Guangzhou, China

Functional distinction of rat liver natural killer cells from spleen natural killer cells under normal and acidic conditionsin vitro

Li-Hong Lv, Jian-Dong Yu, Guo-Lin Li, Tian-Zhu Long, Wei Zhang, Ya-Jin Chen, Jun Min and Yun-Le Wan

Guangzhou, China

BACKGROUND: The microenvironment within solid tumors has often been shown to exhibit an acidic extracellular pH. Although the morphologic and functional differences in natural killer (NK) cells of the liver and spleen have been reported previously under physiological conditions, the difference under acidic conditions is still unclear. This study was to investigate the differences in the morphological and functional characteristics between rat liver and spleen NK cells under normal and acidic conditionsin vitro.

METHODS: Liver and spleen NK cells were isolated and purified from Sprague-Dawley rats by density gradient centrifugation and the Dynabeads? FlowCompTMFlexi system, and stimulated for 4 days with or without IL-2 or treated with low pH or control for different times. Morphology was examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM), cell death and proliferation assays were performed by flow cytometry, IFN-γ production was tested by ELISA, and cytotoxic activity was evaluated by lactate dehydrogenase (LDH) release assay.

RESULTS: Liver NK cells had significantly higher levels of cytotoxic activity than spleen NK cells under normal and acidic conditions, and the maximum difference was observed at pH 5.6. Further analysis revealed that the cytotoxic activity of NK cells was correlated with morphology, cell death, proliferative activity and IFN-γ production. By TEM, liver NK cells contained a greater number of electron-dense granules per cell at pH 5.6. Moreover, a modest elevation of cell death and reduction of proliferation of liver NK cells occurred within a range of 5.6-7.2. Interestingly, an acidic extracellular pH only marginally, and not significantly, suppressed IFN-γ production by liver NK cells.

CONCLUSION: The sharp morphological and functional differences shown by the two types of NK cellsin vitroindicate that liver NK cells are unexpectedly resistant to pH shock.

(Hepatobiliary Pancreat Dis Int 2012;11:285-293)

natural killer cells; liver; spleen; tumor microenvironment; extracellular pH

Introduction

Natural killer (NK) cells are critical effectors of the innate immune system that provide an early host response to invading pathogenic organisms as well as tumor cells, and play a central role in bridging the innate and adaptive arms of the immune response.[1,2]Years ago, NK cells were shown to be bone marrow-derived lymphocytes that are histologically and functionally defined by their ability to recognize and kill certain transformed cells and virus-infected cells in a spontaneous fashion without the need for prior sensitization, as well as to produce a battery of cytokines and chemokines.[3]As opposed to target recognition by cytotoxic T lymphocytes, recognition of target cells by NK cells is not restricted to major histocompatibility complex (MHC) antigen presentation. A microscopic feature of NK cells is the presence of distinct azurophilic granules in the cytoplasm, and we thus named them "large granule leukocytes" (LGLs). As well as in peripheral blood, NK cells are found in the spleen, liver, lymph nodes, bone marrow, intestine and lung. The liver has a unique anatomical environment that maybe closely correlated with its immunologic function, compared with peripheral blood and other organs like the spleen.

The liver is composed of hepatocytes and several types of non-parenchymal cells, including endothelial cells, dendritic cells, stellate cells, Kupffer cells and NK cells. Liver lymphocytes are abundant in NK cells, accounting for 25%-35% of rat liver lymphocytes (NKR-P1A+CD3-), 5%-10% of mouse liver lymphocytes (NK1.1+CD3+or DX5+), and 20%-30% of human liver lymphocytes (CD56+CD3-).[4,5]Liver NK cells reside in sinusoidal lumina, often adhering to endothelial cells and incidentally contacting Kupffer cells. Liver NK cells, also known as pit cells, display markedly distinctive immunophenotypical, morphological and functional characteristics different from those in the peripheral blood and spleen, identifying them as a peculiar subpopulation.[6,7]In 1976, Wisse et al[8]first described "pit cells", a type of rat liver sinusoidal cells so named because they contain the characteristic features of spherical dense granules and rod-cored vesicles in the cytoplasm.

