGA; Table 1) and observed significant co-occurrence between cell dissemination in postpartum women. To test whether mouse PDPN, CD68, and SEMA7A in breast cancer (Table 1). Finally, mammary tumor cell lines transit via the lymphatics during active our gene triad appears to be associated with poor prognosis involution, we injected 66cl4 tumor cells into intact mammary in additional epithelial-derived tumor types, where lymphangio- glands at involution day 1, or into nulliparous controls, and genesis at the primary site has been shown to be important for examined both the primary tumor site (the mammary tissue) for distant metastasis, including ovarian, lung, and gastric (Fig. 6J–L; lymphatic vessels, and distant (axillary) lymph nodes at involu- refs. 39–41). Cumulatively, our results suggest that SEMA7A, tion days 4 and 6 for micrometastasis. We first confirmed lymphatics, and macrophages cooperate to promote metastasis increased LYVE-1þ vessels at days 4 and 6 of involution at the in patients with breast cancer as well as in other cancer types where primary site (Fig. 6A and B). Strikingly, the number of micro- lymphangiogenesis occurs. metastases in the axillary lymph nodes, harvested at days 4 and 6 fi of involution, is signi cantly increased at this early time point Discussion (Fig. 6C). Together, these results suggest that involution lymphatics can support tumor cell transit via the lymphatic vasculature to Previously, our lab identified SEMA7A as low/not expressed in result in early dissemination of postpartum tumors. Consistent normal breast tissue via mRNA analysis of human breast tissues in with this result, we have previously published that normal breast publicly available datasets. We also observed that SEMA7A was tissue from women <1 year postpartum exhibits increased LVD increased in human breast tumor samples and associated with and that patients with breast cancer diagnosed <2 years of recent decreased OS in patients with breast cancer. We attributed the role childbirth have increased lymph node involvement (10). We have of SEMA7A in decreasing patient survival to our observed also shown that macrophage infiltration, measured by CD68 decreases in tumor growth, invasion, and lymphangiogenesis that expression, is increased in breast tissue during postpartum invo- were revealed by our preclinical studies where we silenced lution (11). Here we extend our observations to reveal that SEMA7A in human breast tumor xenografts. Here, we identify normal epithelium in breast tissue from women <5 years out a role for SEMA7A in modulation of normal mammary tissue from most recent pregnancy exhibits increased expression, on lymphangiogenesis and the TME, using immunecompetent average, of SEMA7A compared with tissue from women who have murine models whereby SEMA7A expression promotes

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6481

Elder et al.

Figure 6. Lymphatic trafﬁcking of tumor cells is driven by involution and SEMA7A. A, LVD in mouse mammary tissues injected with tumor cells at involution day 1 compared with those injected into nulliparous hosts. B, Representative image of lymphatics at involution day 6. C, Clonogenic assay reveals presence of micrometastasis/axillary lymph node during active involution. D, Quantitative IHC for SEMA7A expression in normal mammary tissue from never-been-pregnant women or women within 5 years of most recent childbirth. E, CD68, PDPN, and SEMA7A in a patient diagnosed within 1-year postpartum, who was also posititve for LVI and LN involvement. OS analysis using KmPlot for all (n ¼ 626; F), LN negative (n ¼ 122; G), and LN positive (n ¼ 177; H) patients. I, Distant-metastatsis-free survival for patients coexpressing SEMA7A, PDPN, and CD68. J, Progression-free survival in ovarian cancer. K and L, Overall survival in lung and gastric cancer. Unpaired t test: , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001. Scale bars, 50 mm.

