J. Lipid Res.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1194/jlr.M500151-JLR200 on June 16, 2005

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
M500151-JLR200v1
46/9/1983    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kleinfeld, A. M.
Right arrow Articles by Okada, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kleinfeld, A. M.
Right arrow Articles by Okada, C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Journal of Lipid Research, Vol. 46, 1983-1990, September 2005
Copyright © 2005 by American Society for Biochemistry and Molecular Biology

Free fatty acid release from human breast cancer tissue inhibits cytotoxic T-lymphocyte-mediated killing

Alan M. Kleinfeld1 and Clifford Okada

Torrey Pines Institute for Molecular Studies, San Diego, CA 92121

Published, JLR Papers in Press, June 16, 2005. DOI 10.1194/jlr.M500151-JLR200

1 To whom correspondence should be addressed. e-mail: akleinfeld{at}tpims.org


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Immune-mediated antitumor activities confront a variety of tumor-mediated defense mechanisms. Here, we describe a new mechanism involving FFA that may allow breast cancer to evade immune clearance. We determined the IC50 at which unbound free fatty acids (FFAu) inhibit murine cytotoxic T-lymphocyte (CTL)-mediated killing to assess the physiologic relevance of this phenomenon. We found that the IC50 for unbound oleate is 125 ± 30 nM, ~200-fold greater than normal plasma levels. FFA inhibition, however, may play an important role in breast cancer because we found that large quantities of FFAs are released constitutively into the media surrounding samples of human breast cancer but not normal or benign tissue. FFAu concentration ([FFAu]) increased to at least 25 nM in 20 of 22 cancer tissue samples and exceeded 100 nM in 11 patients. Media from these samples inhibited CTL-mediated killing. Extrapolation from our in vitro conditions suggests that for tumor interstitial fluid, in vivo [FFAu] may be 300-fold greater than we observed in vitro.

Although breast cancer release of FFA may suppress effector cell antitumor activity, strategies that reduce interstitial [FFAu] may significantly improve antitumor immune therapies.

Abbreviations: ADIFAB, acrylodan-labeled intestinal fatty acid binding protein; CHTN, Cooperative Human Tissue Network; CTL, cytotoxic T-lymphocyte; FAFBSA, fatty acid-free bovine serum albumin; FCS, fetal calf serum; FFAu, unbound free fatty acids; [FFAu], unbound free fatty acid concentration; OAu, unbound oleate concentration

Supplementary key words unbound free fatty acid • free fatty acid release • immunotherapy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Antitumor cytotoxic T-lymphocytes (CTLs) are central to immunotherapeutic anticancer strategies (1–5). Tumor-specific CTLs can be induced by various immunotherapeutic strategies (reviewed in 3), and tumor-infiltrating lymphocytes are found "naturally" in tumors (6). Nevertheless, immunotherapeutic methods are rarely effective in mediating the regression or clearance of established tumors. This lack of effectiveness suggests that factors other than CTL activation may prevent CTL-mediated clearance of tumor in vivo. A number of factors have been identified that reduce CTL effectiveness, including loss of tumor antigen, major histocompatibility complex class I downregulation, and a physical barrier separating CTL and tumor (3). In the present study, we provide evidence for a new mechanism: inhibition of CTL killing of tumor cells by FFAs produced by human breast cancer tumors.

Inhibition of CTL activity by increased levels of total FFAs has been demonstrated in several in vitro studies (7–16). Increased levels of cis unsaturated FFAs inhibit CTL signaling pathways and, in particular, inhibit CTL-mediated killing of cognate target cells. These inhibitory effects of cis FFAs are attributable to an immediate physical perturbation of the CTL; the effects occur within seconds and are reversed upon extracting FFAs with extracellular albumin. In contrast, saturated FFAs have no effect on signaling or killing. The differential effects of cis and saturated FFAs does not involve FFA metabolism because CTL inhibition can be detected before metabolite levels are significant and because the predominant metabolites of oleic acid (the most potent FFA inhibitor) cannot be extracted with extracellular albumin (9). The differential effects of cis and saturated FFAs, however, are well correlated with differential effects of cis and saturated FFAs on membrane lipid order (10, 14), consistent with a physical mechanism of FFA-mediated CTL inhibition.

In biological fluids, most FFA is bound to albumin and/or cells (17, 18). Although only a small fraction of total FFA is unbound free fatty acids (FFAu) (19), it is the FFAu that determine the degree of FFA inhibition of cellular function (14). FFAu levels that inhibit CTL-mediated killing have not been determined. Estimates have been obtained for FFAu levels (IC50) that inhibit signaling in murine CTL, in this case the concanavalin A-stimulated increase in intracellular calcium (14). The IC50 for signaling inhibition is FFA type-dependent and is ~200 nM for oleate, probably the most important FFA for suppression of the CTL response in vivo because it is the most abundant cis unsaturated FFA. Normal human plasma total FFAu levels are ~1.4 nM (20), of which ~0.5 nM is oleate; therefore, normal unbound oleate levels are 400-fold smaller than the in vitro IC50 for oleate inhibition of signaling.

Specific physiologic conditions, however, may increase plasma FFAu levels sufficiently to inhibit CTL activity. Lipid/heparin infusions in healthy volunteers generate plasma FFA increases that inhibit T-cell signaling (16), and plasma FFAu levels can exceed the IC50 for CTL signaling in acute cardiac ischemia (20, 21). Increased plasma FFAs also occur in the cachexia that is often associated with cancer (22), and lymphocyte function is inhibited in cancer patients with increased plasma FFA levels (23). Because FFAu increases exponentially with increasing total FFA to albumin (17, 18), total FFA increases of 4- to 6-fold above normal, as observed in the studies described above, would be expected to generate inhibitory levels of FFAu.