A considerable set of data demonstrates that NK cells are a heterogeneous population of lymphocytes.[9]Compared to NK cells freshly isolated from the peripheral blood and spleen, rat liver NK cells are larger and contain more rod-cored vesicles and more, but smaller, cytoplasmic granules.[7]The morphological features of liver NK cells suggest that they are a more mature form of peripheral blood and spleen NK cells. Liver NK cells exhibit a higher level of cytotoxic activity against a variety of tumor cells,[10,11]similar to IL-2-activated NK cells, and a higher level of expression of cell adhesion molecules than peripheral blood and spleen NK cells. Rat liver NK cells are more cytotoxic against Yac-1 lymphoma cells and CC531s colon carcinoma cells than freshly isolated peripheral blood and spleen NK cells, although their cytotoxicity is lower than that of IL-2-activated NK cells.[6,12]In contrast to freshly isolated peripheral blood and spleen NK cells, liver NK cells have higher cytotoxic activity and can effectively lyse NK-resistant tumor cells, such as P815 mastocytoma cells, which are sensitive to IL-2-activated NK cells.[6,10,12,13]Human liver NK cells have a higher cytotoxicity against NK-sensitive K562 erythromyeloblastoid leukemia cells and can kill NK-resistant Raji and Daudi cells originating from Burkitt's lymphoma.[14]Therefore, these observations seem to support the classical view that liver NK cells are a subset of naturally activated NK cells and their cytotoxic function is greater than that of spleen NK cells. In recent years, studies have provided new insights into the functions of immune cells in the tumor microenvironment, as well as in the physiological microenvironment.

It has long been recognized that many solid tumors develop a pathophysiological microenvironment characterized by high lactate concentration, low extracellular pH (pHe), low oxygen tension (pO2), low glucose concentration and/or energy deprivation.[15-20]The average pHe in a variety of tumors is about 0.5 unit lower than the corresponding normal tissue, whereas the intracellular pH (pHi) of both tissues is similar.[15,16,21]So, the influences of pHe are becoming increasingly germane to research in tumor immunology. Studies has reported the negative effects of an acidic microenvironment on immune cells and their functions.[22]Severin et al[23]showed decreased cytotoxicity of human lymphokineactivated killer cells against K562 cells with a lowered pHe. Loeffler et al[24,25]found a similar inhibition of murine NK cell activity and IL-2-stimulated lymphocyte proliferation by an acidic pHe. Likewise, a direct correlation between diminished lysis of various tumor cell lines by cytotoxic T lymphocytes and pHe has been described by Redegeld et al.[26]However, perhaps surprisingly, most of the work to date has focused primarily on immune cell function, with relatively few studies on the function of liver NK cells under acidic conditions. Moreover, there is an increasing awareness among immunologists and oncologists that the characteristics of liver NK cells are functionally distinct from spleen NK cells under tumor conditions.

Efforts to improve the therapeutic gain of current cancer treatments focus on strategies that exploit the properties of immune cells in the tumor microenvironment. For this reason, our experiments were undertaken to explore the morphological and functional characteristics of liver NK cells and the differences in NK cell characteristics between liver and spleen NK cells under normal and acidic conditionsin vitro. To this end, we examined the morphology, cell death, proliferative activity, IFN-γ production and cytotoxic activity of NK cells. These studies were done in rats because this species provides an excellent source of liver and spleen NK cells, and is used in many models of antitumor immunity.

Methods

Cell line and culture conditions

The mouse lymphoma cell line Yac-1, purchased from the American Type Culture Collection, was routinely cultured in complete roswell park memorial institute (RPMI) culture medium 1640 (4.5 g/L D-glucose, 300mg/L L-glutamine, 1 mmol sodium pyruvate and 10 mmol HEPES; Gibco) supplemented with 10% fetal bovine serum (FBS; Biological Industries), penicillin (100 IU/mL) and streptomycin (100 μg/mL, both from Sigma-Aldrich) at 37 ℃ in a humidified incubator with 95% air/5% CO2designated as culture at the steady-state condition. Cell viability was assessed using the trypan blue exclusion test and routinely found to contain ＜5% dead cells.

Animals

Male Sprague-Dawley rats, 6-8 weeks old and weighing 180-220 g, were purchased from the Experimental Animal Center of Sun Yat-Sen University. The rats were housed in laminar flow and specific pathogen-free cabinets, under controlled temperature and humidity and a 12-hour light/dark cycle with sterile food and water availablead libitum. Rat handling and experimental procedures were conducted in accordance with the institutional guidelines for animal care and use.

