Tamoxifen Response at Single-Cell Resolution in Estrogen Receptor-Positive Primary Human Breast Tumors

PURPOSE

In estrogen receptor-positive (ER+)/HER2- breast cancer, multiple measures of intratumor heterogeneity are associated with a worse response to endocrine therapy. We sought to develop a novel experimental model to measure heterogeneity in response to tamoxifen treatment in primary breast tumors.

EXPERIMENTAL DESIGN

To investigate heterogeneity in response to treatment, we developed an operating room-to-laboratory pipeline for the collection of live normal breast specimens and human tumors immediately after surgical resection for processing into single-cell workflows for experimentation and genomic analyses. Live primary cell suspensions were treated ex vivo with tamoxifen (10 μmol/L) or control media for 12 hours, and single-cell RNA libraries were generated using the 10X Genomics droplet-based kit.

RESULTS

In total, we obtained and processed normal breast tissue from two women undergoing reduction mammoplasty and tumor tissue from 10 women with ER+/HER2- invasive breast carcinoma. We demonstrate differences in tamoxifen response by cell type and identify distinctly responsive and resistant subpopulations within the malignant cell compartment of human tumors. Tamoxifen resistance signatures from resistant subpopulations predict poor outcomes in two large cohorts of ER+ breast cancer patients and are enriched in endocrine therapy-resistant tumors.

CONCLUSIONS

This novel ex vivo model system now provides the foundation to define responsive and resistant subpopulations within heterogeneous human tumors, which can be used to develop precise single cell-based predictors of response to therapy and to identify genes and pathways driving therapeutic resistance.

Single-Cell Transcriptional and Epigenetic Profiles of Male Breast Cancer Nominate Salient Cancer-Specific Enhancers

Male breast cancer represents about 1% of all breast cancer diagnoses and, although there are some similarities between male and female breast cancer, the paucity of data available on male breast cancer makes it difficult to establish targeted therapies. To date, most male breast cancers (MBCs) are treated according to protocols established for female breast cancer (FBC). Thus, defining the transcriptional and epigenetic landscape of MBC with improved resolution is critical for developing better avenues for therapeutic intervention. In this study, we present matched transcriptional (scRNA-seq) and epigenetic (scATAC-seq) profiles at single-cell resolution of two treatment naïve MBC tumors processed immediately after surgical resection. These data enable the detection of differentially expressed genes between male and female breast tumors across immune, stromal, and malignant cell types, to highlight several genes that may have therapeutic implications. Notably, MYC target genes and mTORC1 signaling genes were significantly upregulated in the malignant cells of MBC compared to the female counterparts. To understand how the regulatory landscape of MBC gives rise to these male-specific gene expression patterns, we leveraged the scATAC-seq data to systematically link changes in chromatin accessibility to changes in gene expression within each cell type. We observed cancer-specific rewiring of several salient enhancers and posit that these enhancers have a higher regulatory load than lineage-specific enhancers. We highlight two examples of previously unannotated cancer-cell-specific enhancers of ANXA2 and PRDX4 gene expression and show evidence for super-enhancer regulation of LAMB3 and CD47 in male breast cancer cells. Overall, this dataset annotates clinically relevant regulatory networks in male breast tumors, providing a useful resource that expands our current understanding of the gene expression programs that underlie the biology of MBC.

Tamoxifen Response at Single Cell Resolution in Estrogen Receptor-Positive Primary Human Breast Tumors

In ER+/HER2- breast cancer, multiple measures of intra-tumor heterogeneity are associated with worse response to endocrine therapy. To investigate heterogeneity in response to treatment, we developed an operating room-to-laboratory pipeline for the collection of live human tumors and normal breast specimens immediately after surgical resection for processing into single-cell workflows for experimentation and genomic analyses. We demonstrate differences in tamoxifen response by cell type and identify distinctly responsive and resistant subpopulations within the malignant cell compartment of human tumors. Tamoxifen resistance signatures from 3 distinct resistant subpopulations are prognostic in large cohorts of ER+ breast cancer patients and enriched in endocrine therapy resistant tumors. This novel ex vivo model system now provides a foundation to define responsive and resistant sub-populations within heterogeneous tumors, to develop precise single cell-based predictors of response to therapy, and to identify genes and pathways driving resistance to therapy.

