LOX, but not LOXL2, promotes bone metastasis formation and bone destruction in triple-negative breast cancer

The primary function of the lysyl oxidase (LOX) family, including LOX and its paralogue LOX-like (LOXL)-2, is to catalyze the covalent crosslinking of collagen and elastin in the extracellular matrix. LOX and LOXL2 are also facilitating breast cancer invasion and metastatic spread to visceral organs (lungs, liver) in vivo. Conversely, the contribution of LOX and LOXL2 to breast cancer bone metastasis remains scant. Here, using gene overexpression or silencing strategies, we investigated the role of LOX and LOXL2 on the formation of metastatic osteolytic lesions in animal models of triple negative breast cancer. In vivo, the extent of radiographic metastatic osteolytic lesions in animals injected with LOX-overexpressing [LOX(+)] tumor cells was 3-fold higher than that observed in animals bearing tumors silenced for LOX [LOX(-)]. By contrast, the extent of osteolytic lesions between LOXL2(+) and LOXL2(-) tumor-bearing animals did not differ, and was comparable to that observed with LOX(-) tumor-bearing animals. In situ, TRAP staining of bone tissue sections from the hind limbs of LOX(+) tumor-bearing animals was substantially increased compared to LOX(-), LOXL2(+) and LOXL2(-)-tumor-bearing animals, which was indicative of enhanced active-osteoclast resorption. In vitro, tumor-secreted LOX increased osteoclast differentiation induced by RANKL, whereas LOXL2 seemed to counteract LOX's pro-osteoclastic activity. Furthermore, LOX (but not LOXL2) overexpression in tumor cells induced a robust production of IL-6, the latter being a pro-osteoclastic cytokine. Based on these findings, we propose a model in which LOX and IL-6 secreted from tumor cells act in concert to enhance osteoclast-mediated bone resorption that, in turn, promotes metastatic bone destruction in vivo.

Liposome-encapsulated zoledronate increases inflammatory macrophage population in TNBC tumours

BACKGROUND

Tumour associated macrophages (TAMs) are important players in breast tumour progression and metastasis. Clinical and preclinical evidence suggests a role for zoledronate (ZOL) in breast cancer metastasis prevention. Further, zoledronate is able to induce inflammatory activation of monocytes and macrophages, which can be favourable in cancer treatments. The inherent bone tropism of zoledronate limits its availability in soft tissues and tumours. In this study we utilised an orthotopic murine breast cancer model to evaluate the possibility to use liposomes (EMP-LIP) to target zoledronate to tumours to modify TAM activation.

METHODS

Triple-negative breast cancer 4T1 cells were inoculated in the 4th mammary fat pad of female Balb/c mice. Animals were divided according to the treatment: vehicle, ZOL, EMP-LIP and liposome encapsulated zoledronate (ZOL-LIP). Treatment was done intravenously (with tumour resection) and intraperitoneally (without tumour resection). Tumour growth was followed by bioluminescence in vivo imaging (IVIS) and calliper measurements. Tumour-infiltrating macrophages were assessed by immunohistochemical and immunofluorescence staining. Protein and RNA expression levels of inflammatory transcription factors and cytokines were measured by Western Blotting and Taqman RT-qPCR.

RESULTS

Liposome encapsulated zoledronate (ZOL-LIP) treatment suppressed migration of 4T1 cell in vitro. Tumour growth and expression of the angiogenic marker CD34 were reduced upon both ZOL and ZOL-LIP treatment in vivo. Long-term ZOL-LIP treatment resulted in shift towards M1-type macrophage polarization, increased CD4 T cell infiltration and activation of NF-κB indicating changes in intratumoural inflammation, whereas ZOL treatment showed similar but non-significant trends. Moreover, ZOL-LIP had a lower bisphosphonate accumulation in bone compared to free ZOL.

CONCLUSION

Results show that the decreased bisphosphonate accumulation in bone promotes the systemic anti-tumour effect of ZOL-LIP by increasing inflammatory response in TNBC tumours via M1-type macrophage activation.

Dysregulation of MicroRNAs in Adult Osteogenesis Imperfecta: The miROI Study

As epigenetic regulators of gene expression, circulating micro-RiboNucleic Acids (miRNAs) have been described in several bone diseases as potential prognostic markers. The aim of our study was to identify circulating miRNAs potentially associated with the severity of osteogenesis imperfecta (OI) in three steps. We have screened by RNA sequencing for the miRNAs that were differentially expressed in sera of a small group of OI patients versus controls and then conducted a validation phase by RT-qPCR analysis of sera of a larger patient population. In the first phase of miROI, we found 79 miRNAs that were significantly differentially expressed. We therefore selected 19 of them as the most relevant. In the second phase, we were able to validate the significant overexpression of 8 miRNAs in the larger OI group. Finally, we looked for a relationship between the level of variation of the validated miRNAs and the clinical characteristics of OI. We found a significant difference in the expression of two microRNAs in those patients with dentinogenesis imperfecta. After reviewing the literature, we found 6 of the 8 miRNAs already known to have a direct action on bone homeostasis. Furthermore, the use of a miRNA-gene interaction prediction model revealed a 100% probability of interaction between 2 of the 8 confirmed miRNAs and COL1A1 and/or COL1A2. This is the first study to establish the miRNA signature in OI, showing a significant modification of miRNA expression potentially involved in the regulation of genes involved in the physiopathology of OI. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

MiR-662 is associated with metastatic relapse in early-stage breast cancer and promotes metastasis by stimulating cancer cell stemness

BACKGROUND

Breast cancer (BC) metastasis, which often occurs in bone, contributes substantially to mortality. MicroRNAs play a fundamental role in BC metastasis, although microRNA-regulated mechanisms driving metastasis progression remain poorly understood.

METHODS

MiRome analysis in serum from BC patients was performed by TaqMan™ low-density array. MiR-662 was overexpressed following MIMIC-transfection or lentivirus transduction. Animal models were used to investigate the role of miR-662 in BC (bone) metastasis. The effect of miR-662-overexpressing BC cell conditioned medium on osteoclastogenesis was investigated. ALDEFLUOR assays were performed to study BC stemness. RNA-sequencing transcriptomic analysis of miR-662-overexpressing BC cells was performed to evaluate gene expression changes.

RESULTS

High levels of hsa-miR-662 (miR-662) in serum from BC patients, at baseline (time of surgery), were associated with future recurrence in bone. At an early-stage of the metastatic disease, miR-662 could mask the presence of BC metastases in bone by inhibiting the differentiation of bone-resorbing osteoclasts. Nonetheless, metastatic miR-662-overexpressing BC cells then progressed as overt osteolytic metastases thanks to increased stem cell-like traits.

CONCLUSIONS

MiR-662 is involved in BC metastasis progression, suggesting it may be used as a prognostic marker to identify BC patients at high risk of metastasis.

Integrin alpha5 in human breast cancer is a mediator of bone metastasis and a therapeutic target for the treatment of osteolytic lesions

Bone metastasis remains a major cause of mortality and morbidity in breast cancer. Therefore, there is an urgent need to better select high-risk patients in order to adapt patient's treatment and prevent bone recurrence. Here, we found that integrin alpha5 (ITGA5) was highly expressed in bone metastases, compared to lung, liver, or brain metastases. High ITGA5 expression in primary tumors correlated with the presence of disseminated tumor cells in bone marrow aspirates from early stage breast cancer patients (n = 268; p = 0.039). ITGA5 was also predictive of poor bone metastasis-free survival in two separate clinical data sets (n = 855, HR = 1.36, p = 0.018 and n = 427, HR = 1.62, p = 0.024). This prognostic value remained significant in multivariate analysis (p = 0.028). Experimentally, ITGA5 silencing impaired tumor cell adhesion to fibronectin, migration, and survival. ITGA5 silencing also reduced tumor cell colonization of the bone marrow and formation of osteolytic lesions in vivo. Conversely, ITGA5 overexpression promoted bone metastasis. Pharmacological inhibition of ITGA5 with humanized monoclonal antibody M200 (volociximab) recapitulated inhibitory effects of ITGA5 silencing on tumor cell functions in vitro and tumor cell colonization of the bone marrow in vivo. M200 also markedly reduced tumor outgrowth in experimental models of bone metastasis or tumorigenesis, and blunted cancer-associated bone destruction. ITGA5 was not only expressed by tumor cells but also osteoclasts. In this respect, M200 decreased human osteoclast-mediated bone resorption in vitro. Overall, this study identifies ITGA5 as a mediator of breast-to-bone metastasis and raises the possibility that volociximab/M200 could be repurposed for the treatment of ITGA5-positive breast cancer patients with bone metastases.