Isolation and purification of liver and spleen NK cells

Under sodium pentobarbital anesthesia, the portal vein was cannulated and the inferior vena cava was clamped. First, the liver was perfused with phosphatebuffered saline (PBS) through the portal vein at physiologic pressure (10-12 cmH2O) to remove the blood. Perfusion with complete RPMI culture medium 1640 containing collagenase IV (0.2 mg/mL) and DNase I (0.02 mg/mL, both from Sigma) was resumed at an elevated pressure (50 cmH2O), and the perfusate was collected. After perfusion, we homogenized the liver by mincing it finely with scissor and forcing it through a metal strainer; the tissue was then completely digested by digestion buffer (complete RPMI culture medium 1640 containing type IV collagenase and DNase I) at 37 ℃ for 30 minutes. The majority of the hepatocytes were removed by centrifugation at 30×g for 5 minutes. Mononuclear cells (MNCs) were isolated from the spleen, removed aseptically, crushed with the hub of a syringe in complete RPMI culture medium 1640, and gently pressed through a stainless steel sieve. After collection by density gradient centrifugation at 400×g for 30 minutes at room temperature over Ficoll-PaqueTMPREMIUM (1.084 g/mL, GE Healthcare) and resuspension in complete RPMI 1640, liver and spleen MNCs were isolated by the same steps. After centrifugation, MNCs present at the interface were collected, resuspended in complete RPMI 1640, and passed through a nylon wool column to deplete adherent cells, such as B lymphocytes, macrophages and monocytes. MNCs were harvested, washed twice with PBS, and suspended in complete RPMI 1640 for further purification.

Dynabeads? FlowCompTMFlexi (110.60D, Invitrogen) is an indirect magnetic labeling system for the isolation of NK cells from MNCs. Briefly, MNCs were incubated with DSB-X labeled antibodies (CD3 and NKR-P1A) and then mixed with FlowComp Dynabeads. Finally, the bead-bound cells were subjected to two consecutive isolations using a magnet: negative selection of magnetic CD3-labeled cells and positive selection of magnetic NKR-P1A-labeled cells, according to the manufacturer's protocol. The purity of the isolated NK cells (CD3-NKRP1A+) was analyzed with flow cytometric analysis by staining with anti-CD3-fluorescein isothiocyanate (FITC, IgM, clone IF4) and anti-NKR-P1A-phycoerythrin (PE, IgG1, clone 10/78, both from Caltag) mouse anti-rat monoclonal antibodies, and was approximately 95%. Cell viability was determined by trypan blue exclusion and was always ＞95%.

IL-2-stimulation of liver and spleen NK cells

Liver and spleen NK cells were obtained as described above and plated in 6-well plates at 1×106cells/mL. The cells were allowed to recover for 1 hour in complete RPMI 1640 with 10% FBS, followed by incubation with or without IL-2 (100 IU/mL, PeproTech) at 37 ℃ in a humidified atmosphere containing 5% CO2for 4 days, and then used for experiments as described below.

Exposure of liver and spleen NK cells to low pH

Liver and spleen NK cells (2×106cells/well) were obtained as above and seeded in 6-well plates. After a recovery period of 1 hour incubation in complete RPMI 1640 with 10% FBS, the cells were exposed to low pH (5.6 or 6.5, both from Genom Co.) or 7.2 as control for different times. The medium and cells were harvested at 4, 8, 12 and 16 hours after exposure to different cultural conditions.

Electron microscopy

The morphology of liver and spleen NK cells, obtained after fresh isolation, stimulation with IL-2 for 4 days or exposure to low pH or control for up to 16 hours, was examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM).

For SEM, the cells were first fixed with 2.5% glutaraldehyde at room temperature for 2 hours and washed twice with PBS. Next, a postfixation step was performed for 1 hour with 1% OsO4. The post-fixed cells were dehydrated step by step with ethanol, before critical-point drying (CPD, Balzers Union) from CO2. The dried cells were coated with gold under vacuum andexamined in an XL-30E SEM (Philips) at 15 kV.

For TEM, the cells were fixed with 2.5% glutaraldehyde and postfixed with 1% OsO4. The post-fixed cells were then dehydrated in an ascending-graded series of ethanol, and loaded onto electron microscope (EM) copper grids coated with formvar/carbon. Then, the grids were counterstained with uranyl acetate and lead citrate before examination in a CM10 TEM (Philips) at an accelerating voltage of 80 kV.