The mechanisms described above for increasing plasma FFA reflect the mobilization of FFA from adipose storage. FFAs may also be generated by activation of lipolysis in the tumor cells themselves. In vitro studies reveal that high levels of FFA are released from tumor cells within minutes of CTL attack (24, 25). In the present study, we monitored the constitutive FFA release from human breast tissue samples and determined the effect of this release on murine CTL killing of cognate target cells. Our results, which are the first to demonstrate that FFAs are released from breast cancer tissue, also show that no release occurs from benign tumor and normal breast tissue. In addition, we determined the dose response for FFAu inhibition (IC50) of CTL-mediated killing of cognate tumor cells and found that FFAu levels produced by breast cancer tissue can exceed the IC50. Our results suggest that constitutive FFA release in vivo may allow breast cancer tumors to evade clearance by CTL.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
CTL clones and tumor cell lines
Maintenance and stimulation of murine CTL clones and tumor cell lines was performed as described previously (26). The CTL clone C30 (anti-H2-Kb) was a gift from Dr. Linda Sherman (Scripps Institute of Research, La Jolla, CA), and the CTL clone OE4 (anti-H2d) was a gift from Osami Kanagawa (Lilly Research Laboratories, La Jolla, CA). The CTLs were maintained in vitro by weekly stimulation with irradiated C57BL/6J and BALB/CJ spleen cells, respectively, and propagated in RPMI 1640 plus 10 mM HEPES and 4 mM L-glutamine (all from Cambrex Bio Sciences), 1% MEM nonessential amino acid solution and 1 mM sodium pyruvate (both from Sigma), 10 U/ml recombinant interleukin 2 (National Institutes of Health), 50 mM 2-mercaptoethanol (Serva Feinbiochemica), and 10% heat-inactivated fetal calf serum (FCS; Tissue Culture Biologicals, Tulare, CA). The respective allogeneic cognate tumor target cells, EL4 (H-2Kb) and P815 (H-2Kd), were maintained in vitro in the same media as for CTL with 2.5% FCS and 7.5% calf serum instead of 10% FCS and were periodically restarted from frozen stock.

Assay for lytic activity
Lytic activity was measured by the release of 51Cr from target cells loaded with Na251CrO4 (Perkin-Elmer Life Sciences Products) as described previously (11). CTL-target cell conjugates, at a total volume of ~200 µl on 96-well plates, were formed by centrifugation using a 5:1 ratio of CTL to target cells (5 x 103 cells) in a buffer composed of 20 mM HEPES, 140 mM NaCl, 5.5 mM glucose, 5 mM KCl, 1 mM NaH2PO4, 1 mM CaCl2, and 1 mM MgSO4 at pH 7.4 (C-HEPES) for 5–10 min at 1,000 rpm (160 g). Harvest of the supernatant was done after 4 h of incubation at 37°C, which was then assayed for 51Cr release. Parallel samples containing labeled target cells with buffer in place of CTL and appropriate amounts of testing sample were assayed in the same way to yield a value for spontaneous release. Total counts were determined from 51Cr released upon treatment of the labeled target cells with 2% Triton X-100. Specific lysis was calculated as 100 x (sample spontaneous release)/(total spontaneous release). Measurements for each assay condition were performed using between three and five replicates. The reported values are averages and standard deviations of these replicates.

Breast tissue preparation
Breast tissue samples obtained immediately after surgery from the Cooperative Human Tissue Network (CHTN) were shipped on ice overnight in RPMI 1640 containing penicillin and streptomycin. The protocol for the use of these samples was approved by the Torrey Pines Institute for Molecular Studies Institutional Review Board, and patient informed consent was obtained by the participating CHTN hospitals. Histological analysis was performed by pathologists at the CHTN participating hospitals. Breast tissue samples included 22 cancer samples and, for 14 of these, a matched normal and/or benign tissue sample obtained from the margins of the excised tumor where no carcinoma was detected. Pathology reports indicated that >70% of the cancer samples were ductal carcinomas; the remaining samples were lobular or untyped mammary carcinomas, and metastases were found in ~60% of patients.

Tissue sample weights ranged from 0.3 to ~3 g. Samples were cut into pieces of ~0.25 g after visible adipose, connective, and necrotic tissue had been removed. For the 14 matched samples, the weight (after removal of adipose, connective, and necrotic tissue) of the cancer tissue was less than or equal to that of the matched normal/benign tissue. Each piece of tissue was incubated at 37°C in a CO2/H2O-controlled atmosphere in 15 ml of media containing MEGM (Clonetics Mammary Epithelial Cell Medium with growth factors, cytokines, and supplements; Cambrex, Walkersville, MD) with 10% FCS (Tissue Culture Biologicals) plus penicillin, streptomycin, and gentamycin. Approximately every 1–2 days, for up to 7 weeks, 180–200 µl of the incubation medium was collected, centrifuged to remove cellular debris, and stored at –80°C. The medium remaining at the end of the incubation period was also collected and stored similarly.

FFAu treatment
Cells were treated with defined concentrations of FFAu using FFA-BSA complexes to buffer FFAu as described previously (27). FFA-BSA complexes were formed by titrating solutions of 600 µM BSA (fatty acid-free from Sigma) in C-HEPES with sodium salts of the FFA (Nu-Chek Prep, Elysian, MN). The FFAu concentration ([FFAu]) for each complex was determined using the fluorescent probes acrylodan-labeled intestinal fatty acid binding protein (ADIFAB) or the L72A mutant (ADIFAB2) (FFA Sciences LLC) as described previously (28, 29).