macrophages to a lymphatic promoting phenotype. In addition, shown to mobilize what have been called "monocyte-derived we observe that macrophages, SEMA7A, and lymphatic markers lymphatic endothelial cell progenitors (MLECP)," which are co-occur in patient samples where coexpression also signiﬁcantly LYVE-1 positive and integrate into lymphatic vessels prior to predicts for poor prognosis, suggesting these mechanisms are at sprouting (29). Although lymphatic marker expression on TAMS play in patients with breast cancer. and activated macrophages has been reported (25–30, 35, 42), it TAM are known to promote tumor lymphangiogenesis through has not been investigated in the normal adult mammary gland. both paracrine and cell autonomous modes (25–30). Speciﬁcally, Because alternatively-activated macrophages make up the major- TAMs can express prolymphangiogenic factors that can signal to ity of the macrophages in the adult mammary gland during the preexisting lymphatic vessels and/or tumors and have been postpartum involution (11, 17), we suspected that these

6482 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

Table 1. Analysis of TCGA-RNA-seq data for co-occurrence of PDPN, CD68, and regression, suggest that this mechanism may only prove true SEMA7A during transient lymphatic remodeling, which favors lymphatic TCGA-RNA-seq co-occurence mimicry rather than transdifferentiation. Finally, our studies PDPN CD68 SEMA7A ¼ P < ¼ P ¼ provide the framework for a role for SEMA7A in monocyte PDPN OR 2.253; 0.001 OR 1.191; 0.005 fl CD68 OR ¼ 1.172; P ¼ 0.006 in ux into the mammary gland, which is suggestive of non- tissue-resident macrophages. Future studies will determine whether involution macrophages are derived from resident-tissue "involution macrophages" may be like the MLECPs observed in macrophages or from mobilization and differentiation of tumors and/or capable of "lymphatic mimicry." Our results peripheral monocytes. indicate that involution macrophages are driven, in part, by Our studies also identify, for the first time, that SEMA7A is mammary epithelial-cell expression of SEMA7A, that involution expressed on the normal mammary epithelium in women within macrophages express lymphatic marker PDPN in a SEMA7A- 5 years of recent pregnancy and during the tissue remodeling dependent manner, and that involution macrophages associate phase of postpartum mammary involution in rodents. These with lymphatic vessels during involution. Whether these cells are observations are consistent with our results showing that lym- actually endothelial cell progenitors is not addressed by our phatic vessel density increases during normal involution in current studies and will be addressed by future studies that involve rodents and in normal breast tissue in women after lactation and lineage tracing. remains elevated for up to 6 years later (10). Other studies have We do reveal that the PDPN-expressing macrophages likely similarly suggested that inflammation and angiogenesis associ- participate in lymphangiogenesis by moving toward lymphatic ated gene expression changes are observed in tissue from parous structures and by adhering to lymphatic endothelial cells. We also women, compared with nulliparous, and some of these changes show that in vitro lymphatic network formation is dependent on persist in the breast tissue up to 10 years after pregnancy (45). both PDPN–CLEC-2 interaction and a known receptor for Furthermore, our own studies indicate that tumor cell expression SEMA7A, namely b1 integrin. Inhibition of b1 integrin, using an of SEMA7A can be driven by exposure to the involution micro- antibody that inhibits ligand binding, primarily blocked transwell environment (Fig. 4A). These results suggest that tumors in migration of macrophages in response to SEMA7A, suggesting women whose breast tissue has recently undergone involution that the SEMA7A–b1 integrin interaction functions to stimulate may be "imprinted" upon by mechanisms that are activated macrophage migration during involution. Conversely, the dom- during involution. Consistent with this hypothesis we have inant effect of our PDPN inhibition strategy, which utilized