Cell death assay

Phosphatidylserine exposure, a hallmark of cell death, is directly detectable by recombinant Annexin V. The discrimination of cell death was quantified by propidium iodide (PI, 20 μg/mL, Sigma) and Annexin V-FITC recombinant protein (Bender Medsystems) counterstaining according to the protocol provided by the manufacturer. Briefly, liver and spleen NK cells were collected, exposed to low pH or control for 4, 8, 12 and 16 hours, washed twice with PBS, resuspended in 1× binding buffer (5×105cells/mL), and mixed with 5 μL Annexin V-FITC recombinant protein in a final volume of 200 μL for 15 minutes at room temperature. After washing twice with PBS, the cells were resuspended in 200 μL 1×binding buffer containing 10 μL PI. Data were acquired using a Becton Dickinson FACScalibur Flow Cytometer within 30 minutes after staining and analyzed using CELL-QuestTMsoftware. Cell death was defined as the percentage of Annexin V+PI+cells.

Proliferation assay

After overnight incubation at 4 ℃ in ice-cold 75% ethanol, harvested liver and spleen NK cells (5×105cells), exposed to low pH or control for 4, 8, 12 and 16 hours, were then washed twice with PBS, and stained with 450 μL PI (50 μg/mL) and 50 μL RNase (5 mg/mL, Sigma) for 30 minutes in the dark at 4 ℃. Samples were then processed on a flow cytometer using CELL-QuestTMdata acquisition and analysis software. The proliferation index was calculated as follows: Proliferation index (%)= (S+G2/M)/(G0/G1+S+G2/M)×100.

IFN-γ ELISA

Cell-culture supernatants were harvested at 4, 8, 12 and 16 hours from liver and spleen NK cells after exposure to low pH or control and centrifuged at 900× g for 15 minutes to remove cells. The concentration of IFN-γ in each sample was measured using a rat IFN-γ ELISA Kit (KRC4021, Invitrogen), with the subsequent assay done as recommended by the manufacturer. Absorbance was read by a spectrophotometric microplate reader (MK-3 Microplate Reader, Thermo labsystems) at an absorbance of 450 nm.

Cytotoxic assay

NK cell-mediated cytotoxic activity was determined in a colorimetric assay based on the measurement of LDH released from the cytosol of lysed Yac-1 target cells into the supernatant with the CytoTox 96? Non-Radioactive Cytotoxicity Assay (Promega) according to the manufacturer's instructions. During the cytolysis assay, Yac-1 target cells (104cells/well) were coincubated with liver or spleen NK cells, prestimulated for 4 days with or without IL-2 (100 IU/mL) at indicated effectorto-target (E:T) cell ratios ranging from 40:1 to 0.625:1, or exposed to low pH (5.6 or 6.5) or control (7.2) for different times (4, 8, 12 and 16 hours) at 10:1 E:T cell ratio in 96-well round-bottom plates. After a 4-hour incubation period in a humidified chamber (37 ℃, 5% CO2), the plates were centrifuged at 250×g for 4 minutes, and aliquots of supernatant were then transferred to fresh 96-well flat-bottom plates. The reconstituted Substrate Mix was added to each well, and incubated at room temperature for 30 minutes with foil to protect it from light. Finally, stop solution was added to each well, and the absorbance was measured at 490 nm using a spectrophotometric microplate reader within 1 hour. The percentage of specific lysis was expressed according to the following formula: Cytotoxicity (%)=(Experimental-Effector Spontaneous-Target Spontaneous)/(Target Maximum-Target Spontaneous)×100.

Statistical analysis

Results are expressed as mean±SD. The differences between experimental groups were analyzed using the paired samples Student'sttest or repeated measures analysis of variance where appropriate. Values ofP＜0.05 were considered statistically significant.

Results

Ultrastructural features of liver and spleen NK cells under different culture conditions

In the scanning electron microscope (data not shown), IL-2-activated spleen NK cells were substantially larger than freshly isolated spleen NK cells. Stimulation of liver NK cells with IL-2 resulted in a minor increase in size. The freshly isolated liver NK cells were intermediate in size between freshly isolated and IL-2-activated spleen NK cells. Liver NK cells had a low response to IL-2, while spleen NK cells showed a marked response.

Fig. 1. Ultrastructural features of freshly isolated liver NK cells (A) and liver NK cells following stimulation with IL-2 (B) for 4 days or treatment with control (pH=7.2, C) or low (pH=6.5, D or pH=5.6, E) after 16 hours in culture. Change in a number of granules is illustrated by TEM. Randomly selected specimens were examined.

Fig. 2. Ultrastructural features of freshly isolated spleen NK cells (A) and spleen NK cells after IL-2 stimulation (B) for 4 days or treatment with control (pH=7.2, C) or low (pH=6.5, D or pH=5.6, E) after 16-hour incubation. Change in a number of granules is shown by TEM. Randomly selected specimens were examined.