FFA treatment of CTL was performed by suspending the CTL in C-HEPES, generally at 2.5 x105 cells/ml, to which the appropriate (defined [FFAu] value) FFA-BSA complex was added to make the final BSA concentration, in the CTL-target mixture, between 60 and 300 µM. The CTLs were incubated in this medium for ~10 min at 37°C before the formation of CTL-target cell complexes. After the 4 h killing assay, FFAu levels were measured in the CTL-target and the target (spontaneous release) media. The ADIFAB fluorescence measurements used ~150 µl of media to 1.35 ml of C-HEPES, yielding a final [BSA] of >=6 µM. Under these conditions the FFAu provides an accurate estimate of the undiluted media (20).

With sufficiently high [BSA], the FFAu levels generated by the FFA-BSA complexes should be well buffered (invariant) with or without cells present (30). In the present studies, we found that the degree of [FFAu] buffering on the plastic 96-well plates used for the cytolysis assays was FFA type-dependent. For oleate and linoleate, the difference in [FFAu] values with or without (FFA-BSA complexes only) cells present was not significant. Moreover, for oleate and linoleate, [FFAu] values for FA-BSA complexes was virtually identical on 96-well plates and in cuvettes. However, for palmitate, the combined effects of binding to the well surfaces and cells was significant and limited the maximum achievable FFAu level to ~200 nM.

CTL-target cell complexes were also treated with media from normal, benign, and cancer breast tissue obtained at the end of the incubation period. This medium, which contains 10% FBS, is ~60 µM albumin. To maximize the FFA-BSA buffering capacity, the volume of breast tissue medium added to the CTL-target cell complexes in C-HEPES was between 60% and 90% of the total volume, yielding ~40–50 µM albumin. The actual [FFAu] values of the breast tissue/CTL-target cell complexes were determined at the completion of the killing assay by direct measurement of media samples, as described previously for plasma (20). These [FFAu] values are the sum of concentrations of the five different FFAs shown in Fig. 4 below that are released into the media. We estimate, using albumin binding constants, that the fraction of cis [FFAu] is ~75% of the total [FFAu], greater than the 60% of total FFA (see Fig. 4) because of the higher affinities of the saturated FFAs.



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 4. Distribution of FFAs released from cancer and normal tissues. Gas chromatography was used to determine FFA distribution in cancer and matched normal breast tissue media. The FFA fraction was obtained from tissue media taken from the last day of the incubation period (generally when [FFAu] was greatest). Results are averages and standard deviations of three separate determinations for each sample, and the normal sample results are 10 times the measured values. [FFAu] values were 4 and 410 nM for the normal and cancer samples, respectively. FFAs are designated by number of carbons and double bond number. The corresponding trivial names are as follows: palmitate (16:0), palmitoleate (16:1), stearate (18:0), oleate (18:1), and linoleate (18:2).

 
Fatty acid analysis of breast tissue media
Sample aliquots of 1.0–9.0 ml of breast tissue media were extracted for lipid analysis using chloroform-methanol as described previously (25). The extract was dried under nitrogen, resuspended in chloroform, and fractionated by silica gel TLC using petroleum ether-ethyl ether-acetic acid (80:20:1) as the developing solvent. TLC plates were stained with iodine, and the FFA fraction was eluted, dried under nitrogen, and resuspended in 50–100 µl of ethanol.

The extracted FFAs were analyzed using an HP-5890 gas chromatograph equipped with a 10 m HP-FFAP capillary column (Agilent Technologies, Wilmington, DE). Detector and injector temperatures were 250°C, and helium was the carrier gas. FFAs were eluted from the column using a splitless injection technique and a three-step temperature ramp from 180°C to 240°C. Chromatograms were referenced to standards (Nu-Chek Prep) and analyzed using Agilent ChemStation software.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
CTL-mediated killing is inhibited at FFAu levels that greatly exceed normal plasma concentrations
Previous studies have demonstrated that CTL-mediated killing of target cells was inhibited at high levels of total FFAs (7, 8, 11, 16). To assess the physiologic relevance of FFA-mediated inhibition, it is necessary to determine the [FFAu] dose dependence, which was not possible in these earlier studies. We have reexamined the inhibition of CTL-mediated killing by FFA and used the ADIFAB method to determine the FFAu dose dependence for inhibition.

We determined the effect of unbound oleate, linoleate, and palmitate on the degree of target cell lysis mediated by two different murine cognate CTL-target systems: C30 (anti kb)-EL4 (kb) and OE4 (anti kd)-P815 (kd). CTLs were incubated for 5–10 min before target cell conjugation with FFA-BSA complexes designed to maintain the extracellular media at defined [FFAu]. Measurements of CTL-mediated lysis were performed by assessing 51Cr release from target cells with increasing [FFAu] in the cell media (see Methods).

Our measurements of 51Cr release reveal that CTL-mediated lysis decreases with increasing unbound oleate concentration ([OAu]) (Fig. 1). The average IC50 for inhibition was ~120 nM for both CTL-target cognate pairs (Table 1). Results for linoleate revealed a similar behavior but with a larger IC50 of ~250 nM (Table 1). Palmitate, in contrast to the cis unsaturated oleate and linoleate, caused no inhibition up to the maximum unbound palmitate level achievable (Fig. 1). Inhibition by cis unsaturated FFAu but not palmitate is consistent with previous results for total FFAs (11, 16). However, the concentrations of cis FFAu that generate significant inhibition are much greater than the molecular species average [FFAu] of 1.4 nM in normal human plasma (95th percentile ~ 2.5 nM) (20). Thus, inhibition should not occur in plasma under normal physiologic conditions.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1. Dose dependence for unbound free fatty acid (FFAu) inhibition of cytotoxic T-lymphocyte (CTL)-mediated killing. Specific lysis of 51Cr-labeled murine tumor targets EL4 and P815 by their cognate CTLs, C30 and OE4, respectively, was measured as a function of the unbound concentrations of oleate and palmitate. FFAu concentration ([FFAu]) levels in the CTL-target cell suspensions were maintained by FFA-BSA complexes for which [BSA] was >60 µM. A: C30 CTL killing of EL4 cells. The specific lysis scales for oleate and palmitate are the same. B: OE4 killing of P815 cells. The specific lysis scale for oleate is indicated on the left and that for palmitate on the right vertical axis. The average specific lysis (percent) and standard deviations from three to five replicates for each [FFAu] (in nM) are shown for oleate; for ease in presentation, only the average values are shown for palmitate. The standard deviations for palmitate are similar to those for oleate and average 14%.

