Published online before print September 28, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cancer Prevention & Research Center, Department of Pharmaceutical Sciences, College of Pharmacy, Washington State University, Pullman, Washington, USA
1 Correspondence: Cancer Prevention & Research Center, Department of Pharmaceutical Sciences, College of Pharmacy, Washington State University, Box 646713, 110 McCoy Office Trailer, Pullman, WA 99164-6713, USA. E-mail: meadows{at}wsu.edu
ABSTRACT
Alcohol consumption reduces peripheral NK cell numbers and compromises NK cell cytolytic activity; however, the underlying mechanism is not understood completely. It was found recently that the peripheral NK cell pool consists largely of bone marrow (BM)-derived and thymus-derived cells, which are phenotypically and functionally different. The effects of alcohol consumption on these subpopulations have not been studied previously. Using a well-established alcohol-feeding model, we found that chronic alcohol consumption decreases the percentage and number of peripheral NK cells, especially those expressing a mature phenotype. Alcohol consumption did not alter NK cells in the thymus. NK cells in the BM were increased significantly; however, proliferation rate was not altered by alcohol consumption, which increased CD127+ and decreased Ly49D+ NK cells in the spleen but not in the BM. Chronic alcohol consumption increased IFN-
-producing NK cells and GATA-3 expression in splenic NK cells. Collectively, these results indicate that chronic alcohol consumption perturbs the balance between thymus-derived and BM-derived NK cells. The increased proportion of thymus-derived NK cells in the spleen likely results from impaired NK cell release from the BM.
Key Words: immune • phenotype • distribution
INTRODUCTION
NK cells are a small population of lymphocytes, which comprise 1–5% of splenic lymphocytes and 5–10% of PBL in humans and mice. They are an important component of the innate immune system. NK cells interact with dendritic cells to regulate their maturation and cytokine and chemokine production during inflammation [1 2 3 ]. NK cells also produce cytokines and chemokines, which in turn, regulate T cell functions further [4 5 6 ]. Thus, NK cells play important roles in innate and adaptive immunity.
Based on the relative expression of the cell surface marker CD56, human NK cells can be divided into two subsets, CD56hi and CD56dim cells. These two subsets have distinct functions. The CD56hi NK cells produce large amount of cytokines such as IFN- but express low cytolytic activity. The CD56dim cells express greater cytolytic activity; however, they produce negligible cytokines after activation [7
8
9
]. CD56hi NK cells comprise only 10% of the total peripheral blood NK cells and are mainly found in lymph nodes [7
].
Mice do not express CD56; however, Vosshenrich et al. [10 ] reported recently that the presence or absence of CD127 surface expression can distinguish different subsets of NK cells. CD127+ NK cells, which also exhibit high expression of the transcription factor GATA3, are generated in the thymus, and CD127– NK cells, which express low levels of GATA3, originate in the bone marrow (BM). CD127+ NK cells are found mainly in the lymph nodes, and low amounts are found in the spleen and BM [10 ]. These cells lack expression of membrane-activated complex 1 (Mac-1), CD43, and Ly49, especially Ly49D, and the expression of c-Kit is up-regulated [10 ]. With these features, the CD127+ NK cells phenotypically and functionally resemble immature NK cells [11 , 12 ]. They exhibit low cytolytic activity and enhanced cytokine production [10 ]. Thus, the murine CD127+ NK cells are a counterpart to the human CD56hi NK cells [10 , 13 ].
Alcohol is a well-known immune suppressor. We and others have reported that chronic alcohol consumption, up to 2.5 months, reduces peripheral NK cell number, and it decreases NK cell cytolytic activity, which is related to decreased expression of perforin and granzymes A and B [14 15 16 17 ]. The underlying mechanism associated with these effects is unknown. In this study, we found that chronic alcohol consumption decreases the percentage and number of CD3– NK1.1+ cells in the spleen, increases the NK cells in the BM, and increases the proportion of thymus derived to the total population of NK cells in the spleen. These data support the hypothesis that chronic alcohol consumption interferes with the release of NK cells from the BM.