In the transmission electron micrographs (Figs. 1 and 2), liver and spleen NK cells shared several similar ultrastructural characteristics: a pronounced asymmetry with an eccentric and kidney-shaped nucleus, a conspicuous nucleolus, and characteristic membranebound electron-dense granules with variable number, shape and density, mainly localized in the nuclear notch. The main morphological characteristic of liver NK cells, compared to spleen NK cells, was the presence of more abundant, but smaller granules in the cytoplasm. The IL-2-activated counterparts were richer in granules than in freshly isolated liver and spleen NK cells. The number and size of granules of freshly isolated liver NK cells were intermediate between freshly isolated and IL-2-activated spleen NK cells. The number of granules in liver NK cells was reduced when the pH was decreased from 7.2 to 5.6 (Fig. 1). Similar results were obtained with spleen NK cells (Fig. 2). However, at pH 5.6 (Figs. 1E and 2E), liver NK cells contain a higher number of granules per cell than spleen NK cells. Morphologic analysis showed that liver NK cells may be more resistant to acidic pH, as defined by morphological alteration.

Cell death of liver and spleen NK cells under normal and acidic conditions

Fig. 3. Acidic pHe for 16 hours induces an increase in cell death of spleen NK cells compared to liver NK cells. Data shown are percentage of NK cells that were Annexin V+and PI+doublepositive. *:P＜0.05.

After a 16-hour incubation period, both liver and spleen NK cells incubated in acidic pH showed increased cell death compared with those in normal pH (P＜0.05; Fig. 3) in a time-dependent manner (data not shown). Especially in spleen NK cells, the cell death rate was 23.74% at pH 7.2, 27.28% at pH 6.5, and 34.12% at pH 5.6. In addition, under the same conditions, the cell death rate was lower in liver NK cells than in spleen NK cells (P＜0.05). Collectively, these data demonstrated that spleen NK cells are more sensitive to acidic pHe, which clearly induced and enhanced cell death, as compared with liver NK cells.

Proliferative activity of liver and spleen NK cells under normal and acidic conditions

Fig. 4. Differences in proliferative activity between liver and spleen NK cells exposed to low pH or control for 16 hours. *:P＜0.05.

Fig. 5. A strong inhibition of IFN-γ production by spleen NK cells was observed after exposure to low pHe as compared with control for 16 hours, but no inhibition of IFN-γ production by liver NK cells. *:P＜0.05.

Although acidification of the microenvironment induced a marked inhibition of the proliferative activity of spleen NK cells within a 16-hour incubation period, its effects were much less pronounced in liver NK cells (P＜0.05, Fig. 4). A strong inhibition of the cellular proliferation was detected at the lowest pHe of 5.6 in spleen NK cells: 22.85% at pHe 7.2 versus 14.55% at pHe 5.6, a decrease of about 36%. Furthermore, exposure to the acidic environment suppressed the cellular proliferation in a time-dependent manner (data not shown). Acidic pHe showed a significant decrease in cellular proliferation of spleen NK cells as compared to liver NK cells, which also suggesting that liver NK cells are more resistant to acidic pHe.

Production of IFN-γ from liver and spleen NK cells under normal and acidic conditions

The production of IFN-γ is a vital function of NK cells,[3]which plays an important role in NK cellmediated cytotoxic activity. High-level secretion of IFN-γ was detected from liver NK cells compared with spleen NK cells in normal or acidic pH medium up to 16 hours (Fig. 5). Liver NK cells secreted more IFN-γ than spleen NK cells, i.e. 135.91 pg/mL versus 81.22 pg/mL at pHe 7.2,respectively. Furthermore, liver NK cells as well as spleen NK cells already constitutively secreted a detectable amount of IFN-γ within 4 hours, and this continued to accumulate in a time-dependent manner with the highest IFN-γ levels after 16 hours (data not shown). Exposure of liver NK cells to acidic pHe only marginally reduced IFN-γ production (P＞0.05, Fig. 5). However, acidic pH had a strong inhibitory effect on IFN-γ release by spleen NK cells, from 81.22 pg/mL (pHe 7.2) to 73.5 pg/mL (pHe 5.6), reduced by about 10% (P＜0.05). The results showed that IFN-γ secretion by liver NK cells was stable in the pH range used, from 7.2 to 5.6.

Fig. 6. The cytotoxicity of liver NK cells was enhanced following exposure to IL-2, but the increase was much lower than that observed in IL-2-activated spleen NK cells (P＞0.05).