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. IC50 values for oleate and linoleate inhibition of CTL-mediated killing

 
For each [FFAu], the level of 51Cr release was also measured from a matched quantity of target cells without CTL. This "spontaneous" release was subtracted from the total 51Cr release to obtain the level of specific lysis. We observed an increase in spontaneous lysis at [OAu] or linoleate concentration significantly greater than the IC50 values (data not shown). In contrast to target cells, CTL revealed little or no lysis (as monitored by trypan blue) at these higher [FFAu] values.

Inhibition can be reversed by removing FFAu
Our previous studies indicated that FFA-mediated inhibition of CTL signaling could be reversed by reducing FFAu levels (9). Therefore, we investigated whether inhibition of CTL-mediated lysis can also be reversed by reducing [FFAu]. To do this, we carried out [FFAu] titration experiments similar to those used in Fig. 1, but for each [FFAu] we added fatty acid-free bovine serum albumin (FAFBSA) to the CTL-target sample at increasing times after CTL-target conjugate formation. The results indicate that inhibition is reversed, and lysis is restored, if FAFBSA is added within 1.5 h of conjugation (Fig. 2 shows the results for FAFBSA addition at 45 min). The apparent impairment of reversibility at longer times may simply reflect the resetting of the "clock" for lysis after FAFBSA addition, thereby effectively reducing the assay period. Adding FAFBSA within 2 h from the start of the incubation also blocked the increase in spontaneous target cell lysis at higher [FFAu] (data not shown). This indicates that CTLs remain functional if FFAu is removed and that even relatively high FFAu levels do not significantly lyse targets or CTLs.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. Reversibility of FFAu inhibition. Specific lysis of EL4 cells by C30 CTLs as a function of unbound oleate concentration ([OAu]) either with (open squares) or without (closed squares) the addition of 120 µM fatty acid-free bovine serum albumin (FAFBSA) at 45 min after C30-EL4 conjugation. Values were normalized to specific lysis at zero [OAu], which was 30% and 45% for cells to which FAFBSA was or was not added, respectively. [OAu] was determined in cells without added FAFBSA. Values shown are averages ± SD.

 
Large quantities of FFAs are constitutively released from breast cancer but not normal breast tissue
Because we observed previously that CTL attack activates FFA release from cognate tumor cells (25), we speculated that rapidly metabolizing tumor cells might release significant levels of FFAs. In this event, FFAu levels might be increased sufficiently in the interstitial fluid surrounding the tumor to interfere with CTL clearance of the tumor. We investigated whether FFA was released from human breast tissue by monitoring the time course of FFAu levels in the media surrounding the tissue (see Methods). This approach for investigating breast tumor interstitial fluid is similar to that described recently for the proteomic characterization of this fluid (31).

We found that [FFAu] increased significantly ({Delta}FFAu > 10 nM) for 19 of the 22 cancer samples but in only 2 of the normal samples and in neither of the 2 benign tumor samples (Fig. 3 and data not shown). The difference in tumor/normal breast tissue FFA release is illustrated in Fig. 3 for six of the matched tumor/normal tissue pairs. Although the magnitude and rate of FFA release varied considerably for different patients, most cancer samples revealed significant ({Delta}FFAu > 10 nM) FFAu increases within the first 2 days and most increased to 50 nM or greater within ~10 days. As explained in Discussion, these in vitro levels likely represent a severe underestimate of actual tumor interstitial fluid levels. Normal FFAu levels did not change with time and were generally <5 nM. These results suggest that FFAu released by the cancer tissue is an intrinsic property of the malignant cells, because little or no FFA was released from normal tissue, although the mass of normal tissue was greater than or equal to that of the cancer tissue (see Methods).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3. Time courses for FFA release from cancer and matched normal breast tissues. [FFAu] was measured in the tissue media for matched cancer (closed squares) and normal (open squares) breast tissue samples from six different patients (A–F). Results shown are averages of two or more measurements, and the standard deviations were ~5%.

 
Evidence supporting the reliability of the ADIFAB determination of [FFAu] was obtained by repeating the [FFAu] measurements after adding 10 µM FAFBSA to each sample. In every case, [FFAu] decreased significantly, especially for samples with high pre-FAFBSA FFAu levels (see Fig. 5 below). This is consistent with the expected nonlinear dependence of [FFAu] on the total FFA-BSA ratio at high [FFAu] (17, 18).