an shown that human and murine tumor cells implanted during antibody that blocks the interaction of PDPN with its receptor postpartum involution in rodents obtain (and retain, after invo- CLEC-2, was on adhesion of SEMA7A-treated macrophages to lution is complete) multiple protumorigenic attributes such as endothelial cell monolayers, suggesting that this ligand receptor increased growth, invasion, expression of proinflammatory mole- interaction is required for macrophage adherence to lymphatics. cules, induction of lymphangiogenesis, and ability to seed met- Cumulatively, our results suggest a mechanism by which SEMA7A astatic sites (8, 10, 33). These models of postpartum breast cancer promotes mammary macrophage migration through activation of have provided us with mechanistic insight into why postpartum b1 integrin and adhesion through promoting the association breast cancers, diagnosed within 5 years of recent pregnancy, are between PDPN and CLEC-2. This mechanism is consistent with nearly three times more likely to develop metastasis when com- results from the literature, where b1 integrin is known to be pared to never been pregnant patients with similar clinical char- expressed on monocyte-derived cells and to promote monocyte acteristics, such as age and stage (46). Based on our data presented recruitment and activation. We predict that once the monocytes herein showing that SEMA7A is increased during involution, we have been recruited from the bone marrow to the mammary gland are investigating whether SEMA7A is a primary driver of postpar- they become differentiated into activated macrophages. Then, tum breast cancer aggressiveness in our University of Colorado under continued exposure to the SEMA7A produced by the Young Women's cohort (46). Interestingly, a recent in silico mammary epthelium, they express PDPN and adhere to existing analysis revealed two involution gene signatures that were signif- lymphatic vessels to result in lymphatic expansion, which is icantly upregulated in inflammatory breast cancer (IBC; ref. 47). facilitated through PDPN–CLEC-2 interaction. This observation IBC is a highly aggressive type of breast cancer that is characterized is also supported by the literature whereby PDPN–CLEC-2 inter- by tumor emboli in dermal lymphatics, which is reminiscent of action has been shown to be important for platelet adhesion to what we have observed in mammary lymphatics of our postpar- lymphatic endothelium (43). Additionally, both PDPN–CLEC-2 tum patients (Fig. 6E; ref. 10). Importantly, when we mined the and SEMA7A–b1 integrin downstream signaling pathways lead to publicly available data, we identified that expression of SEMA7A activation of pro-proliferative and prosurvival pathways, includ- was observed in one of the two signatures that are associated with ing Src and PI3K, thus, these interactions could result in prolif- IBC. Given the multiple similarities between IBC and involu- eration and survival of the mammary macrophages and/or the tion—including macrophages, lymphatics, and the proinflamma- lymphatic endothelial cells. Published data also suggest that tory enzyme COX-2 (8, 10, 11, 33, 48)—it is possible that PDPN mediates differentiation of keratinocytes by downregula- SEMA7A plays a significant role in the unique biology of both tion of b1 integrin (44), suggesting there might be cross-talk inflammatory and postpartum breast cancers. between the two pathways. Additional studies are necessary Our analysis of additional cancer types also suggests that to determine the stability of the interaction between PDPN- SEMA7A may play a general role in tumor metastasis and should expressing macrophages and LECs, and the downstream pathways therefore be investigated as a novel target for cancer therapeutics. that are activated and/or deactivated, in order to conclude wheth- Furthermore, our data showing that SEMA7A is a significant driver er the effects of the interaction are transient or long lasting. Our of decreased OS for LNþ patients suggest that SEMA7A primarily results showing peak LVD during mid-involution, followed by drives distant metastasis via the lymphatic vasculature, which is