A Signature of Circulating miRNAs Associated With Fibrous Dysplasia of Bone: the mirDys Study

Fibrous dysplasia (FD) is a rare bone disease caused by activating mutations of GNAS encoding the Gsα protein, enhancing cyclic adenosine monophosphate (cAMP) production by overstimulation of adenylyl cyclase and impairing osteoblastic differentiation. The clinical presentation ranges from asymptomatic to polyostotic forms with severe disability, explained by the mosaic distribution of the GNAS mutation. Physicians have to deal with the gap of knowledge in FD pathogenesis, the absence of prognostic markers and the lack of specific treatment. The identification of specific biomarkers for FD is an important step to improve the clinical and therapeutic approaches. An epigenetic regulation driven by microRNAs (miRNAs), known as promising biomarkers in bone disease, could be involved in FD. We have sought circulating miRNAs that are differentially expressed in FD patients compared to controls and would reflect dysregulations of osteogenesis-related genes and bone disorder. The global miRNA profiling was performed using Next Generation Sequencing in patient serum collected from a discovery cohort of 20 patients (10 polyostotic and 10 monostotic) and 10 controls. From these, we selected 19 miRNAs for a miRNA validation phase from serum of 82 patients and 82 controls, using real-time qPCR. Discovery screening identified 111 miRNAs differentially expressed in patient serum, after adjusting for the false discovery rate (FDR). Among the 82 patients, 55% were polyostotic, and 73% were women with a mean age of 42 years. Six miRNAs (miR-25-3p, miR-93-5p, miR-182-5p, miR-324-5p, miR-363-3p, and miR-451a) were significantly overexpressed in serum, with FDR <0.05. The expression level of these six miRNAs was not associated with the FD severity. In conclusion, we identified a signature of circulating miRNAs associated with FD. These miRNAs are potential negative regulators of gene expression in bone cell progenitors, suggesting their activity in FD by interfering with osteoblastic and osteoclastic differentiation to impair bone mineralization and remodeling processes. © 2020 American Society for Bone and Mineral Research.

Association of circulating microRNAs with prevalent and incident knee osteoarthritis in women: the OFELY study

Objectives

In the context of the scarcity of biomarkers for knee osteoarthritis (OA), we examined the associations of prevalent and incident OA with the expression levels of serum miRNAs in subjects with and without OA.

Methods

With a next-generation sequencing approach, we compared the miRome expression of 10 women with knee OA and 10 age-matched healthy subjects. By real-time qPCR, we analyzed the expression levels of 19 miRNAs at baseline selecting 43 women with prevalent knee OA (Kellgren Lawrence score of 2/3), 23 women with incident knee OA over a 4-year follow-up and 67 healthy subjects without prevalent or incident OA matched for age and body mass index.

Results

Serum miR-146a-5p was significantly increased in the group of prevalent knee OA compared with controls (relative quantification (RQ); median [Interquartile range] 1.12 [0.73; 1.46] vs 0.85 [0.62; 1.03], p = 0.015). The likelihood of prevalent knee OA was significantly increased (odds ratio [95% confidence interval (CI)] 1.83 [1.21–2.77], p = 0.004) for each quartile increase in serum miR-146a-5p. The women with miR-146a-5p levels above the median (0.851) had a higher risk of prevalent knee OA compared to those below the median [95% CI] 4.62 [1.85–11.5], p = 0.001. Moreover, we found a significant association between the baseline level of serum miR-186-5p and the risk of incident knee OA (Q4 vs Q1–3; odds ratio [95% CI] 6.13 [1.14–32.9], p = 0.034).

Conclusion

We showed for the first time that miR-146a-5p and miR-186-5p are significantly associated with prevalent and incident knee OA, respectively.

Lack of Association Between Select Circulating miRNAs and Bone Mass, Turnover, and Fractures: Data From the OFELY Cohort

Postmenopausal osteoporosis is characterized by the occurrence of fragility fracture with an increase in morbidity and mortality. Recently, microRNAs (miRNAs) have raised interest as regulators of translational repression, mediating a number of key processes, including bone tissue in both physiological and diseased states. The aim of this study was to examine the serum levels of 32 preselected miRNAs with reported function in bone and their association with osteoporotic fracture. We performed cross-sectional and longitudinal analyses from the OFELY Cohort. Serum levels of the miRNAs were quantified by qRT-PCR in 682 women: 99 premenopausal and 583 postmenopausal women, with 1 and 122 women with prevalent fragility fractures in each group, respectively. We have collected clinical variables (such as age, prevalent, and incident fractures), bone turnover markers (BTMs), BMD by dual X-ray absorptiometry, and bone microarchitecture with HRpQCT. We observed a number of miRNAs to be associated with fragility fractures (prevalent or incident), BTMs, BMD, and microarchitecture. This effect, however, was negated after age adjustment. This may be because age was also strongly associated with the serum levels of the 32 miRNAs (correlation coefficient up to 0.49), confirming previous findings. In conclusion, in a well-characterized prospective cohort with a sizeable sample size, we found no evidence that these 32 preselected miRNAs were not associated with BTMs, BMD, microarchitecture, and or fragility fractures. © 2019 American Society for Bone and Mineral Research.

ERRα promotes breast cancer cell dissemination to bone by increasing RANK expression in primary breast tumors

Bone is the most common metastatic site for breast cancer. Estrogen-related-receptor alpha (ERRα) has been implicated in cancer cell invasiveness. Here, we established that ERRα promotes spontaneous metastatic dissemination of breast cancer cells from primary mammary tumors to the skeleton. We carried out cohort studies, pharmacological inhibition, gain-of-function analyses in vivo and cellular and molecular studies in vitro to identify new biomarkers in breast cancer metastases. Meta-analysis of human primary breast tumors revealed that high ERRα expression levels were associated with bone but not lung metastases. ERRα expression was also detected in circulating tumor cells from metastatic breast cancer patients. ERRα overexpression in murine 4T1 breast cancer cells promoted spontaneous bone micro-metastases formation when tumor cells were inoculated orthotopically, whereas lung metastases occurred irrespective of ERRα expression level. In vivo, Rank was identified as a target for ERRα. That was confirmed in vitro in Rankl stimulated tumor cell invasion, in mTOR/pS6K phosphorylation, by transactivation assay, ChIP and bioinformatics analyses. Moreover, pharmacological inhibition of ERRα reduced primary tumor growth, bone micro-metastases formation and Rank expression in vitro and in vivo. Transcriptomic studies and meta-analysis confirmed a positive association between metastases and ERRα/RANK in breast cancer patients and also revealed a positive correlation between ERRα and BRCA1mut carriers. Taken together, our results reveal a novel ERRα/RANK axis by which ERRα in primary breast cancer promotes early dissemination of cancer cells to bone. These findings suggest that ERRα may be a useful therapeutic target to prevent bone metastases.

The C-Terminal Intact Forms of Periostin (iPTN) Are Surrogate Markers for Osteolytic Lesions in Experimental Breast Cancer Bone Metastasis

Periostin is an extracellular matrix protein that actively contributes to tumor progression and metastasis. Here, we hypothesized that it could be a marker of bone metastasis formation. To address this question, we used two polyclonal antibodies directed against the whole molecule or its C-terminal domain to explore the expression of intact and truncated forms of periostin in the serum and tissues (lung, heart, bone) of wild-type and periostin-deficient mice. In normal bones, periostin was expressed in the periosteum and specific periostin proteolytic fragments were found in bones, but not in soft tissues. In animals bearing osteolytic lesions caused by 4T1 cells, C-terminal intact periostin (iPTN) expression disappeared at the invasive front of skeletal tumors where bone-resorbing osteoclasts were present. In vitro, we found that periostin was a substrate for osteoclast-derived cathepsin K, generating proteolytic fragments that were not recognized by anti-periostin antibodies directed against iPTN. In vivo, using an in-house sandwich immunoassay aimed at detecting iPTN only, we observed a noticeable reduction of serum periostin levels (- 26%; P < 0.002) in animals bearing osteolytic lesions caused by 4T1 cells. On the contrary, this decrease was not observed in women with breast cancer and bone metastases when periostin was measured with a human assay detecting total periostin. Collectively, these data showed that mouse periostin was degraded at the bone metastatic sites, potentially by cathepsin K, and that the specific measurement of iPTN in serum should assist in detecting bone metastasis formation in breast cancer.

miRNA-30 Family Members Inhibit Breast Cancer Invasion, Osteomimicry, and Bone Destruction by Directly Targeting Multiple Bone Metastasis-Associated Genes

miRNAs are master regulators of gene expression that play key roles in cancer metastasis. During bone metastasis, metastatic tumor cells must rewire their biology and express genes that are normally expressed by bone cells (a process called osteomimicry), which endow tumor cells with full competence for outgrowth in the bone marrow. Here, we establish miR-30 family members miR-30a, miR-30b, miR-30c, miR-30d, and miR-30e as suppressors of breast cancer bone metastasis that regulate multiple pathways, including osteomimicry. Low expression of miR-30 in primary tumors from patients with breast cancer were associated with poor relapse-free survival. In addition, estrogen receptor (ER)-negative/progesterone receptor (PR)-negative breast cancer cells expressed lower miR-30 levels than their ER/PR-positive counterparts. Overexpression of miR-30 in ER/PR-negative breast cancer cells resulted in the reduction of bone metastasis burden in vivoIn vitro, miR-30 did not affect tumor cell proliferation, but did inhibit tumor cell invasion. Furthermore, overexpression of miR-30 restored bone homeostasis by reversing the effects of tumor cell-conditioned medium on osteoclastogenesis and osteoblastogenesis. A number of genes associated with osteoclastogenesis stimulation (IL8, IL11), osteoblastogenesis inhibition (DKK-1), tumor cell osteomimicry (RUNX2, CDH11), and invasiveness (CTGF, ITGA5, ITGB3) were identified as targets for repression by miR-30. Among these genes, silencing CDH11 or ITGA5 in ER-/PR-negative breast cancer cells recapitulated inhibitory effects of miR-30 on skeletal tumor burden in vivo Overall, our findings provide evidence that miR-30 family members employ multiple mechanisms to impede breast cancer bone metastasis and may represent attractive targets for therapeutic intervention.Significance: These findings suggest miR-30 family members may serve as an effective means to therapeutically attenuate metastasis in triple-negative breast cancer. Cancer Res; 78(18); 5259-73. ©2018 AACR.