Abstract 5754: Determining the regulatory logic of breast cancer cells using single-cell multi-omics

Cancer cells rewire regulatory elements scattered throughout the genome (such as enhancers) to drive aberrant gene expression. Thus, deconvoluting the regulatory mechanisms that contribute to oncogenic gene expression in cancer cells is key to understanding tumor biology. To this end, we have charted the transcriptional and epigenetic landscape of breast cancer at single-cell resolution to quantitatively link variation in chromatin accessibility to gene expression across malignant and non-malignant cell types. Our comprehensive dataset profiles the chromatin landscape (scATAC-seq) in concert with the transcriptional profiles (scRNA-seq) of 4 breast cancer cell lines, 12 primary breast tumors, and 4 normal mammary reduction tissue specimens collected and processed immediately after surgical resection. This dataset, encompassing over 250,000 individual cells, allowed us to define the regulatory logic of cancer cells by 1) revealing how the epigenome underlies cellular heterogeneity of these tumors in comparison to normal mammary tissue, 2) defining how malignant cells hijack enhancer elements to drive key transcriptional programs in a subtype-specific manner, and 3) annotating which cancer-specific enhancer-to-gene connections portend a worse outcome in patients. Notably, we discovered that cancer cells acquire de novo non-coding enhancer elements to modulate hallmark cancer pathways that were previously hidden using bulk genomics approaches. This highlights the potential for cancer-specific enhancers to serve as markers with diagnostic and prognostic potential, or even serve as tractable targets for therapeutic intervention. Together these data enable the annotation of the cellular composition, transcriptional, and epigenetic landscape of breast tumors to help pinpoint clinically relevant mechanisms of tumorigenesis.

Citation Format: Matthew J. Regner, Aatish Thennavan, Susana Garcia-Recio, Kamila Wisniewska, Philip M. Spanheimer, Joel S. Parker, Charles M. Perou, Hector L. Franco. Determining the regulatory logic of breast cancer cells using single-cell multi-omics. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5754.

Abstract PD4-08: PD4-08 A Novel Single Cell Model of Tamoxifen Response in Primary Human Breast Tumors

A Novel Single Cell Model of Tamoxifen Response in Primary Human Breast Tumors Austin Whitman, Hyunsoo Kim, Kamila Wisniewska, Rasha Kakati, Susana Garcia Recio, Hector Franco, Charles Perou, Philip Spanheimer Background: Resistance to endocrine therapy is a primary cause of treatment failure and death in patients with estrogen receptor (ER)-positive breast cancer. Intratumor heterogeneity is associated with resistance to therapy across tumors, and specifically in ER+/HER2- breast cancer, heterogeneity in ER and PR expression is associated with a worse response to endocrine therapy. We hypothesize that subpopulations within and across ER+/HER2- human breast tumors have distinct responses to tamoxifen and that discerning heterogeneity in response will improve understanding of inherent and emerging resistance to endocrine therapy. Methods: We developed an operating room-to-laboratory pipeline immediately after surgical resection for studies using alive tissue. Tissue samples were obtained and single cell suspensions created using physical and enzymatic dissociation. Cells were treated with tamoxifen (10 M) or control media for 12 hours in suspension and single cell RNA libraries generated using the 10X Genomics droplet-based kit and sequenced using the Illumina NextSeq2000. Results: We obtained normal breast tissue from 2 women undergoing reduction mammoplasty and tumor tissue from 10 women with ER+/HER2- invasive breast carcinoma. In tamoxifen treated and control matched pairs, a total of 22,195 cells from normal breast and 94,558 cells from tumor samples were sequenced. Computational analysis using consensus clustering was performed and cell types assigned using canonical correlation. Both tumor and normal samples identified clustering by cell type and not by patient revealing significant variability in cell type abundance between samples. In the normal breast samples, we performed differentially expressed genes (DEG) analysis comparing tamoxifen treatment to control for each cell type (Immune cells, fibroblasts, basal epithelial cells, luminal progenitor cells, and mature luminal cells) and enrichment analysis of up- and down-regulated genes performed. Strong depletion of estrogen induced genes was observed in tamoxifen-treated normal luminal progenitor and mature luminal cells, but not in basal epithelial cells or fibroblasts, demonstrating distinct, subpopulation-specific response to tamoxifen. In the 10 tumor matched pairs, 4 had a high epithelial proportion and tumor cells identified using inferred copy number variation. Tumor cells in 3 of these 4 samples showed significant down regulation of estrogen response genes with tamoxifen treatment. Using scBCSubtype to assign PAM50 subtype to individual tumor cells, the 3 responsive tumors were comprised primarily of LumA cells while the unresponsive tumor was predominantly LumB. Finally, we developed a novel score to quantify responsiveness at the single cell level based on downregulation of estrogen response genes with tamoxifen treatment relative to matched cluster-specific untreated expression. This analysis demonstrated heterogeneity in response to tamoxifen in tumor cells and identified distinct subpopulations of responsive and unresponsive tumor cells to tamoxifen treatment. Conclusion: We developed a novel ex vivo model to determine heterogeneity in therapeutic response to tamoxifen in normal human breast tissue and primary human breast tumors. We demonstrate differences in tamoxifen response by cell type and identify distinctly responsive and resistant subpopulations within human tumors. This provides a foundation to define features of responsive and resistant populations on the individual cell and specimen basis, and should allow us to develop precise, single cell-based predictors of response to endocrine therapy, and to identify genes and pathways driving resistance to therapy.