View larger version (46K):
[in this window]
[in a new window]
  		Fig. 5. Effect of breast tissue media on CTL killing. C30-CTL killing of EL4 cells was measured in the following media with corresponding [FFAu] (total of all FFAu) and percent lysis: first bar, C-HEPES (60 µM FAFBSA, [FFAu] = 0, 69 ± 6%); second bar, C-HEPES plus oleate/BSA (60 µM BSA, [OAu] = 83 nM, 17 ± 6%); third bar, normal breast tissue media ([BSA] {approx} 45 µM, [FFAu] = 2 nM, 50 ± 4%); fourth bar, benign breast tissue media ([BSA] {approx} 45 µM, [FFAu] = 3 nM, 51 ± 4%); fifth bar, cancer breast tissue media ([BSA] {approx} 45 µM, [FFAu] = 206 nM, 7 ± 9%); and sixth bar, cancer breast tissue media plus 60 µM FAFBSA ([BSA] {approx} 105 µM, [FFAu] = 8 nM, 44 ± 9%).

 
We used gas chromatography to determine the distribution of molecular species of FFAs released from several cancer and normal tissues after lipid extraction and thin-layer chromatography of the respective media. The results indicate that the distribution of the major FFAs was approximately similar in cancer and normal tissue (Fig. 4). The predominant FFA released from breast cancer tissues is oleic acid, constituting ~30–40% of the total. As expected from the FFAu results, the mass of FFA in cancer media was considerably (>10-fold) greater than in normal media.

Media from breast cancer but not normal tissue inhibits CTL-mediated lysis
Because increased levels of cis FFAu inhibit CTL-mediated killing and cis FFAu levels are increased in breast cancer tissue media, we compared CTL-mediated killing of EL4 and P815 cells in media from normal, benign, and cancer breast tissue. The media used were obtained from the last day of incubation, generally when [FFAu] in the cancer medium had achieved its maximum value. The results of these measurements indicate that media from breast cancer tissue, with high [FFAu], but not from normal or benign breast tissue inhibit CTL-mediated lysis, as indicated in Fig. 5 for C30 killing of EL4 cells. Inhibition was greater in the cancer media compared with OA-BSA because of the large fraction of cis FFAu in the media (see Methods). Figure 5 also shows that inhibition can be reversed by reducing FFAu levels with FAFBSA. These results support FFAs as the mediators of inhibition by breast cancer media.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, we provide evidence that FFA release from human breast cancer tumors represents a previously unrecognized mechanism by which tumors may evade immune-mediated clearance. This conclusion derives from our observation that large quantities of FFAs were released constitutively from human breast cancer tissue but not from benign tumor or normal tissue. FFAu levels in cancer tissue media often exceeded 120 nM, the IC50 for oleate inhibition of murine CTL-mediated killing of cognate tumor cells, and oleate was the most abundant FFAu released from the breast cancer tissue. Media from cancer but not normal or benign tissues inhibited CTL-mediated killing of cognate tumor cells, and this inhibition could be reversed by reducing FFAu levels. These observations suggest that [FFAu] is increased in the interstitial fluid of breast cancer tumors and that reducing [FFAu] might improve the efficacy of immune-mediated clearance of the tumors.

The effects of FFAs released from tumor cells have not been reported previously. Earlier studies, however, have described relationships between exogenously added FFAs and tumor proliferation. For example, exogenously added arachidonate was found to inhibit the growth of breast cancer and colorectal cancer cell lines in vitro, whereas linoleate uptake stimulated the growth of a rat hepatoma in vivo (32–35). A complication in interpreting results derived from exogenously added FFAs, as emphasized by Hardy, Langelier, and Prentki (35) and the present study, is that without careful control of [FFAu], the physiologic relevance of such studies cannot be gauged accurately. In the study of Hardy, Langelier, and Prentki (35), for which FFA/albumin ratios were maintained at physiologic levels, oleate stimulated proliferation but palmitate inhibited growth of the MDA-MB-231 breast cancer cell line in vitro. In addition, Hardy, Langelier, and Prentki (35) and others (36) have found that oleate protects cells against palmitate-mediated apoptosis. This may be relevant to the results of the present study, because both in CTL-mediated tumor cell release of FFA (25) and in the constitutive release from breast cancer tissue (Fig. 4), significantly more oleate than palmitate is released.

Additional evidence that FFAs are essential to tumor growth derives from studies demonstrating that FAS is highly expressed in many tumor cells, including breast cancer, and that inhibition of FAS inhibits tumor growth (37). These studies are consistent with the notion that high levels of FFA synthesis are necessary for the viability of rapidly metabolizing cancer cells. This high level of metabolism might also be needed to replenish the large quantities of FFAs that we find secreted from tumor cells. However, we did not find, in studies of selected tissue samples, that the FAS inhibitor C57 reduced the level of FFAu secreted from breast cancer tissue (data not shown).

Except for adipocytes, fatty acid release from nontransformed cells appears to occur only upon loading cells with triacylglycerol, and even then, net release requires virtually zero extracellular FFAu (38). Ehrlich ascites tumor cells release FFAs without loading, but net release also occurs only when extracellular FFAu is virtually zero (39). In contrast, net release from breast cancer, but not benign or normal tissue, increases virtually independently of extracellular FFAu levels (Fig. 3).

The difference in FFA release from cancer and normal/benign tissue is likely a reflection of tumorigenesis rather than, for example, a different distribution of normal cells in the different tissues. Breast tissue is predominantly a mixture of adipose, fibroblastic, epithelial, and myoepithelial cells. Our results were based on samples from 22 different patients (22 cancer and 14 matched normal/benign), each of which is a mixture of the different cell types. All cancers were of epithelial origin (carcinomas), which are also constituents of the normal/benign samples. Thus, FFA release from cancer tissue is likely a reflection of epithelial cell transformation.