In Vitro Incorporation and Metabolism of Icosapentaenoic and Docosahexaenoic Acids in Human Platelets - Effect on Aggregation

Washed human platelets were pre-loaded with icosapentaenoic acid (EPA), docosahexaenoic acid (DHA) or EPA + DHA and tested for their aggregation response in comparison with control platelets. In fatty acid-rich platelets, an inhibition of the aggregation could be observed when induced by thrombin, collagen or U-46619. The strongest inhibition was observed with DHA-rich platelets and it was reduced when DHA was incorporated in the presence of EPA.

Study of fatty acid distribution in cell lipids after loading showed that around 90% of EPA or DHA taken up was acylated into phospholipids and a very small amount (less than 2%) remained in their free and hydroxylated forms. DHA was more efficiently acylated into phosphatidylethanolamine (PE) than into phosphatidylinositol (PI) in contrast to what observed with EPA, and both acids were preferentially incorporated into phosphatidylcholine (PC). EPA inhibited total incorporation of DHA and increased its relative acylation into PE at the expense of PC. In contrast, DHA did not affect the acylation of EPA. Upon stimulation with, thrombin, EPA was liberated from phospholipids and oxygenated (as judged by the formation of its monohydroxy derivative) whereas DHA was much less metabolized, although consistently transferred into PE.

It is concluded that EPA and DHA might affect platelet aggregation via different mechanisms when pre-loaded in phospholipids. Whereas EPA is known to alter thromboxane A2 metabolism from endogenous arachidonic acid, by competing with it, DHA might act directly at the membrane level for inhibiting aggregation.

Abstract 486: MicroRNA-30 family members inhibit breast cancer invasion, osteomimicry, and bone destruction by directly targeting multiple bone metastasis-associated genes

MicroRNAs (miRNAs), which are master regulators of gene expression, have emerged as key players in cancer metastasis. Here, we established miR-30 family members (miR-30a, miR-30b, miR-30c, miR-30d and miR-30e) as breast cancer bone metastasis suppressor genes. Low expression levels of miR-30s in primary tumors of breast cancer patients were associated with poor relapse-free survival. Additionally, miR-30s levels were significantly lower in estrogen receptor (ER)-negative/progesterone receptor (PR)-negative tumors than in their hormone-responsive counterparts. Overexpression of miR-30s in ER/PR-negative breast cancer cells resulted in the reduction of bone metastasis burden in vivo. MiR-30s overexpression also reduced tumor outgrowth, when tumor cells were implanted subcutaneously in animals. In vitro, miR-30s did not affect tumor cell proliferation, but inhibited tumor cell invasion. Furthermore, miR-30s restored bone homeostasis by reversing the in vitro effects of the tumor cell conditioned medium on stimulation of osteoclastogenesis and inhibition of osteoblastogenesis. We identified a number of genes associated with osteoclastogenesis stimulation (IL-8, IL-11), osteoblastogenesis inhibition (DKK-1), tumor cell osteomimicry (RUNX2, CDH11) and invasiveness (CTGF, ITGA5, ITGB3) that were direct and/or indirect targets for repression by miR-30s. Among these genes, integrin ITGA5 was a previously unknown miR-30 target. By silencing ITGA5 in ER-/PR-negative breast cancer cells, colonization of the bone marrow by tumor cells was drastically reduced in vivo. Overall, our data indicate that miR-30s employ multiple mechanisms to impede breast cancer bone metastasis. These findings may pave the way to a new field of therapeutic interventions in breast cancer patients with bone metastasis.

Citation Format: Martine Croset, Francesco Pantano, Casina Kan, Edith Bonnelye, Francoise Descotes, Catherine Alix-Panabières, Charles Lecellier, Richard Bachelier, Nathalie Allioli, Saw See Hong, Kai Bartkowiak, Klaus Pantel, Philippe Clézardin. MicroRNA-30 family members inhibit breast cancer invasion, osteomimicry, and bone destruction by directly targeting multiple bone metastasis-associated genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 486.

Abstract 29: RANK-RANKL signaling inhibition delays early breast cancer bone metastasis formation

Denosumab is a fully human monoclonal antibody to human RANKL, currently approved in the management of established bone metastases, mainly due to its anti-resorptive activity which blocks tumor-induced osteoclastogenesis and osteolysis.

Our aims are to determine the role of RANK expression in breast cancer (BC) bone metastasis and the efficacy of RANKL blocking by AMG161 (IgG1 equivalent to Denosumab). AMG161 does not recognize murine RANKL, so we use a BalB/c nude knock-in (KI) mouse model, which expresses a chimeric murine/human RANKL. Moreover, non-invasive in vivo imaging techniques such as bioluminescence and µCT were used to monitor metastatic tumor burden and bone degradation, respectively. To evaluate bone marrow early colonization by BC cells, DiD labelling and ex vivo laser scanning multiphoton confocal microscopy were used.

Long-term studies showed that intra-arterial (i.a.) injection of MDA-MB231 RANK versus MDA-MB231 leads to a poorer overall survival (OS), with the most striking differences seen in the earlier time-points. In this respect, the extent of radiographic osteolytic lesions was higher in the RANK overexpressing group than in the control group. Similar findings were observed following intra-tibial inoculation of RANK-expressing MCF-7 cells, when compared to that observed with parental MCF-7 cells. However, at later time points, bone and lung metastatic burden was similar between the two groups, suggesting that the poor OS of the RANK overexpressing group was due to a faster tumor progression owing to enhanced bone homing and colonization. Indeed, using short-term studies (5 and 14 days), i.a. injection of MDA-MB231 RANK led to higher incidence and extension of bone marrow micro-metastases than MDA-MB231, whereas there was no difference for lung micro-metastases. Daily subcutaneous injections of 1,5mg/kg AMG161 antibody to MDA-MB231RANK tumor-bearing animals, decreased bone micro-metastases and early bone marrow colonization without affecting lung micro-metastasis.

We therefore hypothesized that RANK overexpressing BC cells interact with RANKL produced by osteocytes, osteoblasts and other cells in the osteogenic niche, favoring survival and colonization of BC cells in bone. To test this hypothesis, we made use of heterotypic mammosphere assays in which BC cells were cultured with or without cells of the osteogenic niche (e.g. pre-osteoblasts, mesenchymal stem cells), recombinant RANKL and anti-RANK antibody or AMG161. Co-culture with recombinant RANKL or a cellular source of RANKL increased BC cell mammosphere formation, and blocking RANK/RANKL signaling impaired it.

In conclusion, RANK/RANKL signaling promotes the early engraftment of BC cells in the bone marrow. Blocking this signaling with AMG161 decreases bone marrow micro-metastasis formation in vivo, suggesting that adjuvant treatment of early-stage breast cancer patients with denosumab may prevent bone relapse.

Citation Format: Sofia Sousa, Evelyne Gineyts, Sandra Geraci, Martine Croset, Philippe Clézardin. RANK-RANKL signaling inhibition delays early breast cancer bone metastasis formation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 29.

Estrogen related receptor alpha in castration-resistant prostate cancer cells promotes tumor progression in bone

Bone metastases are one of the main complications of prostate cancer and they are incurable. We investigated whether and how estrogen receptor-related receptor alpha (ERRα) is involved in bone tumor progression associated with advanced prostate cancer. By meta-analysis, we first found that ERRα expression is correlated with castration-resistant prostate cancer (CRPC), the hallmark of progressive disease. We then analyzed tumor cell progression and the associated signaling pathways in gain-of-function/loss-of-function CRPC models in vivo and in vitro. Increased levels of ERRα in tumor cells led to rapid tumor progression, with both bone destruction and formation, and direct impacts on osteoclasts and osteoblasts. VEGF-A, WNT5A and TGFβ1 were upregulated by ERRα in tumor cells and all of these factors also significantly and positively correlated withERRα expression in CRPC patient specimens. Finally, high levels of ERRα in tumor cells stimulated the pro-metastatic factor periostin expression in the stroma, suggesting that ERRα regulates the tumor stromal cell microenvironment to enhance tumor progression. Taken together, our data demonstrate that ERRα is a regulator of CRPC cell progression in bone. Therefore, inhibiting ERRα may constitute a new therapeutic strategy for prostate cancer skeletal-related events.

Lysyl Oxidase Is a Strong Determinant of Tumor Cell Colonization in Bone

Lysyl oxidase (LOX) is a secreted copper-dependent amine oxidase whose primary function is to drive collagen crosslinking and extracellular matrix stiffness. LOX in colorectal cancer synergizes with hypoxia-inducible factor-1 (HIF-1) to promote tumor progression. Here we investigated whether LOX/HIF1 endows colorectal cancer cells with full competence for aggressive colonization in bone. We show that a high LOX expression in primary tumors from patients with colorectal cancer was associated with poor clinical outcome, irrespective of HIF-1 In addition, LOX was expressed by tumor cells in the bone marrow from colorectal cancer patients with bone metastases. In vivo experimental studies show that LOX overexpression in colorectal cancer cells or systemic delivery of the conditioned medium from LOX-overexpressing colorectal cancer cells promoted tumor cell dissemination in the bone marrow and enhanced osteolytic lesion formation, irrespective of HIF-1 Conversely, silencing or pharmacologic inhibition of LOX activity blocked dissemination of colorectal cancer cells in the bone marrow and tumor-driven osteolytic lesion formation. In vitro, tumor-secreted LOX supported the attachment and survival of colorectal cancer cells to and in the bone matrix, and inhibited osteoblast differentiation. LOX overexpression in colorectal cancer cells also induced a robust production of IL6. In turn, both LOX and IL6 were acting in concert to promote RANKL-dependent osteoclast differentiation, thereby creating an imbalance between bone resorption and bone formation. Collectively, our findings show that LOX supports colorectal cancer cell dissemination in the bone marrow and they reveal a novel mechanism through which LOX-driven IL6 production by colorectal cancer cells impairs bone homeostasis. Cancer Res; 77(2); 268-78. ©2016 AACR.