Citation Format: Hyunsoo Kim, Austin Whitman, Kamila Wisniewska, Susana Garcia-Recio, Rasha Kakati, Hector L. Franco, Charles M. Perou, Philip Spanheimer. PD4-08 A Novel Single Cell Model of Tamoxifen Response in Primary Human Breast Tumors [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD4-08.

Abstract P5-14-02: Identifying oncogenic enhancer elements in TNBC of the Basal-like subtype using single-cell ATAC-seq and RNA-seq

Identification of the cis-regulatory elements controlling oncogenic transcriptional programs is critical to understanding tumor biology. To find cis-regulatory elements (i.e. gene enhancers) of oncogenic dependencies in Triple-Negative Breast Cancers (TNBC) of the Basal-like gene expression subtype, we generated matched single-cell transcriptome (scRNA-seq) and chromatin accessibility (scATAC-seq) profiles for two human Basal-like tumors and four normal mammary reduction samples. This unique dataset enabled us to correlate variations in chromatin structure with variations in gene expression revealing putative enhancers that are specifically active within cancer cells, but not within normal mammary ductal epithelial cells. We then leveraged the Cancer Dependency Map (DepMap) portal at the BROAD Institute to infer gene expression dependencies in breast cancer cell lines of the Basal-like molecular subtype. Putative cancer-specific enhancers were prioritized based on the transcriptional dependency of their target gene(s) in Basal-like cell lines as reported by the DepMap portal. Based on our preliminary analyses, we report several cancer-specific enhancers that drive the expression of important transcription factors such as EN1 and SOX4. These transcription factors are known to have profound effects on tumor biology, especially considering that high expression of EN1 is associated with brain metastasis and SOX4 is known to regulate immune evasion and PI3K/Akt signaling. Moreover, both of these transcription factors portend a worse outcome in TNBC patients. Thus, our analysis suggests that high levels of expression of these transcription factors is sustained specifically within the malignant cell types of these tumors, by the activity of these cancer-specific enhancers that are not typically active in normal epithelial cells. We are now performing CRISPR dCas9-KRAB experiments to epigenetically silence these cancer-specific enhancers and measure the consequences on expression of their predicted target genes. Additionally, we are investigating the trans-acting transcription factors that may physically bind to these enhancers to further regulate oncogenic transcription. By defining the regulatory logic of cancer cells at single-cell resolution, our work highlights the importance of cancer-specific and clinically relevant oncogenic regulatory elements in TNBC of the Basal-like subtype.

Citation Format: Matthew J. Regner, Aatish Thennavan, Kamila Wisniewska, Susana Garcia-Recio, Raul Mendez-Giraldez, Philip Spanheimer, Charles M. Perou, Hector L. Franco. Identifying oncogenic enhancer elements in TNBC of the Basal-like subtype using single-cell ATAC-seq and RNA-seq [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-14-02.