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6483

Elder et al.

supported by our preclinical data that SEMA7A drives lymphan- Writing, review, and/or revision of the manuscript: A.M. Elder, B.A.J. giogenesis. Two recent papers in Science identify a blood- Tamburini, L.S. Crump, V.M. Wessells, P.J. Schedin, V.F. Borges, T.R. Lyons lymphatic hybrid route for metastatic spread in preclinical Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A.M. Elder, V.F. Borges, T.R. Lyons models of melanoma and breast cancer (49, 50). Because Study supervision: T.R. Lyons SEMA7A has been shown to affect both angio- and lymphangiogenesis, it is possible that SEMA7A facilitates this hybrid route of Acknowledgments metastasis and should be targeted to prevent metastasis. Given Cancer League of Colorado AWD#173586-TL (T.R. Lyons, B.A.J. Tam- that SEMA7A expression is low in most adult tissues, targeting burini, A.M. Elder), Colorado Clinical Translational Sciences Institute SEMA7A could have minimal toxicities. Furthermore, it is possi- KL2TR001080 (T.R. Lyons, S.A. Black, A.M. Elder), American Cancer Society ble that SEMA7A could be a biomarker for patients who are at high RSG-16-171-010CSM (T.R. Lyons, A.M. Elder, L.S. Crump, V.M. Wessells), National Cancer Institute R01 CA211696-01A1 (T.R. Lyons, A.M. Elder, L.S. risk for metastasis, and therefore should receive more aggressive Crump, V.M. Wessells), R21CA185226-01 (T.R. Lyons, S.A. Black, V.M. therapy; because SEMA7A can be shed into the tumor microen- Wessells), AACR-BCRF 09-06-26 (V.F. Borges, P.J. Schedin, V.M. Wessells), vironment, it is also possible that SEMA7A could be detected in NIH1R01CA169175-01 (V.F. Borges, P.J. Schedin, T.R. Lyons, A.M. Elder, the blood. In summary, SEMA7A may be a promising target and and V.M. Wessells), and Grohne Family Foundation Grant (V.F. Borges, P.J. biomarker for metastatic cancer. Schedin, T.R. Lyons). Imaging experiments were performed in the Univer- sity of Colorado Anschutz Medical Campus Advance Light Microscopy Disclosure of Potential Conflicts of Interest Core. L.S. Crump and T.R. Lyons were supported in part by NIH/NCATS Colorado CTSI Grant TR001082 and RR025780. Tissue Bio banking fl No potential con icts of interest were disclosed. occurred in the Shared Resource of Colorado's NIH/NCI Cancer Center Support Grant P30CA046934. The authors would also like to thank the Disclaimer following persons for all of their help in the production of this manuscript: Contents are the authors' sole responsibility and do not necessarily represent Holly Martinson, Justin Pruitt, Charlene Tilton, Cera Neito, Kim Jordan, official NIH views. Michelle Borakove, Jeffrey Finlon, Gavin Ryan, Peter Hensen, and the UC Cohort Patients. Authors' Contributions Conception and design: A.M. Elder, B.A.J. Tamburini, V.F. Borges, T.R. Lyons Development of methodology: A.M. Elder, B.A.J. Tamburini, V.M. Wessells, The costs of publication of this article were defrayed in part by the P.J. Schedin, V.F. Borges, T.R. Lyons payment of page charges. This article must therefore be hereby marked advertisement Acquisition of data (provided animals, acquired and managed patients, in accordance with 18 U.S.C. Section 1734 solely to indicate provided facilities, etc.): A.M. Elder, L.S. Crump, S.A. Black, P.J. Schedin, this fact. V.F. Borges, T.R. Lyons Analysis and interpretation of data (e.g., statistical analysis, biostatistics, Received June 3, 2018; revised August 15, 2018; accepted September 20, 2018; computational analysis): A.M. Elder, B.A.J. Tamburini, L.S. Crump, S.A. Black, published first September 25, 2018. V.F. Borges, T.R. Lyons