Tumour-derived miRNAs and bone metastasis

Skeletal metastases are complications of epithelial cancers, among which breast, prostate and lung carcinomas are the most osteotropic. In primary tumours, a subset of cancer cells undergoes epithelial-mesenchymal transition, acquires mobility to migrate into the surrounding stroma and seeds at distant sites to grow. The specific development of bone metastasis requires the recruitment of circulating tumour cells in the bone marrow, their adaptation to survive in the surrounding microenvironment where they alter the functions of osteoclasts and osteoblasts, and hijack signals coming from the bone matrix. Each of the molecular pathways underlining these steps is regulated by multiple factors, through the tight control of genes expressed by cancer cells interacting with cells from the bone microenvironment. In this context, miRNAs can act as master regulators of gene expression to control multiple aspects of bone metastasis formation, including cancer cell escape from the primary tumour site, cancer cell dissemination to bone and invasion of the bone marrow, as well as secondary outgrowth and tumour-stroma cell interactions. In the clinic, specific miRNA signatures have been identified in osteotropic cancer cells, raising the possibility that miRNAs could be used as biomarkers of bone metastasis. The regulatory activity of miRNAs in the bone microenvironment also suggests that miRNAs could be promising therapeutic targets.

TWIST1 expression in breast cancer cells facilitates bone metastasis formation

The transcription factor TWIST1 induces epithelial-mesenchymal transition and/or escape to the oncogenic-induced failsafe program, facilitating the intravasation of breast cancer cells in the systemic circulation and their dissemination to the lungs. Its involvement in breast cancer bone metastasis is unknown. To address this question, human osteotropic MDA-MB-231/B02 breast cancer cells were stably transfected with a Tet-inducible vector encoding for TWIST1, whose expression was specifically repressed in the presence of doxycycline (dox). The intra-arterial inoculation of transfectants expressing TWIST1 in immunodeficient mice substantially increased the extent of osteolytic lesions in these animals, being 50% larger than that of animals bearing mock-transfected tumors, as determined by radiography. This difference was accompanied by a sharp reduction of the bone volume (indicating a higher bone destruction) and a twofold increase in the tumor volume compared with mice bearing mock-transfected tumors, as determined by histomorphometry. Importantly, the suppression of TWIST1 expression in MDA-MB-231/B02 cells in the presence of dox abolished the stimulatory effect of TWIST1 on bone metastasis formation in vivo. Additionally, examination of the bone marrow from untreated and dox-treated animals on day 7 after tumor cell inoculation, at which time there was no evidence of radiographic osteolytic lesions, revealed that the number of tumor cell colonies that were recovered from the bone marrow of untreated mice was dramatically increased compared with that of dox-fed animals. In vitro, TWIST1 expression promoted tumor cell invasion and enhanced microRNA 10b (miR-10b) expression, a proinvasive factor, but was dispensable for growth of tumor cells. In vivo, the repression of miR-10b substantially decreased the presence of TWIST1-expressing breast cancer cells in the bone marrow. Overall, these results establish that TWIST1 facilitates breast cancer bone metastasis formation through a mechanism dependent of miR-10b, which leads to increase tumor burden and bone destruction.

Abstract P5-04-13: Antibody-based therapy targeting integrin a5 is an effective strategy to treat experimental breast cancer bone metastasis

Background: Integrin a5β1 is a specific fibronectin receptor that is often upregulated in breast cancer cells undergoing epithelial-to-mesenchymal transition. Preclinical evidence showed that α5β1-fibronectin interaction promotes in vitro the survival of growth-arrested breast cancer cells in the bone marrow. This is in line with the observation that disseminated tumor cells (DTCs) in the bone marrow, that are the earliest sign of development of metastatic disease in patients, express α5β1 integrin. However, the role of α5β1 in bone marrow micrometastasis formation remains unknown. Here, we investigated its role in experimental breast cancer bone metastasis, using an antibody-based strategy targeting the α5 subunit.

Methods: Integrin α5 mRNA expression levels were quantified by qRT-PCR, using radically resected primary tumors of 427 breast cancer patients. Comparison between Kaplan-Meier curves was performed using the log-rank test. The Cox model was used for multivariate analysis. For experimental metastasis, female BALB/c immunodeficient mice were treated with a chimeric IgG4 monoclonal antibody that specifically binds to human integrin subunit α5 (M200;15mg/kg) or the vehicle. Treatment was performed 3 times per week starting the day before intra-arterial inoculation with luciferase-expressing human MDA-MB-231/B02 breast cancer cells, which selectively metastasize to bone. Vehicle and M200-treated mice were analyzed by radiography and bioluminescence. On day 28 after tumor cell inoculation, animals were sacrificed and histomorphometric analysis of metastatic legs was performed. Alternatively, animals were culled on day 7 after tumor cell inoculation, and the bone marrow was flushed for DTC colony assay. In-vitro cell-based assays were conducted to explore the functions of α5β1 on adhesion, invasion and survival. Statistical analysis was carried out by performing a Mann-Whitney U test.

Results: Median durations of metastasis-free survival were 6.2 (high α5 expression) and 9.5 years (low α5 expression) (P = 0.0008). Additionally, compared with low expression, a high α5 expression was associated with shorter bone metastasis-free survival, both in univariate (P = 0.024) and multivariate analysis (HR = 1.65,95%CI = 1.02 to 2.67; P = 0.04).The treatment of tumor-bearing animals with M200 antibody statistically significantly delayed the onset of skeletal lesions (P = 0.02) and caused a 50% reduction in the extent of osteolytic lesions (P = 0.038), compared with vehicle. This difference was accompanied with a sharp reduction of tumor burden (P = 0.02), as determined by bioluminescence. Histomorphometric analysis of metastatic legs showed that M200 treatment decreased skeletal tumor burden (P = 0.027) and increased the bone volume (P = 0.02), compared with vehicle. Additionally, the number of DTC colonies in the bone marrow from M200-treated mice was dramatically decreased compared with vehicle (P&lt;0.001). In vitro, M200 antibody did not affect tumor cell survival. By contrast, it specifically inhibited fibronectin-mediated tumor cell adhesion and invasion.

Conclusion: Our results suggest that α5β1 integrin expression in breast cancer cells facilitates bone marrow micrometastasis formation and the subsequent development of osteolytic lesions.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-04-13.

Impact of connexin32 deletion on E7 or RET/PTC3 oncogene-driven growth and neoplastic transformation of the thyroid gland

Connexins (Cx) form gap junctions and allow direct cell-to-cell communication. Cx through gap junctions or by themselves play regulatory roles on cell growth and differentiation. Using genetically modified mice, we previously found that Cx32 acts as a down-regulator of growth in normal thyroid gland. In this study, we examined the impact of Cx32 ablation on oncogene-driven thyroid growth and neoplastic transformation. Cx32 knockout (Cx32-KO) mice were crossed with transgenic mice expressing, selectively in the thyroid gland, either the E7 or RET/PTC3 (RP3) oncogene. As already described, Cx32-KO mice had no detectable thyroid alteration in physiological conditions and mice expressing E7 or RP3 exhibited time-dependent thyroid hypertrophy and variable changes in expression of differentiation. The thyroid of E7 mice evolved towards a large colloid goitre whereas RP3 mice developed a hyperplastic thyroid of variable size, and the largest glands (about 40% of total) represented a profound tissue remodeling with proliferative papillary formations. E7-induced thyroid hypertrophy was reduced by about 40% in Cx32-KO mice as compared with wild-type (WT) littermates. On the contrary, thyroid hypertrophy induced by thyrotropin stimulation (in response to goitrogen treatment) was enhanced by about 40% in Cx32-KO mice as compared with WT mice. Thyroid hypertrophy of RP3 mice and the proportion of glands showing extensive tissue remodeling were drastically reduced in mice devoid of Cx32. Our data show that Cx32, which negatively controls thyroid growth activated by thyrotropin via the cAMP pathway, would act as a positive effector of thyroid growth triggered by oncogenes acting through other signaling cascades.

Effect of dietary argan oil on fatty acid composition, proliferation, and phospholipase D activity of rat thymocytes

OBJECTIVE

Argan oil is receiving increasing attention due to its potential health benefits in the prevention of cardiovascular risk, but no information to date is available about its possible effect on immune cells and functions.

METHODS

To address this issue male rats were fed one of five diets that contained fish oil, argan oil, olive oil, coconut oil, or sunflower oil for 4 wk. The fatty acid composition of plasma and thymocyte lipids was then analyzed in relation to the mitogen-induced proliferation and phospholipase D (PLD) activity of thymocytes.

RESULTS

The 18:2omega-6 proportion in thymocyte phospholipids from rats fed argan oil was significantly lower than that observed in phospholipids from rats fed sunflower oil and fish oil but higher than that found in the olive oil and coconut oil groups. Further, a significant positive linear relation was found between thymocyte proliferation and the 18:2omega-6 proportion in thymocyte phospholipids, whatever the diet. The proliferation response of thymocytes to mitogenic activation was also inversely correlated to PLD activity measured in intact thymocytes. Subsequent western blotting experiments indicated that the diet-induced variations in PLD activity mainly reflected variations in the expression of PLD2 protein.