A multi-omic dissection of super-enhancer driven oncogenic gene expression programs in ovarian cancer

The human genome contains regulatory elements, such as enhancers, that are often rewired by cancer cells for the activation of genes that promote tumorigenesis and resistance to therapy. This is especially true for cancers that have little or no known driver mutations within protein coding genes, such as ovarian cancer. Herein, we utilize an integrated set of genomic and epigenomic datasets to identify clinically relevant super-enhancers that are preferentially amplified in ovarian cancer patients. We systematically probe the top 86 super-enhancers, using CRISPR-interference and CRISPR-deletion assays coupled to RNA-sequencing, to nominate two salient super-enhancers that drive proliferation and migration of cancer cells. Utilizing Hi-C, we construct chromatin interaction maps that enable the annotation of direct target genes for these super-enhancers and confirm their activity specifically within the cancer cell compartment of human tumors using single-cell genomics data. Together, our multi-omic approach examines a number of fundamental questions about how regulatory information encoded into super-enhancers drives gene expression networks that underlie the biology of ovarian cancer.

Enhancer RNA Transcription Is Essential for a Novel CSF1 Enhancer in Triple-Negative Breast Cancer

Enhancers are critical regulatory elements in the genome that help orchestrate spatiotemporal patterns of gene expression during development and normal physiology. In cancer, enhancers are often rewired by various genetic and epigenetic mechanisms for the activation of oncogenes that lead to initiation and progression. A key feature of active enhancers is the production of non-coding RNA molecules called enhancer RNAs, whose functions remain unknown but can be used to specify active enhancers de novo. Using a combination of eRNA transcription and chromatin modifications, we have identified a novel enhancer located 30 kb upstream of Colony Stimulating Factor 1 (CSF1). Notably, CSF1 is implicated in the progression of breast cancer, is overexpressed in triple-negative breast cancer (TNBC) cell lines, and its enhancer is primarily active in TNBC patient tumors. Genomic deletion of the enhancer (via CRISPR/Cas9) enabled us to validate this regulatory element as a bona fide enhancer of CSF1 and subsequent cell-based assays revealed profound effects on cancer cell proliferation, colony formation, and migration. Epigenetic silencing of the enhancer via CRISPR-interference assays (dCas9-KRAB) coupled to RNA-sequencing, enabled unbiased identification of additional target genes, such as RSAD2, that are predictive of clinical outcome. Additionally, we repurposed the RNA-guided RNA-targeting CRISPR-Cas13 machinery to specifically degrade the eRNAs transcripts produced at this enhancer to determine the consequences on CSF1 mRNA expression, suggesting a post-transcriptional role for these non-coding transcripts. Finally, we test our eRNA-dependent model of CSF1 enhancer function and demonstrate that our results are extensible to other forms of cancer. Collectively, this work describes a novel enhancer that is active in the TNBC subtype, which is associated with cellular growth, and requires eRNA transcripts for proper enhancer function. These results demonstrate the significant impact of enhancers in cancer biology and highlight their potential as tractable targets for therapeutic intervention.

A multi-omic single-cell landscape of human gynecologic malignancies

Deconvolution of regulatory mechanisms that drive transcriptional programs in cancer cells is key to understanding tumor biology. Herein, we present matched transcriptome (scRNA-seq) and chromatin accessibility (scATAC-seq) profiles at single-cell resolution from human ovarian and endometrial tumors processed immediately following surgical resection. This dataset reveals the complex cellular heterogeneity of these tumors and enabled us to quantitatively link variation in chromatin accessibility to gene expression. We show that malignant cells acquire previously unannotated regulatory elements to drive hallmark cancer pathways. Moreover, malignant cells from within the same patients show substantial variation in chromatin accessibility linked to transcriptional output, highlighting the importance of intratumoral heterogeneity. Finally, we infer the malignant cell type-specific activity of transcription factors. By defining the regulatory logic of cancer cells, this work reveals an important reliance on oncogenic regulatory elements and highlights the ability of matched scRNA-seq/scATAC-seq to uncover clinically relevant mechanisms of tumorigenesis in gynecologic cancers.