References 1. Yamada A, Kubo K, Takeshita T, Harashima N, Kawano K, Mine T, et al. dependent mouse mammary epithelial cells. Mol Biol Cell 2012; Molecular cloning of a glycosylphosphatidylinositol-anchored molecule 23:3873–81. CDw108. J Immunol 1999;162:4094–100. 10. Lyons TR, Borges VF, Betts CB, Guo Q, Kapoor P, Martinson HA, et al. 2. Pasterkamp RJ, Peschon JJ, Spriggs MK, Kolodkin AL. Semaphorin 7A Cyclooxygenase-2-dependent lymphangiogenesis promotes nodal metas- promotes axon outgrowth through integrins and MAPKs. Nature tasis of postpartum breast cancer. J Clin Invest 2014;124:3901–12. 2003;424:398–405. 11. O'Brien J, Lyons T, Monks J, Lucia MS, Wilson RS, Hines L, et al. Alterna- 3. Suzuki K, Okuno T, Yamamoto M, Pasterkamp RJ, Takegahara N, tively activated macrophages and collagen remodeling characterize the Takamatsu H, et al. Semaphorin 7A initiates T-cell-mediated inflam- postpartum involuting mammary gland across species. Am J Pathol matory responses through alpha1beta1 integrin. Nature 2007;446: 2010;176:1241–55. 680–4. 12. Jindal S, Gao D, Bell P, Albrektsen G, Edgerton SM, Ambrosone CB, et al. 4. Reilkoff RA, Peng H, Murray LA, Peng X, Russell T, Montgomery R, et al. Postpartum breast involution reveals regression of secretory lobules medi- Semaphorin 7aþ regulatory T cells are associated with progressive idio- ated by tissue-remodeling. Breast Cancer Res 2014;16:R31. pathic pulmonary fibrosis and are implicated in transforming growth 13. Atabai K, Sheppard D, Werb Z. Roles of the innate immune system in factor-beta1-induced pulmonary fibrosis. Am J Respir Crit Care Med mammary gland remodeling during involution. J Mammary Gland Biol 2013;187:180–8. Neoplasia 2007;12:37–45. 5. Scott GA, McClelland LA, Fricke AF. Semaphorin 7a promotes spreading 14. Clarkson RW, Wayland MT, Lee J, Freeman T, Watson CJ. Gene expression and dendricity in human melanocytes through beta1-integrins. J Invest profiling of mammary gland development reveals putative roles for death Dermatol 2008;128:151–61. receptors and immune mediators in post-lactational regression. Breast 6. Ma B, Herzog EL, Lee CG, Peng X, Lee CM, Chen X, et al. Role of chitinase 3- Cancer Res 2004;6:R92–109. like-1 and semaphorin 7a in pulmonary melanoma metastasis. Cancer Res 15. Li M, Liu X, Robinson G, Bar-Peled U, Wagner KU, Young WS, et al. 2015;75:487–96. Mammary-derived signals activate programmed cell death during the first 7. Garcia-Areas R, Libreros S, Amat S, Keating P, Carrio R, Robinson P, et al. stage of mammary gland involution. Proc Nat Acad Sci USA 1997;94: Semaphorin7A promotes tumor growth and exerts a pro-angiogenic effect 3425–30. in macrophages of mammary tumor-bearing mice. Front Physiol 16. Lund LR, Romer J, Thomasset N, Solberg H, Pyke C, Bissell MJ, et al. Two 2014;5:17. doi: 10.3389/fphys.2014.00017. distinct phases of apoptosis in mammary gland involution: proteinase- 8. Black SA, Nelson AC, Gurule NJ, Futscher BW, Lyons TR. Semaphorin 7a independent and -dependent pathways. Development 1996;122: exerts pleiotropic effects to promote breast tumor progression. Oncogene 181–93. 2016;35:5170–8. 17. Martinson HA, Jindal S, Durand-Rougely C, Borges VF, Schedin P. Wound 9. Allegra M, Zaragkoulias A, Vorgia E, Ioannou M, Litos G, Beug H, et al. healing-like immune program facilitates postpartum mammary gland Semaphorin-7a reverses the ERF-induced inhibition of EMT in Ras- involution and tumor progression. Int J Cancer 2015;136:1803–13.