Most of the FFAs released from breast cancer tissue probably originates from cellular membranes. Microscopic examination of dispersions of cells from selected breast tumor samples did not reveal significant amounts of intracellular lipid droplets (data not shown). Previous studies indicate that the phospholipid content is 4-fold higher and the triacylglycerol content is 65% lower in cancer compared with noncancerous parts of the breast tissue (40). Moreover, the fatty acid composition of the cancer tissue phospholipids was highly enriched in unsaturated FFAs compared with noncancerous tissue, whereas triacylglycerol revealed no significant difference between cancer and normal tissue (40). Consistent with these observations, cis unsaturated FFAs are the predominant FFAs released from breast cancer tissue (Fig. 4). Our earlier study of CTL-mediated FFA release from murine tumor cell lines also revealed a phospholipid origin of the released FFAs and enrichment of cis unsaturated FFAs (25). Thus, our earlier study and the present study are consistent with a FFA release mechanism involving phospholipase activity.

Although phospholipid may be the source of the FFAs, the mechanism that activates lipolysis and the mechanism that releases FFAs from the tumor tissue are unknown. Necrotic tissue is present especially in high-grade breast cancer lesions and is thought to be induced by hypoxia within the tissue (41). Loss of membrane integrity in necrotic cells would increase intracellular calcium levels and thereby activate calcium-sensitive phospholipases. FFAs might then be released from the cells simply by diffusing through the membrane defects created during necrosis. Alternatively, FFAs may be released from the cell by transport through an intact membrane. To achieve high extracellular [FFAu] would require high (>50 nM) intracellular [FFAu] and/or an ATP-driven FFA pump having an orientation opposite the one we have described in adipocytes (30, 42). Determination of the specific mechanism is important for developing methods to reduce release, in the event that this phenomenon can be confirmed in vivo. Future studies will attempt to determine the mechanism of breast tumor release of FFAs.

The results of this study suggest that FFAu levels in the interstitial fluid surrounding a breast cancer tumor may be increased sufficiently to inhibit the cytolytic activity of infiltrating CTLs. To estimate in vivo interstitial FFAu levels from our results, we compared the FFA-buffering capacity of the in vitro extracellular media with that expected in interstitial fluid in vivo. In our measurements, the volume of extracellular media was ~60-fold greater than the volume of tissue mass (15 ml to 0.25 g; see Methods). Measurements of the ratio of interstitial volume to adipose and skeletal muscle tissue volume yield ~0.1 (43), ~600-fold less than in our in vitro measurements. On the other hand, the interstitial albumin concentration (44), ~130 µM, is roughly twice the 60 µM (10% FCS) seen in our extracellular media. Combining these two factors suggests that the buffering capacity of interstitial fluid is ~300-fold less than in our extracellular media. Therefore, if in vivo tumor production and efflux of FFAs are similar to those of isolated tissue, the magnitude and rate of increase of FFAu levels may be significantly (up to 300-fold) greater in interstitial fluid than we observe in vitro.

It follows that at relatively early stages in tumor development, interstitial FFAu levels may be increased sufficiently to inhibit CTL activity. Even if activated CTLs are generated and are able to infiltrate a tumor, increased interstitial FFAu levels may prevent CTL-mediated clearance of the tumor. Because CTLs also stimulate very rapid and large amounts of FFA release from cognate tumor targets (24, 25), any level of CTL-mediated killing may actually further increase interstitial FFAu levels. Plasma FFAu may also be increased in cachexia, diabetes, and cardiac ischemia (FFA release from the tumor, because of its small mass, probably does not contribute significantly to plasma [FFAu]). Any increase in plasma FFAu would further diminish the buffering capacity of the interstitial fluid and thereby enhance immune suppression. If in situ tumor interstitial fluid [FFAu] is increased and if therapeutic strategies can be developed to limit the tumor release of FFAs, the success of immune-mediated tumor clearance may be improved.


   ACKNOWLEDGMENTS
 
This study was supported by the United States Army Medical Research and Material Command through Grant DAMD17-001-0470 and by funds from the Alzheimer's and Aging Research Center.

Manuscript received April 19, 2005 and in revised form June 9, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
  1. Pardoll, D. 2001. T cells and tumours. Nature. 411: 1010–1012.

  2. Ochsenbein, A. F., S. Sierro, B. Odermatt, M. Pericin, U. Karrer, J. Hermans, S. Hemmi, H. Hengartner, and R. M. Zinkernagel. 2001. Roles of tumour localization, second signals and cross priming in cytotoxic T-cell induction. Nature. 411: 1058–1064.[CrossRef][Medline]

  3. Ochsenbein, A. F. 2002. Principles of tumor immunosurveillance and implications for immunotherapy. Cancer Gene Ther. 9: 1043–1055.[CrossRef][Medline]

  4. Brossart, P., S. Wirths, G. Stuhler, V. L. Reichardt, L. Kanz, and W. Brugger. 2000. Induction of cytotoxic T-lymphocyte responses in vivo after vaccinations with peptide-pulsed dendritic cells. Blood. 96: 3102–3108.[Abstract/Free Full Text]

  5. Hernando, J. J., T. W. Park, and W. C. Kuhn. 2003. Dendritic cell-based vaccines in breast and gynaecologic cancer. Anticancer Res. 23: 4293–4303.[Medline]

  6. Rosenberg, S. A., P. Spiess, and R. Lafreniere. 1986. A new approach to the adoptive immunotherapy of cancer with tumor-infiltrating lymphocytes. Science. 233: 1318–1321.[Abstract/Free Full Text]

  7. Samlaska, C. P. 1979. Linoleic acid inhibition of naturally occurring lymphocytotoxicity to breast cancer-derived cells measured by a chromium-51 release assay. J. Natl. Cancer Inst. 62: 1427–1433.