Cytotoxic activity of liver and spleen NK cells under different culture conditions

NK cells have been shown to respond to IL-2, including augmenting NK cell cytotoxic function, inducing NK cell proliferation, and producing NK-relevant molecules.[27,28]In agreement with many previous studies,[6,10-14]freshly isolated liver NK cells were more cytotoxic against Yac-1 target cells than spleen NK cells. The cytotoxicity of liver NK cells against Yac-1 target cells was enhanced following exposure to IL-2, but the increase was much lower than that observed in IL-2-activated spleen NK cells (P＜0.05, Fig. 6). Therefore, liver and spleen NK cells differ in their cytotoxic response to IL-2.

The effects of pH on the cytotoxic activity of liver and spleen NK cells are summarized in Fig. 7. The cytotoxicity of both NK cell populations linearly decreased as the pH was lowered from pH 7.2 to 5.6 (P＜0.05), and gradually decreased in a time-dependent manner (data not shown). Liver NK cells showed higher levels of cytotoxic activity than spleen NK cells at an E:T ratio of 10:1 in control or acidic environment (P＜0.05), and the maximum difference was obtained at pH 5.6. After 16 hours incubation at pH 5.6, liver NK cells still showed higher cytotoxic activity, with approximateto 37.5% cytotoxicity against Yac-1 cells. In contrast, there was an almost 20% decline in spleen NK cell cytotoxicity from 37.3% (pHe 7.2) to 30.08% (pHe 5.6).

Fig. 7. Liver NK cells were more cytotoxic against Yac-1 cells than spleen NK cells under normal and acidic conditions, especially acidic pHe. *:P＜0.05.

Discussion

Liver NK cells constitute an especial resident population in the liver sinusoids, and their morphological, immunophenotypical and functional characteristics differ from those of peripheral blood and spleen NK cells.[7]Consistent with a considerable set of data,[9-12]morphologic and cytotoxic differences were found among liver and spleen NK cells under physiological conditions in our experiments. By morphological investigation, freshly isolated NK cells from the liver were characterized by a relatively larger size and the presence of more abundant, but smaller cytoplasmic granules (Figs. 1A and 2A) than spleen NK cells. Unlike liver NK cells, spleen NK cells become substantially larger in size and richer in granules after activation with IL-2 (Figs. 1 and 2). Furthermore, freshly isolated liver NK cells are more cytotoxic against Yac-1 target cells than spleen NK cells. IL-2 treatment significantly increased the cytotoxicity of spleen NK cells, but slightly increased the cytotoxicity of liver NK cells (Fig. 6). Therefore, the above observations confirmed that liver NK cells are naturally activated by the microenvironment of the liver sinusoid.

In the past few years, a growing accumulation of data has mainly focused on the function of tumor-infiltrating leukocytes in the tumor microenvironment. Compared with normal tissues, solid neoplasms often possess regions with an acidic microenvironment. Thistlethwaite et al[29]pointed out that the pHe of human tumors is below normal physiological levels with values varying from 5.6 to 7.7 and an average of 6.8. Experimental evidence is gradually emerging for an inhibition of leukocyte activity when the surrounding pH of tumors is reduced.[23-26]Nevertheless, a better understanding of the relation between tumors and tumor-infiltrating leukocytes still seems necessary to develop effective therapeutic approaches, especially for advanced cancer. The liver is the most common location of distant cancer metastasis. NK cells are a predominant lymphocyte population in the liver, and liver NK cells are thought to exert non-MHC-restricted tumor cell killing as a first line of the antitumor immune response. However, NK cell functions are usually impaired in cancer patients,[30]and pHe seems to be one important factor contributing to this impairment. In the majority of studies, for practical reasons, NK cells are derived from spleen or peripheral blood. In addition, little is known about the morphological and functional characteristics of liver NK cells compared with spleen NK cells in an acidic microenvironment. To gain insight into the differences among liver and spleen NK cells, we analyzed in detail the influence of an acidic pHe on NK cell morphology, cell death, proliferative activity, IFN-γ production and cytotoxic activity.