CONCLUSIONS

On the whole, the present study shows that the effects of argan oil on immune cells are very similar to those of olive oil, and that, as a consequence, argan oil can be used as a balanced dietary supply without marked adverse effects on immune cell function.

Oxidized phospholipid: POVPC binds to platelet-activating-factor receptor on human macrophages. Implications in atherosclerosis

Atherosclerosis as a chronic inflammatory disease resulting from the imbalance of the pro- and anti-inflammatory factors in the vessel wall. PAF and PAF-like oxidized phospholipids generated upon LDL oxidation in the intima of the arteries may interact with infiltrated monocytes/macrophages and lead to the alteration of gene expression patterns accompanied by an impaired production of chemokines, interleukins and proteolytic and lipolytic enzymes. The aim of this study was to evaluate the binding capacity of the major component of PAF-like oxidized phospholipids, namely the 1-palmitoyl-2-oxovaleroyl-sn-glycero-3-phosphorylcholine (POVPC) to PAF-receptor (PAF-R) on the surface of human monocytes/macrophages and to further characterize the gene expression induced by such binding. We show that, POVPC binds to cultured human macrophages via PAF-R and transduces the signals leading to the intracellular Ca(2+) fluxes and modifies the transcription levels of numerous pro-inflammatory and pro-atherogenic genes. Although a some similarity of the gene expression patterns was observed when macrophages were activated with POVPC versus PAF, we observed that only POVPC treatment induced a several-fold activation of IL-8 gene. In turn, only PAF activated PAF-R, matrix metalloproteinase-13 and 15-lipoxygenase mRNA accumulation. Thus, we suggest, that POVPC signals in mature macrophages only in part through the PAF-R, a part of its effects may involve other receptors.

Induction of control genes in intestinal gluconeogenesis is sequential during fasting and maximal in diabetes

We studied in rats the expression of genes involved in gluconeogenesis from glutamine and glycerol in the small intestine (SI) during fasting and diabetes. From Northern blot and enzymatic studies, we report that only phosphoenolpyruvate carboxykinase (PEPCK) activity is induced at 24 h of fasting, whereas glucose-6-phosphatase (G-6-Pase) activity is induced only from 48 h. Both genes then plateau, whereas glutaminase and glycerokinase strikingly rebound between 48 and 72 h. The two latter genes are fully expressed in streptozotocin-diabetic rats. From arteriovenous balance and isotopic techniques, we show that the SI does not release glucose at 24 h of fasting and that SI gluconeogenesis contributes to 35% of total glucose production in 72-h-fasted rats. The new findings are that 1) the SI can quantitatively account for up to one-third of glucose production in prolonged fasting; 2) the induction of PEPCK is not sufficient by itself to trigger SI gluconeogenesis; 3) G-6-Pase likely plays a crucial role in this process; and 4) glutaminase and glycerokinase may play a key potentiating role in the latest times of fasting and in diabetes.

Lysophosphatidylcholine as a preferred carrier form of docosahexaenoic acid to the brain

The metabolic fate of docosahexaenoic acid (DHA) was evaluated from its intake as a nutrient in triglycerides and phosphatidylcholines to its uptake by target tissues, especially the brain. Several approaches were used including the kinetics and tissue distribution of ingested 13C-labeled DHA, the incorporation of radiolabeled DHA injected as its nonesterified form compared to the fatty acid esterified in lysophosphatidylcholine (lysoPC), and the capacity of the two latter forms to cross a reconstituted blood-brain barrier (BBB) consisting of cocultures of brain-capillary endothelial cells and astrocytes. The results obtained allow us to raise the hypothesis that lysoPC may represent a preferred physiological carrier of DHA to the brain.

Rat small intestine is an insulin-sensitive gluconeogenic organ

At variance with the current view that only liver and kidney are gluconeogenic organs, because both are the only tissues to express glucose-6-phosphatase (Glc6Pase), we have recently demonstrated that the Glc6Pase gene is expressed in the small intestine in rats and humans and that it is induced in insulinopenic states such as fasting and diabetes. We used a combination of arteriovenous balance and isotopic techniques, reverse transcription-polymerase chain reaction, Northern blot analysis, and enzymatic activity assays. We report that rat small intestine can release neosynthesized glucose in mesenteric blood in insulinopenia, contributing 20-25% of total endogenous glucose production. Like liver glucose production, small intestine glucose production is acutely suppressed by insulin infusion. In the small intestine, glutamine and, to a much lesser extent, glycerol are the precursors of glucose, whereas alanine and lactate are the main precursors in liver. Accounting for these metabolic fluxes: 1) the phosphoenolpyruvate carboxykinase gene (required for the utilization of glutamine) is strongly induced at the mRNA and enzyme levels in insulinopenia; 2) the glycerokinase gene is expressed, but not induced; 3) the pyruvate carboxylase gene (required for the utilization of alanine and lactate) is repressed by 80% at the enzyme level in insulinopenia. These studies identify small intestine as a new insulin-sensitive tissue and a third gluconeogenic organ, possibly involved in the pathophysiology of diabetes.

Induction of PEPCK gene expression in insulinopenia in rat small intestine

PEPCK is a key enzyme of gluconeogenesis in liver and kidney. Recently, we have shown that small intestine also contributes to the endogenous glucose production in insulinopenia in rats and that glutamine is the main precursor of glucose synthesized in this tissue. The expression of the PEPCK gene in rat and human small intestine and the effect of streptozotocin-induced diabetes and fasting have been studied using reverse transcriptase-polymerase chain reaction, Northern blot analysis, and determination of enzyme activity. The PEPCK gene is expressed along the whole small intestine in adult rat and human. The abundance of PEPCK mRNA was increased approximately 30 times in the duodenum, 15 times in the jejunum, and only 3 times in the ileum from diabetic rats. PEPCK mRNA was downregulated in all parts of the tissue upon insulin treatment for 10 h. In 48-h fasted rats, the PEPCK mRNA abundance was increased 17 times in the duodenum and the jejunum and 3 times in the ileum, and it was normalized upon refeeding for 7 h. PEPCK activity was also increased 2-5 times in diabetic and fasted rats in the duodenum and jejunum but not in the ileum. We conclude that PEPCK is a crucial enzyme contributing to the induction of gluconeogenesis in small intestine, just as it is well known to be in the liver and kidney.

Characterization of plasma unsaturated lysophosphatidylcholines in human and rat

Unsaturated lysophosphatidylcholines (lysoPtdCho) bound to albumin circulate in blood plasma and seem to be a novel transport system for carrying polyunsaturated fatty acids (PUFA) to tissues that are rich in these fatty acids, such as the brain. The potential of these lysoPtdCho as a significant source of PUFA for cells has been assessed by comparing their plasma concentration with that of unsaturated non-esterified fatty acids (NEFA) bound to albumin. In humans, the PUFA concentration was 25.9+/-3.1 nmol/ml for these lysoPtdCho, compared with 33.4+/-9.6 nmol/ml for NEFA; in rats the equivalent values are 14.2+/-0.6 and 13.1+/-1.1 nmol/ml respectively (means+/-S.E.M.). The lysoPtdCho arachidonic acid content was 2-fold (human) and 5-fold (rat) higher than that of NEFA. In human and rat plasma, unsaturated lysoPtdCho were associated mainly with albumin rather than lipoproteins. The rate and extent of the acyl group shift from the sn-2 to sn-1 position of these lysoPtdCho were studied by the incubation of 1-lyso, 2-[(14)C]C(18:2)n-6-glycerophosphocholine (GPC) with plasma. The rapid isomerization of this lipid occurred at pH 7 (20% isomerization within 2 min) and was not prevented by its association with albumin. The position of the acyl group in the lysoPtdCho circulating in plasma was studied by collecting blood directly in organic solvents containing 1-lyso,2-[(14)C]C(18:2)n-6-GPC as a marker of isomerization that occurred during sampling and analysis. Approx. 50% of the PUFA was located at the sn-2 position, demonstrating that substantial concentrations of 2-acyl-lysoPtdCho are present in plasma and are available for tissue uptake, where they can be reacylated at the sn-1 position to form membrane phospholipids.

Synthesis of acetyl,docosahexaenoyl-glycerophosphocholine and its characterization using nuclear magnetic resonance

Docosahexaenoic acid (DHA) circulates in mammals in lipoproteins and bound to serum albumin as a nonesterified fatty acid as well as esterified in lysophosphatidylcholine (lysoPC). 1-Lyso,2-DHA-glycerophosphocholine (GPC) is an unstable isomer because of a primary alcohol at the sn-1 position. To keep DHA at the sn-2 position of lysoPC, its usual position for the corresponding lysoPC to be acylated into PC in tissues, we synthesized 1-acetyl,2-DHA-GPC and confirmed its structure by use of nuclear magnetic resonance (NMR) spectroscopy in comparison with its positional isomer, 1-DHA,2-acetyl-GPC. 1-Lyso,2-DHA-GPC was prepared from 1-stearoyl,2-DHA-GPC by enzymatic hydrolysis and purified by high-performance liquid chromatography. The isomerization of 1-lyso,2-DHA-GPC into 1-DHA,2-lyso-GPC was obtained by keeping the former overnight at room temperature under nitrogen. Both lysoPC isomers were acetylated by acetic anhydride into 1-acetyl,2-DHA-GPC and 1-DHA,2-acetyl-GPC, respectively, and the resulting phospholipids were fully characterized by NMR. In particular, the 1,2 substitution pattern of the acetyl and DHA chains could be easily detected by 2D heteronuclear multibond correlation. We conclude that 1-acetyl,2-DHA-GPC might be considered as a stable form of 1-lyso,2-DHA-GPC for its delivery to tissues, if the latter exhibits acetyl hydrolase activity.