  8. Taylor, A. S., R. C. Howe, A. R. Morrison, H. Sprecher, and J. H. Russel. 1985. Inhibition of cytotoxic T lymphocyte-mediated lysis by ETYA: effect independent of arachidonic acid metabolism. J. Immunol. 134: 1130–1135.[Abstract]

  9. Richieri, G. V., and A. M. Kleinfeld. 1989. Free fatty acid perturbation of transmembrane signaling in cytotoxic T lymphocytes. J. Immunol. 143: 2302–2310.[Abstract]

  10. Richieri, G. V., M. F. Mescher, and A. M. Kleinfeld. 1990. Short term exposure to cis unsaturated free fatty acids inhibits degranulation of cytotoxic T lymphocytes. J. Immunol. 144: 671–677.[Abstract]

  11. Richieri, G. V., and A. M. Kleinfeld. 1990. Free fatty acids inhibit cytotoxic T lymphocyte mediated lysis of allogeneic target cells. J. Immunol. 145: 1074–1077.[Abstract]

  12. Chow, S. C., I. J. Ansotegui, and M. Jondal. 1990. Inhibition of receptor-mediated calcium influx in T cells by unsaturated non-esterified fatty acids. Biochem. J. 267: 727–732.[Medline]

  13. Breittmayer, J. P., C. Pelassy, J. L. Cousin, A. Bernard, and C. Aussel. 1993. The inhibition by fatty acids of receptor-mediated calcium movements in Jurkat T-cells is due to increased calcium extrusion. J. Biol. Chem. 268: 20812–20817.[Abstract/Free Full Text]

  14. Anel, A., G. V. Richieri, and A. M. Kleinfeld. 1993. Membrane partition of fatty acids and inhibition of T cell function. Biochemistry. 32: 530–536.[CrossRef][Medline]

  15. Kleinfeld, A. M. 1993. The role of free fatty acids in CTL-target cell interactions. In Cytotoxic Cells: Generation, Triggering, Effector Functions, Methods. M. Sitkovsky, editor. Birkhauser, Boston. 321–328.

  16. Stulnig, T. M., M. Berger, M. Roden, H. Stingl, D. Raederstorff, and W. Waldhausl. 2000. Elevated serum free fatty-acid concentrations inhibit T lymphocyte signaling. FASEB J. 14: 939–947.[Abstract/Free Full Text]

  17. Spector, A. A. 1975. Fatty acid binding to plasma albumin. J. Lipid Res. 16: 165–179.[Abstract]

  18. Richieri, G. V., A. Anel, and A. M. Kleinfeld. 1993. Interactions of long chain fatty acids and albumin: determination of free fatty acid levels using the fluorescent probe ADIFAB. Biochemistry. 32: 7574–7580.[CrossRef][Medline]

  19. Richieri, G. V., and A. M. Kleinfeld. 1995. Unbound free fatty acid levels in human serum. J. Lipid Res. 36: 229–240.[Abstract]

  20. Apple, F. S., A. M. Kleinfeld, and J. E. Adams. 2004. Unbound free fatty acid concentrations are increased in cardiac ischemia. Clin. Proteomics. 1: 41–44.[CrossRef]

  21. Kleinfeld, A. M., D. Prothro, D. Brown, R. C. Davis, G. V. Richieri, and A. DeMaria. 1996. Increases in serum unbound free fatty acid levels following coronary angioplasty. Am. J. Cardiol. 78: 1350–1354.[CrossRef][Medline]

  22. Legaspi, A., M. Jeevanandam, H. F. Starnes, Jr., and M. F. Brennan. 1987. Whole body lipid and energy metabolism in the cancer patient. Metabolism. 36: 958–963.[CrossRef][Medline]

  23. Brown, R. E., R. W. Steele, D. J. Marmer, J. L. Hudson, and M. A. Brewster. 1983. Fatty acids and the inhibition of mitogen-induced lymphocyte transformation by leukemic serum. J. Immunol. 131: 1011–1016.[Abstract]

  24. Koren, H. S., E. Ferber, and H. Fischer. 1971. Changes in phospholipid metabolism of a tumor target cell during a cell-mediated cytotoxic reaction. Biochim. Biophys. Acta. 231: 520–526.[Medline]

  25. Richieri, G. V., and A. M. Kleinfeld. 1991. Free fatty acids are produced in and secreted from target cells very early in cytotoxic T lymphocyte-mediated killing. J. Immunol. 147: 2809–2815.[Abstract/Free Full Text]

  26. Anel, A., G. V. Richieri, and A. M. Kleinfeld. 1994. A tyrosine phosphorylation requirement for CTL degranulation. J. Biol. Chem. 269: 9506–9513.[Abstract/Free Full Text]

  27. Cupp, D., J. P. Kampf, and A. M. Kleinfeld. 2004. Fatty acid:albumin complexes and the determination of long chain free fatty acid transport across membranes. Biochemistry. 43: 4473–4481.[CrossRef][Medline]

  28. Richieri, G. V., R. T. Ogata, and A. M. Kleinfeld. 1992. A fluorescently labeled intestinal fatty acid binding protein: interactions with fatty acids and its use in monitoring free fatty acids. J. Biol. Chem. 267: 23495–23501.[Abstract/Free Full Text]

  29. Richieri, G. V., R. T. Ogata, and A. M. Kleinfeld. 1999. The measurement of free fatty acid concentration with the fluorescent probe ADIFAB: a practical guide for the use of the ADIFAB probe. Mol. Cell. Biochem. 192: 87–94.[CrossRef][Medline]

  30. Kampf, J. P., and A. M. Kleinfeld. 2004. Fatty acid transport in adipocytes monitored by imaging intracellular FFA levels. J. Biol. Chem. 279: 35775–35780.[Abstract/Free Full Text]