In the present study, we provided evidence that microenvironmental pH sharply interferes with NK cell functions. The inhibition profoundly impacts the ability to kill Yac-1 target cells. In accord with published data,[23-26]we found that NK cells incubated in an acidic medium exhibited poorer cytolytic activity than in a neutral medium (Fig. 7). Furthermore, liver NK cells had significantly greater levels of cytotoxicity than spleen NK cells under normal and acidic conditions, and the maximum difference was found at pH 5.6 (37.5% vs 30.08%). These results suggest that an acidic pHe is more influential in the cytotoxicity of spleen NK cells than that of liver NK cells. However, it should be noted that although displaying differences in cytotoxic activity, the question of whether the acidic microenvironment similarly regulated morphology, cell death, proliferative activity and IFN-γ production between liver and spleen NK cells was unclear.

The effects of external acidity on the morphology of NK cells are illustrated in Figs. 1 and 2. The number of granules in spleen NK cells was markedly reduced as the pH was lowered from 7.2 to 5.6. The change in number of granules of liver NK cells was similar; however, liver NK cells contained a higher number of granules per cell at pH 5.6. Interestingly, a decline in cytotoxic activity seemed to coincide with a decrease in the number of granules. Although the content of granules has not been well determined, they are deeply implicated to NK-mediated cytotoxic activity because they increase in number when NK activity is augmented by biological response modifiers. Another remarkable finding was that the cell death rate was lower at pH 7.2than at pH 5.6, and the cell death of spleen NK cells was elevated to levels higher than that of liver NK cells at pH 5.6 (34.12% vs 27.21%; Fig. 3). Moreover, decreasing external pH from 7.2 to 5.6 resulted in a quite dramatic drop (36%) in the proliferative activity of spleen NK cells (Fig. 4). In contrast, a modest reduction was found in the proliferation of liver NK cells within the range of 5.6-7.2 (18.41% vs 12.23%). Importantly, it should be noted that an acidic pHe only slightly, but not significantly, suppressed the secretion of the important chemokine IFN-γ by liver NK cells, while showing a strong inhibitory effect on IFN-γ production by spleen NK cells, from 81.22 pg/mL (pHe 7.2) to 73.5 pg/mL (pHe 5.6) (Fig. 5). Killing Yac-1 cells by NK cells exposed to normal or acidic conditions, appeared to correlate with the changes in morphology, cell death, proliferative activity and IFN-γ production. The differences in NK cell characteristics between liver and spleen NK cells under acidic conditions may be due to the inability of spleen NK cells to respond to acidosis.

In our experiments, we comparatively analyzed NK cells that derived from the liver and spleen. The sharp morphological and functional differences displayed by the two types of NK cellsin vitroimply that liver NK cells are unexpectedly resistant to pH shock. Therefore, this may constitute a very important finding in the context of tumor immunotherapy, and further consideration of the effects of the tumor microenvironment on immune function would appear to be warranted.

Contributors: LLH, ZW, CYJ, MJ and WYL participated in the research design. YJD, LGL and LTZ performed the experiments. LLH analyzed the data and wrote the paper. LLH and YJD contributed equally to this study. All authors contributed to the interpretation of the study and to further drafts. WYL is the guarantor.

Funding: The research was supported by grants from the National Natural Science Foundation of China (30671987 and 81000065).

Ethical approval: This study was approved by the Ethics Committee at Sun Yat-Sen University, China.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev Immunol 2001;1:41-49.

2 Moretta L, Ferlazzo G, Bottino C, Vitale M, Pende D, Mingari MC, et al. Effector and regulatory events during natural killer-dendritic cell interactions. Immunol Rev 2006;214:219-228.

3 Trinchieri G. Biology of natural killer cells. Adv Immunol 1989;47:187-376.

4 Racanelli V, Rehermann B. The liver as an immunological organ. Hepatology 2006;43:S54-62.

5 Gao B, Jeong WI, Tian Z. Liver: An organ with predominant innate immunity. Hepatology 2008;47:729-736.

6 Vanderkerken K, Bouwens L, Wisse E. Characterization of a phenotypically and functionally distinct subset of large granular lymphocytes (pit cells) in rat liver sinusoids. Hepatology 1990;12:70-75.

7 Wisse E, Luo D, Vermijlen D, Kanellopoulou C, De Zanger R, Braet F. On the function of pit cells, the liver-specific natural killer cells. Semin Liver Dis 1997;17:265-286.

8 Wisse E, van't Noordende JM, van der Meulen J, Daems WT. The pit cell: description of a new type of cell occurring in rat liver sinusoids and peripheral blood. Cell Tissue Res 1976;173:423-435.

9 Lotzová E. Definition and functions of natural killer cells. Nat Immun 1993;12:169-176.

10 Vermijlen D, Luo D, Froelich CJ, Medema JP, Kummer JA, Willems E, et al. Hepatic natural killer cells exclusively kill splenic/blood natural killer-resistant tumor cells by the perforin/granzyme pathway. J Leukoc Biol 2002;72:668-676.