Inhibition of prostaglandin H synthase and activation of 12-lipoxygenase by 8,11,14,17-eicosatetraenoic acid in human endothelial cells and platelets

The effects of the marine fatty acid 20:4n-3, an isomer of arachidonic acid (20:4n-6), have been compared to that of 20:5n-3 on 20:4n-6 oxygenation in human platelets and endothelial cells. In platelets, 20:4n-3 added along with 20:4n-6 was as potent as 20:5n-3 in inhibiting prostaglandin H synthase (PGH synthase) activity. From 2.5- to 10 microM of 20:4n-6, the synthesis of thromboxane B2 and 12-hydroxy-5,8,10-heptadecatrienoic acid, reflecting the PGH/thromboxane synthase activity, was lowered by 5 and 10 microM of both fatty acids. In contrast, 20:4n-3, but not 20:5n-3, strongly stimulated the lipoxygenase activity at each concentration of 20:4n-6 used whatever the amount of 20:4n-3 added. The effects of both n-3 polyunsaturated fatty acids on endothelial cell PGH/prostacyclin synthases were compared after 2- and 24-hr incubation with the cells, leading to moderate (2 hr) and high (24 hr) concentrations of these fatty acids in membrane phospholipids. The incorporation of 20:4n-3 and 20:5n-3 occurred mostly in phosphatidylcholine and phosphatidylethanolamine and did not alter the 20:4n-6 level of phospholipid classes after 2-hr supplementation, whereas it was drastically decreased after 24 hr. The synthesis of prostacyclin obtained after cell stimulation by 0.1 U/mL thrombin was unaffected by the fatty acid modifications induced after 2-hr supplementation, whereas it was strongly depressed after 24 hr. It was concluded that 20:4n-3 is not an agonist for platelet activation, despite its close structural analogy with 20:4n-6, and is as potent as 20:5n-3 in inhibiting PGH synthase activities, showing that the double bond at C5 is not necessary for inhibition. In contrast, the oxygenation of 20:4n-6 by 12-lipoxygenase was stimulated by 20:4n-3 but not by 20:5n-3, which might be related to the efficient oxygenation of 20:4n-3 by this enzyme compared with 20:5n-3.

Human plasma albumin transports [13C]docosahexaenoic acid in two lipid forms to blood cells

Docosahexaenoic acid (22:6) decreases blood platelet function and is highly concentrated in the brain where its depletion leads to functional impairments. Because the platelets and blood brain barrier capillary endothelium cannot hydrolyze the complex lipids for fatty acid (FA) uptake, nonesterified FA (NEFA) bound to albumin are assumed to be the delivery route of FA to these cells. The supply of 13C-labeled 22:6 to blood cells by plasma albumin was studied in humans after a single ingestion of this FA esterified in a triglyceride (TG). The 22:6 13C/12C ratio, measured by gas chromatography combustion-isotope ratio mass spectrometry was measured in lipid classes from albumin, platelets, leukocytes, and erythrocytes (taken as a tentative index of the brain uptake). Nonesterified [13C]22:6 bound to albumin was rapidly produced after ingestion, as a result of the hydrolysis of very low density lipoprotein (VLDL) plus chylomicron TG. We found that albumin carried another source of 22:6, lyso-phosphatidylcholines (lyso-PC), in which [13C]22:6 accumulated while the nonesterified [13C]22:6 reached its minimal plasma concentrations. Computation of the relative contribution of NEFA and lyso-PC for the [13C]22:6 delivery to platelets and erythrocytes showed that the [13C]22:6 supply to platelets occurred uniquely through NEFA, whereas this pool was weakly involved in the delivery to erythrocytes. In contrast, lyso-PC was uniquely concerned with the 22:6 delivery to erythrocytes and represented the major part of this supply. We conclude that plasma albumin carries 22:6 in two lipid forms that are involved differently in the delivery of this FA to target cells.

Retroconversion and metabolism of [13C]22:6n-3 in humans and rats after intake of a single dose of [13C]22:6n-3-triacylglycerols

The apparent retroconversion of docosahexaenoic acid (22:6n-3) to eicosapentaenoic acid (20:5n-3) and docosapentaenoic acid (22:5n-3) was studied in vivo, in rats and humans, after they ingested a single dose of triacylglycerols containing [13C]22:6n-3 ([13C]22:6-triacylglycerol), without 22:6n-3 dietary supplementation. The amount of apparent retroconversion and the distribution of the three n-3 polyunsaturated fatty acids (PUFAs) in plasma lipid classes were followed as a function of time by measuring the appearance of 13C in these PUFAs with gas-chromatography combustion-isotope ratio mass spectrometry. This [13C]22:6n-3 retroconversion, calculated by summing the amounts of [13C]22:5n-3 and [13C]20:5n-3 in plasma lipids, was lower in humans than in rats, reaching a maximum of approximately 9% of the total plasma [13C]22:6n-3 in rats, but only 1.4% in humans. The incorporation of [13C]22:6n-3 and [13C]22:5n-3 in lipid classes followed their endogenous distribution with a maximal accumulation in phospholipids, but a low incorporation into cholesterol esters (CEs), whereas [13C]20:5n-3 was equally present in phospholipids and CEs. The ratio of the amount of HDL-CE to HDL-phosphatidylcholine for [13C]20:5n-3 was higher than for [13C]22:6n-3, indicating a selectivity of the lecithin-cholesterol acyltransferase enzyme with regard to these PUFAs, which may be related to the differences in their biological properties after fish oil feeding. The occurrence of a weak basal 22:6n-3 retroconversion in humans supports feeding this pure PUFA in cases in which 20:5n-3 presents undesirable side effects and when specific alterations of blood lipids are expected.

Metabolic fate of an oral tracer dose of [13C]docosahexaenoic acid triglycerides in the rat

The appearance of 13C in rat lipoprotein, blood cells, and brain lipids was followed as a function of time after the ingestion of triglycerides (TG) containing [13C]22:6n-3. The time course of 13C abundance in 22:6n-3 of various lipid pools, measured by gas chromatography combustion-isotope mass spectrometry, established precursor-product relationships within lipids. The [13C]22:6n-3 was rapidly incorporated into very low density lipoprotein-chylomicron-TG and unesterified fatty acids bound to albumin, with a concomitant maximal appearance at 3 h and further decline. Lysophosphatidylcholines (lysoPC) bound to albumin were also enriched in [13C]22:6n-3, and their labeling appeared to be mainly due to hepatic secretion at the earliest time points. From 12 h postingestion, the synthesis of [13C]22:6n-3-lysoPC was twice as high as that of unesterified [13C]22:6n-3, making lysoPC a potential source of 22:6n-3 supply for tissues. The labeling of platelets, red blood cells, and brain phospholipids presented different kinetics, presumably involving the two lipid forms of [13C]22:6n-3 bound to albumin, to different extents. We conclude that [13C]22:6n-3 esterified in TG is rapidly redistributed within blood lipoproteins and the albumin fraction and that its incorporation in lipid species bound to albumin influences its uptake by target tissues.

Metabolic fate of an oral long-chain triglyceride load in humans

To determine the steps involved in the metabolism of ingested triglycerides (TG), 10 healthy women were studied during 6 h after ingestion of 30 g olive oil labeled with [1,1,1-13C3] triolein. The appearance of 13C was followed in chylomicron-TG (CM-TG), nonesterified fatty acid (NEFA), very low-density lipoprotein (VLDL)-TG, and in expired gas. Indirect calorimetry was used to determine total lipid oxidation. After 90 min, labeling was higher in CM-TG than in NEFA or VLDL. At 180 min, a plateau of enrichment was obtained for CM-TG and NEFA, demonstrating the entry of exogenous lipids in the NEFA pool. After 300 min, a plateau was observed for VLDL-TG with levels of enrichment (0.38 +/- 0.04%) similar to those observed for NEFA (0.36 +/- 0.03%), suggesting a precursor-product relationship. Only 19 +/- 2% of the load was oxidized. From 300 to 360 min, 70% of total lipid oxidation was from exogenous TG. We conclude that, after ingestion of a lipid load, a cycle of fatty acids-TG occurs from CM to NEFA and from NEFA to VLDL. Furthermore, this lipid load has a sparing effect on endogenous lipid stores.