Nitrogen removal and greenhouse gas fluxes from integrated buffer zones treating agricultural drainage water

Integrated buffer zones (IBZ) are novel mitigation measures designed to decrease the loading of nitrogen (N) transported by subsurface drainage systems from agricultural fields to streams. In IBZ, drainage water flows into a pond with free water surface followed by an inundated, vegetated filterbed. This design provides an environment favorable for denitrification and thus a decrease in nitrate concentration is expected as water flow through the IBZ. However, due to the establishment of anaerobic conditions, there is a risk for increasing emissions of the greenhouse gases nitrous oxide (N₂O) and methane (CH₄). In this year-long study, we evaluated the N removal efficiency along with the risk of N₂O and CH₄ emissions from two pilot-scale IBZs (IBZ1 and 2). The two IBZs had very different yearly removal efficiencies, amounting to 29% and 71% of the total N load at IBZ1 and 2, respectively. This was probably due to differences in infiltration rates to the filterbed, which was 22% and 81% of the incoming water at IBZ1 and 2, respectively. The site (IBZ2) with the highest removal efficiency was a net N₂O sink, while 0.9% of the removed nitrate was emitted as N₂O at IBZ1. Both IBZs were net sources of CH₄ but with different pathways of emission. In IBZ1 CH₄ was mainly lost directly to the atmosphere, while waterborne losses dominated in IBZ2. In conclusion, the IBZs were effective in removing N three years after establishment, and although the IBZs acted as greenhouse gas sources, especially due to CH₄, the emissions were comparable to those of natural wetlands and other drainage transport mitigation measures.

Abstract 4496: Cytokine-induced post-translational modifications of FOXA1 affect enhancer selection and estrogen signaling in breast cancer cells

The pioneer transcription factor FOXA1 is a critical determinant for estrogen receptor (ER) function in hormone-dependent breast cancers. Upon estrogen stimulation, liganded ER binds to poised enhancer regions across the genome that are demarcated by FOXA1 and histone modifications such as H3K4me1 and H3K27ac. In a recent publication, we show that proinflammatory signaling, caused by the cytokine TNFa, drives FOXA1 to latent enhancer binding sites to promote chromatin accessibility for subsequent ER binding upon estrogen ligation. These latent enhancers, activated by the combined treatment of estrogen and TNFa, induced the expression of a unique transcriptome with clinical significance. The effects of TNFa treatment on FOXA1 chromatin redistribution and subsequent gene expression occur within 40 minutes, which points to a rapid signaling cascade that culminates in either changes in FOXA1's posttranslational modifications (PTMs) or its binding partners. To understand how proinflammatory TNFa signaling can redirect FOXA1 to new sites across the genome, we started by characterizing the posttranslational modifications (PTMs) of FOXA1. We immunoprecipitated FOXA1 from MCF-7 breast cancer cells that were treated by E2, TNFa or E2+ TNFa, and then examined their posttranslational status using semi-quantitative and quantitative mass spectrometry approaches. Several phosphorylation sites and acetylation sites have been identified near the DNA binding domain of FOXA1, and acetylation of lysine 295 (K295) was found specifically enriched in TNFa treatment. To test if acetylation of FOXA1 at K295 changes its binding preference and genomic distribution, we used the programmable properties of CRISPR/Cas9 to create specific knockin mutations to mimic or prevent acetylation of K295 in MCF-7 cells. More specifically, we mutated K295 to glutamine (K295Q) to mimic acetylation and essentially “lock” FOXA1 into a permanently acetylated state and, for comparison, we created another cell line where K295 was mutated to arginine (K295R) to prevent acetylation of FOXA1. Our preliminary data shows changes in the genomic redistribution of FOXA1 in the knock-in cell lines resulting in altered gene expression programs. These data suggest that inflammation-based acetylation of FOXA1 can affect estrogen signaling pathways in breast cancer cells by altering the enhancer landscape of FOXA1 and consequently the estrogen receptor.

Supported by a grant from the NIH/NCI (R00 CA204628) to H.L.F

Citation Format: Shen Li, Raul Mendez-Giraldez, Joseph P. Garay, Kamila Wisniewska, Colby A. Tubbs, Charles M. Perou, Hector L. Franco. Cytokine-induced post-translational modifications of FOXA1 affect enhancer selection and estrogen signaling in breast cancer cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4496.