11 Ishiyama K, Ohdan H, Ohira M, Mitsuta H, Arihiro K, Asahara T. Difference in cytotoxicity against hepatocellular carcinoma between liver and periphery natural killer cells in humans. Hepatology 2006;43:362-372.

12 Vujanovic NL, Polimeno L, Azzarone A, Francavilla A, Chambers WH, Starzl TE, et al. Changes of liver-resident NK cells during liver regeneration in rats. J Immunol 1995;154: 6324-6338.

13 Hata K, Van Thiel DH, Herberman RB, Whiteside TL. Natural killer activity of human liver-derived lymphocytes in various liver diseases. Hepatology 1991;14:495-503.

14 Winnock M, Garcia-Barcina M, Huet S, Bernard P, Saric J, Bioulac-Sage P, et al. Functional characterization of liverassociated lymphocytes in patients with liver metastasis. Gastroenterology 1993;105:1152-1158.

15 Wike-Hooley JL, Haveman J, Reinhold HS. The relevance of tumour pH to the treatment of malignant disease. Radiother Oncol 1984;2:343-366.

16 Rotin D, Robinson B, Tannock IF. Influence of hypoxia and an acidic environment on the metabolism and viability of cultured cells: potential implications for cell death in tumors. Cancer Res 1986;46:2821-2826.

17 Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res 1989;49:6449-6465.

18 Martin GR, Jain RK. Noninvasive measurement of interstitial pH profiles in normal and neoplastic tissue using fluorescence ratio imaging microscopy. Cancer Res 1994;54:5670-5674.

19 Helmlinger G, Yuan F, Dellian M, Jain RK. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med 1997;3: 177-182.

20 Brown JM, Giaccia AJ. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 1998;58:1408-1416.

21 Gerweck LE, Seetharaman K. Cellular pH gradient in tumor versus normal tissue: potential exploitation for the treatment of cancer. Cancer Res 1996;56:1194-1198.

22 Lardner A. The effects of extracellular pH on immune function. J Leukoc Biol 2001;69:522-530.

23 Severin T, Müller B, Giese G, Uhl B, Wolf B, Hauschildt S, etal. pH-dependent LAK cell cytotoxicity. Tumour Biol 1994; 15:304-310.

24 Loeffler DA, Juneau PL, Heppner GH. Natural killercell activity under conditions reflective of tumor microenvironment. Int J Cancer 1991;48:895-899.

25 Loeffler DA, Juneau PL, Masserant S. Influence of tumour physico-chemical conditions on interleukin-2-stimulated lymphocyte proliferation. Br J Cancer 1992;66:619-622.

26 Redegeld F, Filippini A, Sitkovsky M. Comparative studies of the cytotoxic T lymphocyte-mediated cytotoxicity and of extracellular ATP-induced cell lysis. Different requirements in extracellular Mg2+and pH. J Immunol 1991;147:3638-3645.

27 Robertson MJ, Ritz J. Biology and clinical relevance of human natural killer cells. Blood 1990;76:2421-2438.

28 Whittington R, Faulds D. Interleukin-2. A review of its pharmacological properties and therapeutic use in patients with cancer. Drugs 1993;46:446-514.

29 Thistlethwaite AJ, Leeper DB, Moylan DJ 3rd, Nerlinger RE. pH distribution in human tumors. Int J Radiat Oncol Biol Phys 1985;11:1647-1652.

30 Smyth MJ, Hayakawa Y, Takeda K, Yagita H. New aspects of natural-killer-cell surveillance and therapy of cancer. Nat Rev Cancer 2002;2:850-861.

October 21, 2011

Accepted after revision February 6, 2012

Author Affiliations: Department of Hepatobiliary Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University (Lv LH, Yu JD, Li GL, Chen YJ, Min J and Wan YL); Department of Breast Surgery, Guangzhou Women and Children's Medical Center (Long TZ); Guangzhou 510120, China; Department of Hepatobiliary and Pancreatic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China (Zhang W)

Yun-Le Wan, MD, Department of Hepatobiliary Surgery, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, 107W, Yanjiang Road, Guangzhou 510120, China (Tel: 86-20-34071173; Email: wanyldr@163.com)

? 2012, Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872(12)60162-3

To be able to distinguish, classify, and catalogue-external things on the basis of a secure order already established in the mind-this is at once intelligence and culture.—Maria Montessori