  31. Celis, J. E., P. Gromov, T. Cabezon, J. M. Moreira, N. Ambartsumian, K. Sandelin, F. Rank, and I. Gromova. 2004. Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery. Mol. Cell. Proteomics. 3: 327–344.[Abstract/Free Full Text]

  32. Trimboli, A. J., B. M. Waite, G. Atsumi, A. N. Fonteh, A. M. Namen, C. E. Clay, T. E. Kute, K. P. High, M. C. Willingham, and F. H. Chilton. 1999. Influence of coenzyme A-independent transacylase and cyclooxygenase inhibitors on the proliferation of breast cancer cells. Cancer Res. 59: 6171–6177.[Abstract/Free Full Text]

  33. Chan, T. A., P. J. Morin, B. Vogelstein, and K. W. Kinzler. 1998. Mechanisms underlying nonsteroidal antiinflammatory drug-mediated apoptosis. Proc. Natl. Acad. Sci. USA. 95: 681–686.[Abstract/Free Full Text]

  34. Blask, D. E., L. A. Sauer, R. T. Dauchy, E. W. Holowachuk, M. S. Ruhoff, and H. S. Kopff. 1999. Melatonin inhibition of cancer growth in vivo involves suppression of tumor fatty acid metabolism via melatonin receptor-mediated signal transduction events. Cancer Res. 59: 4693–4701.[Abstract/Free Full Text]

  35. Hardy, S., Y. Langelier, and M. Prentki. 2000. Oleate activates phosphatidylinositol 3-kinase and promotes proliferation and reduces apoptosis of MDA-MB-231 breast cancer cells, whereas palmitate has opposite effects. Cancer Res. 60: 6353–6358.[Abstract/Free Full Text]

  36. Listenberger, L. L., X. Han, S. E. Lewis, S. Cases, R. V. Farese, Jr., D. S. Ory, and J. E. Schaffer. 2003. Triglyceride accumulation protects against fatty acid-induced lipotoxicity. Proc. Natl. Acad. Sci. USA. 100: 3077–3082.[Abstract/Free Full Text]

  37. Kuhajda, F. P., E. S. Pizer, J. N. Li, N. S. Mani, G. L. Frehywot, and C. A. Townsend. 2000. Synthesis and antitumor activity of an inhibitor of fatty acid and synthase. Proc. Natl. Acad. Sci. USA. 97: 3450–3454.[Abstract/Free Full Text]

  38. Figard, P. H., D. P. Hejlik, T. L. Kaduce, L. L. Stoll, and A. A. Spector. 1986. Free fatty acid release from endothelial cells. J. Lipid Res. 27: 771–780.[Abstract]

  39. Spector, A. A., and D. Steinberg. 1966. Release of free fatty acids from Ehrlich ascites tumor cells. J. Lipid Res. 7: 649–656.[Abstract]

  40. Sakai, K., H. Okuyama, J. Yura, H. Takeyama, N. Shinagawa, N. Tsuruga, K. Kato, K. Miura, K. Kawase, and T. Tsujimura. 1992. Composition and turnover of phospholipids and neutral lipids in human breast cancer and reference tissues. Carcinogenesis. 13: 579–584.[Abstract/Free Full Text]

  41. Wykoff, C. C., N. Beasley, P. H. Watson, L. Campo, S. K. Chia, R. English, J. Pastorek, W. S. Sly, P. Ratcliffe, and A. L. Harris. 2001. Expression of the hypoxia-inducible and tumor-associated carbonic anhydrases in ductal carcinoma in situ of the breast. Am. J. Pathol. 158: 1011–1019.[Abstract/Free Full Text]

  42. Kleinfeld, A. M., J. P. Kampf, and C. Lechene. 2004. Transport of 13C-oleate in adipocytes measured using multi imaging mass spectrometry. J. Am. Soc. Mass Spectrom. 15: 1572–1580.[CrossRef][Medline]

  43. Iversen, P. O., E. Berggreen, G. Nicolaysen, and K. Heyeraas. 2001. Regulation of extracellular volume and interstitial fluid pressure in rat bone marrow. Am. J. Physiol. 280: H1807–H1813.

  44. Ellmerer, M., L. Schaupp, G. A. Brunner, G. Sendlhofer, A. Wutte, P. Wach, and T. R. Pieber. 2000. Measurement of interstitial albumin in human skeletal muscle and adipose tissue by open-flow microperfusion. Am. J. Physiol. 278: E352–E356.


Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Lipid Res.Home page
S. R. Shaikh and M. Edidin
Immunosuppressive effects of polyunsaturated fatty acids on antigen presentation by human leukocyte antigen class I molecules
J. Lipid Res., January 1, 2007; 48(1): 127 - 138.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
S. Yakar, N. P. Nunez, P. Pennisi, P. Brodt, H. Sun, L. Fallavollita, H. Zhao, L. Scavo, R. Novosyadlyy, N. Kurshan, et al.
Increased Tumor Growth in Mice with Diet-Induced Obesity: Impact of Ovarian Hormones
Endocrinology, December 1, 2006; 147(12): 5826 - 5834.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
E. Shumilina, N. Klocker, G. Korniychuk, M. Rapedius, F. Lang, and T. Baukrowitz
Cytoplasmic accumulation of long-chain coenzyme A esters activates KATP and inhibits Kir2.1 channels
J. Physiol., September 1, 2006; 575(2): 433 - 442.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
M500151-JLR200v1
46/9/1983    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kleinfeld, A. M.
Right arrow Articles by Okada, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kleinfeld, A. M.
Right arrow Articles by Okada, C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Journal of Biological Chemistry 
 Molecular and Cellular Proteomics   ASBMB Today