Dose-dependent effects of glucocorticoids on pulmonary vascular development in a murine model of hyperoxic lung injury

BACKGROUND

Exposure of neonatal mice to hyperoxia results in pulmonary vascular remodeling and aberrant phosphodiesterase type 5 (PDE5) signaling. Although glucocorticoids are frequently utilized in the NICU, little is known about their effects on the developing pulmonary vasculature and on PDE5. We sought to determine the effects of hydrocortisone (HC) on pulmonary vascular development and on PDE5 in a neonatal mouse model of hyperoxic lung injury.

METHODS

C57BL/6 mice were placed in 21% O2 or 75% O2 within 24 h of birth and received HC (1, 5, or 10 mg/kg subcutaneously every other day) or vehicle. At 14 d, right ventricular hypertrophy (RVH), medial wall thickness (MWT), lung morphometry, and pulmonary artery (PA) PDE5 activity were assessed. PDE5 activity was measured in isolated pulmonary artery smooth muscle cells exposed to 21 or 95% O2 ± 100 nmol/l HC for 24 h.

RESULTS

Hyperoxia resulted in alveolar simplification, RVH, increased MWT, and increased PA PDE5 activity. HC decreased hyperoxia-induced RVH and attenuated MWT. HC had dose-dependent effects on alveolar simplification. HC decreased hyperoxia-induced PDE5 activity both in vivo and in vitro.

CONCLUSIONS

HC decreases hyperoxia-induced pulmonary vascular remodeling and attenuates PDE5 activity. These findings suggest that HC may protect against hyperoxic injury in the developing pulmonary vasculature.

Association study of the 2-bp deletion polymorphism in exon 6 of the CHRFAM7A gene with idiopathic generalized epilepsy

There is evidence of linkage between the 15q13-q14 locus, containing the gene encoding the α7 subunit (CHRNA7) of the neuronal nicotinic acetylcholine receptor (nAChR) and its partially duplicated isoform (CHRFAM7A), and epilepsy. Additionally, a 2-bp deletion polymorphism (c.497-498delTG; rs67158670) in CHRFAM7A, resulting in a frame shift and truncation of the protein product, is associated with some neurological diseases. This study was designed to explore the possibility of an association of the c.497-498delTG polymorphism of CHRFAM7A with idiopathic generalized epilepsies (IGEs) in Polish children and young patients. The study included 197 IGE patients and 258 unrelated healthy individuals. The frequency of the CHRFAM7A c.497-498delTG polymorphism was determined in each group using heteroduplex analysis. An association between the c.497-498delTG polymorphism of CHRFAM7A and IGE was evidenced. It was demonstrated that the frequency of the CHRFAM7A 2-bp deletion carriers was significantly lower in the IGE patients than in the control group. The observed frequency of 2-bp deletion carriers was high in IGE subjects (64%), but significantly higher in control subjects (76%). Carriers of at least one copy of the -2 bp allele had halved their risk of IGE susceptibility (delTG/delTG and delTG/wild-type versus wild-type/wild-type: odds ratio=0.55; 95% confidence intervals=0.365-0.827; p=0.004). Moreover, it has been demonstrated that this polymorphic variant is associated with the c.524-12_524-11insGTT variation (rs10649395) in intron 7 of CHRFAM7A. Our study substantiates the involvement of the α7 subunit of nAChR in the pathophysiology of IGEs and indicates that the CHRFAM7A c.497-498TG deletion or a nearby polymorphism may play a role in the pathogenesis of IGE. Further work should concentrate on ascertaining the exact mechanism of this polymorphism's effect and its relationship with IGE.

Effects of freezing on major arteries

Thirty experiments were performed in dogs to assess the effects of freezing on carotid and femoral arteries. These vessels withstood freezing and thawing without rupture. Blood within the vessels did not clot and was lysed during thawing.

Histological changes of vessel walls consisted of early degeneration followed by a reparative process. Occasional secondary thrombosis occurred close to bifurcation of ligated branches and was related to extent and frequency of freezing and thawing.