6484 Cancer Res; 78(22) November 15, 2018 Cancer Research

SEMA7A Promotes Mammary Tumor-Associated Lymphangiogenesis

18. Ramirez RA, Lee A, Schedin P, Russell JS, Masso-Welch PA. Alterations in 34. Podgrabinska S, Braun P, Velasco P, Kloos B, Pepper MS, Skobe M. mast cell frequency and relationship to angiogenesis in the rat mammary Molecular characterization of lymphatic endothelial cells. PNAS 2002; gland during windows of physiologic tissue remodeling. Dev Dyn 99:16069–74. 2012;241:890–900. 35. Xu H, Chen M, Reid DM, Forrester JV. LYVE-1-positive macrophages are 19. Stein T, Morris JS, Davies CR, Weber-Hall SJ, Duffy MA, Heath VJ, et al. present in normal murine eyes. Invest Ophthalmol Visual Sci 2007; Involution of the mouse mammary gland is associated with an immune 48:2162–71. cascade and an acute-phase response, involving LBP, CD14 and STAT3. 36. Jiang S, Bailey AS, Goldman DC, Swain JR, Wong MH, Streeter PR, et al. Breast Cancer Res 2004;6:R75–91. Hematopoietic stem cells contribute to lymphatic endothelium. PLoS One 20. Strange R, Li F, Saurer S, Burkhardt A, Friis RR. Apoptotic cell death and 2008;3:e3812. tissue remodelling during mouse mammary gland involution. Develop- 37. Mantovani A, Locati M. Tumor-associated macrophages as a paradigm of ment 1992;115:49–58. macrophage plasticity, diversity, and polarization: lessons and open ques- 21. Borges VF, Elder AM, Lyons TR. Deciphering Pro-Lymphangiogenic pro- tions. Arterioscler Thromb Vasc Biol 2013;33:1478–83. grams during mammary involution and postpartum breast cancer. Front 38. Stout RD, Watkins SK, Suttles J. Functional plasticity of macrophages: in Oncol 2016;6:227. doi: 10.3389/fonc.2016.00227. situ reprogramming of tumor-associated macrophages. J Leukoc Biol 22. Becker F, Kurmaeva E, Gavins FNE, Stevenson EV, Navratil AR, Jin L, et al. A 2009;86:1105–9. Critical Role for Monocytes/Macrophages During Intestinal Inflammation- 39. Nakamura Y, Yasuoka H, Tsujimoto M, Kurozumi K, Nakahara M, Nakao K, associated Lymphangiogenesis. Inflamm Bowel Dis 2016;22:1326–45. et al. Importance of lymph vessels in gastric cancer: a prognostic indicator 23. Cursiefen C, Chen L, Borges LP, Jackson D, Cao J, Radziejewski C, et al. in general and a predictor for lymph node metastasis in early stage cancer. VEGF-A stimulates lymphangiogenesis and hemangiogenesis in inflam- J Clin Pathol 2006;59:77–82. matory neovascularization via macrophage recruitment. J Clin Invest 40. Sundar SS, Zhang H, Brown P, Manek S, Han C, Kaur K, et al. Role of 2004;113:1040–50. lymphangiogenesis in epithelial ovarian cancer. Br J Cancer 2006;94: 24. Harvey NL, Gordon EJ. Deciphering the roles of macrophages in devel- 1650–7. opmental and inflammation stimulated lymphangiogenesis. Vascular Cell 41. Rouzaut A, Irigoyen M, Montuenga LM. Lymphangiogenesis and lung 2012;4:15. doi: 10.1186/2045-824X-4-15. cancer. J Thorac Oncol 2007;2:384–6. 25. Jeon BH, Jang C, Han J, Kataru RP, Piao L, Jung K, et al. Profound but 42. Dollt C, Becker K, Michel J, Melchers S, Weis CA, Schledzewski K, et al. The dysfunctional lymphangiogenesis via vascular endothelial growth factor shedded ectodomain of Lyve-1 expressed on M2-like tumor-associated ligands from CD11bþ macrophages in advanced ovarian cancer. macrophages inhibits melanoma cell proliferation. Oncotarget 2017; Cancer Res 2008;68:1100–9. 8:103682–92. 26. Kerjaschki D. The crucial role of macrophages in lymphangiogenesis. 