In vivo compartmental metabolism of 13C-docosahexaenoic acid, studied by gas chromatography-combustion isotope ratio mass spectrometry

The exchange of docosahexaenoic acid (22:6n-3) within lipid pools in rat and human has been followed as a function of time after the ingestion of triglycerides (TG) containing 22:6n-3 labeled with 13C(13C 22:6n-3). The 13C abundance in the fatty acid was measured by gas-chromatography-combustion isotope ratio mass spectrometry which allowed the detection of 0.001 atom 13C percent 12C. The 13C 22:6n-3 appearance was rapid in the TG of very low density lipoprotein plus chylomicron fraction, in which the maximal labeling was observed at 3 and 2 h after ingestion in rat and human, respectively. Concomitant with the TG utilization of this fraction by lipoprotein lipase from tissues, unesterified 13C 22:6n-3 appeared in the plasma albumin. 13C 22:6n-3 bound to albumin was mostly present in unesterified form before 12 h post-ingestion while after that period, lysophosphatidylcholine (lysoPC) bound to albumin carried higher 13C 22:6n-3 concentrations. These lyso-PC were mostly from hepatic origin and might represent a potential source of 22:6n-3 redistribution to tissues. The 13C 22:6n-3 uptake into rat brain PC and phosphatidylethanolamine was still increasing when the concentration of plasma unesterified 13C 22:6n-3 had already dropped to a minimal plateau value and during the period of maximal plasma circulation of 13C 22:6n-3-lysoPC bound to albumin. In contrast, the uptake of 13C 22:6n-3 into blood platelet PC occurred during the phase of important circulation of 13C-22:6n-3 bound to albumin, suggesting the in vivo efficiency of the Lands pathway for this fatty acid. It is concluded that 13C 22:6n-3 esterified in TG is rapidly absorbed and redistributed within plasma lipoproteins and that its redistribution within the two lipid species bound to albumin might influence its uptake by platelets and rat brain.

Selective modifications of the phospholipid fatty acid composition in human platelet membranes using nonspecific and specific lipid transfer proteins

In order to specifically modify the fatty acid composition of cell membrane phospholipids, we have developed an original method based on the transfer of pure phospholipid molecular species to membranes. Phosphatidylcholine (PC) and phosphatidylethanolamine (PE) subclasses containing 18:2n-6 and 22:6n-3 at the sn-2 position were incorporated into human platelet membranes using the endogenous phosphatidylinositol/PC transfer protein (PI/PC-TP) and the phospholipid transfer protein from maize (L-TP), respectively. PI/PC-TP was shown to catalyze a strict exchange of phospholipids between platelet membranes and unilamellar vesicles containing 1,2-diacylglycerophosphocholine (diacyl-GPC; 16:0/18:2-GPC, or 16:0/22:6-GPC). The proportions of 18:2n-6 and 22:6n-3 in diacyl-GPC of platelet membranes were gradually increased from 10.7 to 16.9% and from 0.8 to 10.1%, respectively, whereas the PE and PI fatty acid compositions were not changed. The diacyl-GPC enrichment in 22:6n-3 and 18:2n-6 did not induce changes in membrane fluidity parameters measured by electron-spin resonance of 5- and 16-nitroxy stearic acids. Similarly, 18:2n-6 and 22:6n-3 esterified in 1,2-diacylglycerophosphoethanolamine (diacyl-GPE) have been incorporated in platelet membranes by an apparent exchange process under conditions where donor vesicles had a phospholipid composition equivalent to that of platelet membranes. The proportions of 18:2n-6 and 22:6n-3 were selectively and progressively increased from 6.0 to 21.2% and from 2.2 to 17.2%, respectively, in diacyl-GPE of platelet membranes. Thus, the L-TP- and PI/PC-TP-catalyzed enrichment can be used for studying the modulation of membrane biological activities by defined changes of fatty acid composition of specific phospholipid classes or subclasses.

Decrease of brain phospholipid synthesis in free-moving n-3 fatty acid deficient rats

The autoradiographic method with [14C]-docosahexaenoic acid ([14C]22:6 n-3) was used to determine whether a diet deficient in n-3 fatty acids, inducing a decrease in 22:6 n-3 circulating level, was associated with changes in local rates of phospholipid synthesis in the rat brain. As compared with rats fed a normal diet (peanut plus rapeseed oil), a n-3 fatty acid deficiency [peanut oil group (P group)] induced a generalized decrease (-35 to -76%) of 22:6 n-3 incorporation rates into phospholipids in all the regions examined. This effect was confirmed by using [3H]22:6 n-3 infusion by biochemical analysis and quantifications corrected for the contribution of docosahexaenoate derived from lipid store recycling to the unesterified pool, taken as the precursor pool for phospholipid synthesis in the whole brain. In normal or n-3 fatty acid-deficient rats, the values of the brain-to-plasma 22:6 n-3 specific activity ratio (psi) were similar (0.03), indicating that a considerable endogenous source of 22:6 n-3 (97%), likely derived from phospholipid degradation, dilutes the specific activity of the tracer coming from plasma. Using the specific activity of 22:6 n-3 in plasma instead of brain would thus lead to a gross underestimation of the rate of phospholipid synthesis. The results also demonstrate that the pattern of 14C or 3H distribution in brain lipids was not modified by the n-3 fatty acid-deficient diet. The major lipids labeled were phospholipids, particularly phosphatidylethanolamine. Nevertheless, the unesterified 22:6 n-3 concentrations in plasma and brain were significantly reduced (eight-and threefold, respectively) in the P group. In addition, the proportion of 22:6 n-3 in the brain total lipid fraction, total phospholipids, and phosphatidylcholine, -ethanolamine, and -serine was significantly decreased in n-3 fatty acid-deficient rats. This was partially compensated for by an increase in the 22:5 n-6 level. These results are discussed in relation to the limitation of 22:6 n-3 use to quantify, by the quantitative autoradiographic method, changes in local rates of phospholipid synthesis in rat brain.

Effect of specific phospholipid molecular species incorporated in human platelet membranes on thromboxane A2/prostaglandin H2 receptors

The incorporation of albumin-bound docosahexaenoic acid (22:6n-3), but not linoleic acid (18:2n-6), into cellular phospholipids inhibits platelet aggregation induced by the thromboxane analogue U46619. [3H]U46619 specific binding to thromboxane A2/prostaglandin H2 (TXA2/PGH2) receptors, as well as specific binding of the antagonist [3H]SQ29548 to these sites were also decreased in these modified cells (P. G., Swann et al. 1990. J. Biol. Chem. 265: 21692-21697). More than 80% of the 22:6n-3 incorporated in these cells was esterified in the various endogenous phospholipid classes and the remaining was found in neutral lipids and in the unesterified fatty acid pool. In this study, we determined whether the effects observed could be attributed to the esterification of 22:6n-3 in phospholipids and whether the 22:6n-3 biological activity might depend on its esterification in specific phospholipid classes. Therefore, pure phosphatidylcholine (PC) and phosphatidylethanolamine (PE) molecular species were transferred to platelet membranes, using lipid transfer proteins. PC and PE containing palmitate (16:0) and 22:6n-3 or 16:0 and 18:2n-6 at position sn-1 and sn-2, respectively, were incorporated into membranes only at the expense of the corresponding endogenous phospholipid class, by an apparent exchange process. When such modified membranes were tested for specific binding of U46619 and SQ29548, a significant decrease of the receptor site affinity was only observed in membranes highly enriched with 1-palmitoyl-2-docosahexaenoyl-glycerophosphocholine (16:0/22:6-GPC). Fluidity parameters measured by electron spin resonance of 5- and 16-nitroxy-stearic acids were not significantly different in membranes enriched with 16:0/22:6-GPC relative to those enriched with 16:0/18:2n-6-GPC, arguing against a generalized perturbation of the membrane due to 22:6n-3 incorporation. We conclude that molecular species of PC with 22:6n-3 at the sn-2 position can affect TXA2/PGH2 receptors. The selectivity of the inhibitory effect of PC containing 22:6n-3 is discussed.

Stable isotope tracer and gas-chromatography combustion isotope ratio mass spectrometry to study the in vivo compartmental metabolism of docosahexaenoic acid

A gas-chromatography combustion isotope ratio mass spectrometry (GCC-IRMS) method using carbon 13 (13C)-stable isotope to trace n-3 polyunsaturated fatty acids (PUFA) turnover in vivo is presented. Natural 13C abundance of commercial n-3 PUFA was measured from 100 to 300 ng of fatty acids and was -27.58, -27.83, and -28.16 for 22:6n-3, 22:5n-3, and 20:5n-3, expressed as delta 13C /1000 versus Pee Dee Belemnite (PDB), respectively. Precision of delta 13C /1000 values was comparable for the three PUFA and gave relative standard deviations of 0.95-0.97%. Isotope enrichment of 0.0010 at.% could be detected. Triglycerides enriched in [13C]22:6n-3 ([13C]22:6-TG) were synthesized by growing a microalgae on [1-13C]glucose. [13C]22:6n-3 represented 36 wt.% of total triglyceride fatty acids and had an isotope enrichment of 2.0420 at.%, which was the double of natural abundance. The isotope enrichment of 22:6n-3 in lipids from rat lipoproteins and red cells could be followed as a function of time after ingestion of 3 mg [13C]22:6-TG and showed specific patterns according to the lipid compartments. The retroconversion of [13C]22:6n-3 was also detected in HDL phosphatidylcholine by the appearance of [13C]22:5n-3 and [13C]20:5n-3. On the other hand, 22:6n-3 natural 13C abundance in human lipid classes of lipoproteins and blood cells has been measured using 10 ml plasma, even for the more limiting lipid compartments in terms of 22:6n-3 dose size.(ABSTRACT TRUNCATED AT 250 WORDS)

Incorporation of arachidonic and docosahexaenoic acids into phospholipids of rat brain membranes

The incorporation of [3H]arachidonic acid (20:4n-6) into rat brain membranes and its mobilization in response to norepinephrine, a relevant neuromediator were studied. The most efficient [3H]20:4n-6 incorporation was in inositol glycerophospholipids (PI) where it reached a plateau after 10 min incubation, while this incorporation was very weak in choline glycerophospholipids (PC). In contrast, the esterification of docosahexaenoic acid, another polyunsaturated fatty acid occurring at high level in brain, was similar in PI and PC, the incorporation in PI being 8-fold lower than that of 20:4n-6. The newly esterified [3H]20:4n-6 was exclusively found in the 1,2-diacyl subclasses of PI and PC. The bulk of incorporation was in the 18:0/20:4n-6 molecular species of 1,2-diacyl-glycerophosphoinositol and in 16:0/20:4n-6 + 18:1/20:4n-6 molecular species of 1,2-diacyl-glycerophosphocholine, which agrees with the usual location of 20:4n-6 in brain phospholipid classes. Upon norepinephrine treatment, [3H]20:4n-6 was not released from PC, but was dose-dependently decreased in PI, the release being significant from 10(-5) M of the agonist. These results suggest that 20:4n-6 exhibits a high specific turnover in brain PI and is mobilized from this class upon relevant neuromediator stimulation. The acellular system used preserved the specificity of enzymes catalyzing the polyunsaturated fatty acid incorporation and release and could be helpful for studying their turn over in brain.

Phospholipid molecular species from human placenta lipids

The phospholipid molecular species from a large-scale preparation of human placenta lipids were analyzed. The major placental phospholipids were choline glycerophospholipids (CPL) (53.2 wt%), sphingomyelin (21.7 wt%) and ethanolamine glycerophospholipids (EPL) (14.6 wt%). 1,2-Diacyl-glycerophosphocholine was the most abundant subclass of CPL (91.7 mol%), while EPL contained 1,2-diacyl (54.6 mol%) and 1-alk-1'-enyl-2-acyl (43.8 mol%) subclasses. The level of polyunsaturated fatty acids (PUFA) in total phospholipids was remarkably constant (38.4-39.9 mol%) within all placental batches tested. The long-chain PUFA, mainly 20:4n-6 and 22:6n-3 of the n-6 and n-3 series, respectively, were found in high proportion in all phospholipid classes, especially in EPL (46.7 mol%) and in inositol glycerophospholipids (IPL) (39.9 mol%). CPL and serine glycerophospholipids were much richer in 18:1n-9 and 18:2n-6. High levels of molecular species with arachidonic acid in the sn-2 position were found particularly in 1-alk-1'-enyl-2-acyl-glycerophosphoethanolamine (with 24.0 mol% 16:0 and 22.0 mol% 18:0 in sn-1 position) and in 1,2-diacyl glycerophosphoinositol with 42.6 mol% 18:0 in sn-1 position. EPL subclasses were rich in 22:6n-3, which occurs mainly as 16:0/22:6n-3 (11.7 mol%) in the plasmalogen form and as 18:0/22:6n-3, 16:0/22:6n-3 and 18:1/22:6n-3 in the diacyl forms. Based on their availability and composition, placental phospholipids could be of interest, for example, for supplementing artificial milk preparations with n-3 and n-6 long-chain PUFA for newborn infants with insufficiently developed 18:2n-6 and 18:3n-3 desaturation/elongation.

Interactions between arachidonic and eicosapentaenoic acids during their dioxygenase-dependent peroxidation

Eicosapentaenoic acid (EPA), a major polyunsaturated fatty acid of fish has been widely proposed as a potential nutrient for decreasing platelet-endothelial cell interactions and the subsequent atherogenesis and thrombogenesis. This is mainly based upon the decrease of arachidonic acid (AA) oxygenation into bioactive molecules like thromboxane A2. In addition, EPA may be oxygenated into its own active derivatives via cell dioxygenases. We report evidence for the requirement of specific peroxides, adequately provided by AA, to allow EPA to be oxygenated into its bioactive products like prostaglandin I3, a prostacyclin mimetic. On the other hand, we present some data that argue for a decreased basal AA dioxygenation (specific peroxidation) by small concentrations of EPA. The interactions between AA and EPA are then dual, EPA being able to counteract AA oxygenation whereas EPA requires AA to be efficiently oxygenated.

Modulation of cyclic nucleotide phosphodiesterase by dietary fats in rat heart

Feeding oils of different fatty acid composition modifies the fatty acid composition of cardiac membrane phospholipids, thereby inducing changes in cardiac contractility and altering response of adenylate cyclase to catecholamines. In the present study, the effect of such dietary manipulations on cyclic nucleotide phosphodiesterase, which is involved in the control of cyclic nucleotide intracellular levels and in the control of cardiac contractility, was investigated. Rats were fed either a saturated fatty acid-enriched diet (8 weight percent [%] coconut oil + 2% sunflower oil), an n-6 fatty acid-enriched diet (10% sunflower oil) or an n-3 fatty acid-enriched diet (8% fish oil + 2% sunflower oil). The fatty acid composition of cardiac phospholipids, as well as the nonesterified fatty acid content of heart were markedly altered by the diets. The 18:2n-6 and 20:4n-6 content of cardiac phospholipids was markedly (-49%) depressed by fish oil as compared with sunflower oil feeding, but the nonesterified fatty acid level of heart membrane was lowest in coconut oil-fed rats. In addition, fish oil feeding more drastically depressed the n-6/n-3 fatty acid ratio in the nonesterified fatty acid pool than in cardiac phospholipids. Cyclic AMP phosphodiesterase activity was the lowest in both the particulate and soluble fractions of heart from rats fed sunflower oil, whereas cyclic GMP phosphodiesterase activity was not altered by the diets. Cyclic AMP phosphodiesterase activity was decreased by 18 and 12% in heart membranes of the sunflower oil group as compared to that of the coconut oil and fish oil groups, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Incorporation and turnover of eicosapentaenoic and docosahexaenoic acids in human blood platelets in vitro

Mass changes in the incorporation of linoleic (C18:2), eicosapentaenoic (C20:5) and docosahexaenoic (C22:6) acids in human blood platelet phospholipids were induced by incubating the cells and these fatty acids complexed to albumin. The remodelling of [14C]C18:2, [14C]C20:5 and [14C]C22:6 in classes, subclasses and molecular species of platelet phospholipids was studied in resting and thrombin-stimulated cells. More than 85% of the incorporation was located in phospholipids, representing 5-fold and 2.5-fold increases in the phospholipid C20:5 and C22:6 endogenous content respectively. Thrombin stimulation induced a 30% degradation of 1-acyl-2-C20:5-glycerophosphocholine (GPC) and 1-acyl-2-C22:6-GPC, but did not induce significant release of C18:2 from 1-acyl-2-C18:2-GPC. There was no change in the [14C]fatty acid composition of 1-alkyl-2-acyl-GPC. Thrombin-dependent increases in 1-alkenyl-2-C20:5-glycerophosphoethanolamine (GPE) and 1-alkenyl-2-C22:6-GPE of 2.1-fold and 2.5-fold respectively accounted for the rise in GPE radioactivity and partly compensated for the loss of these fatty acids from 1,2-diacyl-GPC: transfer to 1-alkenyl-2-acyl-GPE was 0.4 and 1.5 nmol/10(9) platelets for C20:5 and C22:6 respectively. [14C]C20:5 and [14C]C22:6 were incorporated into six different species of 1,2-diacyl-GPC, with acylation in the major endogenous forms (C18:1 +C16:0 and C18:0 species) representing 76% and 66% respectively of the total radioactivity present in 1,2-diacyl-GPC. Stimulation by thrombin induced significant release of these fatty acids from the main molecular species of 1,2-diacyl-GPC, but significantly stimulated the synthesis of alkenyl forms of GPE containing C18:1/C22:6 +C16:0/C22:6, C18:0/C22:6 and C18:0/C20:5. C18:0/C18:2, the major endogenous C18:2 molecular species, represented only 10.5% of the incorporation; none of the [14C]C18:2 molecular species was a substrate for transfer towards 1-alkenyl-2-acyl-GPE. It is concluded that when C20:5 and C22:6, but not C18:2, are acylated in 1,2-diacyl-GPC, they participate in thrombin-dependent phospholipid remodelling, and might compete with the turnover and release of arachidonic acid from platelet phospholipids and the subsequent activation of the cells.

Dietary polyunsaturated fatty acids modulate fatty acid composition and early activation steps of concanavalin A-stimulated rat thymocytes

The early biochemical responses to concanavalin A (Con A) of thymocytes from rats fed a saturated (coconut oil), (n-6) (sunflower oil) or (n-3) (fish oil) fatty acid-enriched diet for 3 wk were investigated. Fish oil feeding resulted in greater (n-3) polyunsaturated fatty acid level (PUFA) at the expense of (n-6) PUFA in total and individual thymocyte phospholipids. Such alterations of the fatty acid composition did not affect basal ornithine decarboxylase (ODC), cyclic nucleotide phosphodiesterase (PDE) or gamma-glutamyl transferase activities. However, the fish oil-enriched diet impaired some of the early thymocyte responses to Con A, such as the rapid induction (30 min) of soluble ODC and PDE activities. Synthesis of [3H]20:4(n-6) oxygenated metabolites was not different between the dietary groups; however, the uptake of [3H]20:4(n-6) into phospholipid classes was significantly lower in phosphatidylcholine and greater in phosphatidylethanolamine and phosphatidylinositol after fish oil feeding. Similarly, the Con A-induced remodeling of the [3H]20:4(n-6) esterification in phospholipids differed in sunflower oil- vs. fish oil-fed rats, suggesting a modulation of acyl CoA synthase and/or acyl CoA transferase activities. Thus, the modulation of Con A-induced ODC and PDE stimulation upon in vivo changes of membrane phospholipid fatty acid composition is not related to eicosanoid formation, but rather to the modification of the fatty acid acylation processes, altering phospholipid composition and signal transduction.