43. Bertozzi CC, Schmaier AA, Mericko P, Hess PR, Zou Z, Chen M, et al. J Clin Invest 2005;115:2316–9. Platelets regulate lymphatic vascular development through CLEC-2-SLP- 27. Kim KE, Koh YJ, Jeon BH, Jang C, Han J, Kataru RP, et al. Role of CD11bþ 76 signaling. Blood 2010;116:661–70. macrophages in intraperitoneal lipopolysaccharide-induced aberrant lym- 44. Masaru Honma MF, Takahashi H, Iizuka H. Podoplanin alters J Dermatol phangiogenesis and lymphatic function in the diaphragm. Am J Pathol Sci 2013;69:e1–e46. 2009;175:1733–45. 45. Asztalos S, Gann PH, Hayes MK, Nonn L, Beam CA, Dai Y, et al. Gene 28. Maruyama K, Ii M, Cursiefen C, Jackson DG, Keino H, Tomita M, et al. expression patterns in the human breast after pregnancy. Cancer Prev Res Inflammation-induced lymphangiogenesis in the cornea arises from (Phila) 2010;3:301–11. CD11b-positive macrophages. J Clin Invest 2005;115:2363–72. 46. Callihan EB, Gao D, Jindal S, Lyons TR, Manthey E, Edgerton S, et al. 29. Ran S, Montgomery KE. Macrophage-mediated lymphangiogenesis: the Postpartum diagnosis demonstrates a high risk for metastasis and merits an emerging role of macrophages as lymphatic endothelial progenitors. expanded definition of pregnancy-associated breast cancer. Breast Cancer Cancers (Basel) 2012;4:618–57. Res Treat 2013;138:549–59. 30. Zumsteg A, Baeriswyl V, Imaizumi N, Schwendener R, Ruegg C, Christofori 47. Bambhroliya A, Van Wyhe RD, Kumar S, Debeb BG, Reddy JP, Van Laere S, G. Myeloid cells contribute to tumor lymphangiogenesis. PLoS One et al. Gene set analysis of post-lactational mammary gland involution gene 2009;4:e7067. signatures in inflammatory and triple-negative breast cancer. PLoS One 31. Morote-Garcia JC, Napiwotzky D, Kohler D, Rosenberger P. Endothelial 2018;13:e0192689. Semaphorin 7A promotes neutrophil migration during hypoxia. PNAS 48. Lim B, Woodward WA, Wang X, Reuben JM, Ueno NT. Inflammatory 2012;109:14146–51. breast cancer biology: the tumour microenvironment is key. Nat Rev 32. Holmes S, Downs AM, Fosberry A, Hayes PD, Michalovich D, Murdoch P, Cancer 2018. et al. Sema7A is a potent monocyte stimulator. Scand J Immunol 49. Pereira ER, Kedrin D, Seano G, Gautier O, Meijer EFJ, Jones D, et al. Lymph 2002;56:270–75. node metastases can invade local blood vessels, exit the node, and colonize 33. Lyons TR, O'Brien J, Borges VF, Conklin MW, Keely PJ, Eliceiri KW, et al. distant organs in mice. Science 2018;359:1403–7. Postpartum mammary gland involution drives progression of ductal 50. Brown M, Assen FP, Leithner A, Abe J, Schachner H, Asfour G, et al. Lymph carcinoma in situ through collagen and COX-2. Nat Med 2011;17: node blood vessels provide exit routes for metastatic tumor cell dissem- 1109–15. ination in mice. Science 2018;359:1408–11.

www.aacrjournals.org Cancer Res; 78(22) November 15, 2018 6485

Semaphorin 7A Promotes Macrophage-Mediated Lymphatic Remodeling during Postpartum Mammary Gland Involution and in Breast Cancer

Alan M. Elder, Beth A.J. Tamburini, Lyndsey S. Crump, et al.

Cancer Res 2018;78:6473-6485. Published OnlineFirst September 25, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-18-1642

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2018/09/25/0008-5472.CAN-18-1642.DC1

Cited articles This article cites 49 articles, 16 of which you can access for free at: http://cancerres.aacrjournals.org/content/78/22/6473.full#ref-list-1

Citing articles This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/78/22/6473.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/78/22/6473. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.