MATERIALS AND METHODS
Mice and alcohol administration
Female, C57BL/6 mice were obtained from Charles River Laboratories (Wilmington, MA, USA) and housed in the College of Pharmacy vivarium, which is Association for Assessment and Accreditation of Laboratory Animal Care-accredited. The animal protocol was approved by the Institutional Animal Care and Use Committee at Washington State University (Pullman, WA, USA). Mice were allowed to acclimate for 1 week, given free access to double-distilled, deionized water, and fed sterilized Purina 5001 laboratory chow. Then mice were randomly distributed into groups of eight to 10, based on their similarity in body weight. Each mouse was housed singly in a shoe box-style polycarbonate cage with CareFresh bedding (Absorption Corporation, Bellingham, WA, USA), given free access to the diet and water as described above or to filter-sterilized 20% w/v alcohol (diluted from EverclearTM) as the sole source of fluid. Consumption of alcohol and diet was monitored routinely, and intakes were consistent with previous chronic alcohol-feeding studies [14
]. Mice were maintained on a 12-h light/dark cycle at 22–24°C, and they were used in assays from 1 h to 2 h into the light cycle. We administered alcohol to mice for 2–6 months, as the goal of this project was to determine the effect of chronic alcohol consumption on NK cells. We used the rationale of Song et al. [18
] that human chronic alcoholics typically abuse alcohol for over 20 years. Based on the differences in longevity between mice and humans, we selected these time-points, which roughly correspond to 2–12 years of alcohol abuse in humans. Also, during this period, there are no differences in weight between alcohol-consuming and water-drinking mice [14
] (data not shown).
Antibodies
The following PE, FITC, biotin, and Cy5.5-conjugated anti-mouse antibodies were used in this study: NK1.1 (PK136), CD16 (2.4G2), Mac-1 (M1/70), CD43 (S7), CD3 (145 2C11), CD127 (SB/199), CD117 (2B8), Ly 49D (4E5), and BrdU-FITC. They were purchased from BD PharMingen (San Diego, CA, USA) or BioLegend (San Diego, CA, USA). Anti-mouse GATA3 (H-48) was from Santa Cruz Biotechnology (Santa Cruz, CA, USA).
Isolation of splenic lymphocytes
Mice were killed, and the spleen was removed. A single cell suspension was created from each spleen by pushing it through a wire screen into a sterile culture dish containing PBS + 0.1% BSA (buffer). The cell suspension was then passed through sterile Nitex nylon mesh to remove cell debris. The cell suspension was centrifuged for 10 min at 500 g at 4°C. The resulting pellet was resuspended in buffer and again passed through sterile Nitex, and the collected cells were centrifuged. The spleen cell suspension was diluted to 10 ml in buffer at room temperature and layered over 10 ml Lympholyte-M (Accurate Chemical and Scientific Corp., Westbury, NY, USA) in a 15-ml conical centrifuge tube. The leukocytes were separated from RBC by centrifuging the tube for 20 min at 2000 rpm at room temperature. The leukocytes were recovered by removing the layer of cells at the buffer:Lympholyte-M interface. These cells were washed, pelleted, and resuspended, and a total and viable cell count was determined with a Vi-CELL series cell viability analyzer (Beckman Coulter, Fullerton, CA, USA). Cell viability was >95%.
Isolation of BM cells
BM cells were isolated from the femur and tibia. Two ends of the bone were cut to expose the bone cavity. A 26-G needle attached to a 25cc syringe filled with PBS was inserted into the cavity, and the marrow was flushed with 5–10 ml PBS or until it was completely evacuated. A single cell suspension was obtained by passing the resulting BM and PBS through a 21-G needle. The BM cell suspension was washed with PBS. The cells were resuspended in 10 ml PBS and layered onto 10 ml Lymphocyte-M to remove the RBC as described above. BM cells were washed with buffer and counted with the Vi-CELL series cell viability analyzer.
Calculation of NK cell numbers
NK cell numbers in specific tissues were calculated by multiplying the total cell number in the tissue by the percentage of NK cells, as determined by flow cytometric analysis (below). Throughout this work, NK cells are defined as those that are CD3–NK1.1+. The number of Mac-1hi and CD43hi NK cells was calculated by multiplying the NK cell number by the percentage of Mac-1hi or CD43hi NK1.1+ cells in the total NK cells within a specific tissue, as determined by three-color flow cytometry.
Cell surface staining and flow cytometric analysis
One million cells were added per well to a 96-well plate and mixed with 5 µl anti-CD16 on ice for 10 min to block cell surface FcRs. Cells were washed with PBS + 0.1% BSA + 0.1% NaN3 (FACS buffer) once, and then an appropriate amount of fluorescently labeled antibody was added to the cell mixture to a final volume of 100 µl. The cells were placed on ice in the dark for 20 min, washed once, and then analyzed on a FACScan flow cytometer (BD Immunocytometry Systems, San Jose, CA, USA). Data from 20,000 gated events in the lymphocyte population were analyzed with CellQuest software (BD Biosciences, Mountain View, CA, USA).
BrdU administration and in vivo incorporation assay
The assay for in vivo incorporation of BrdU generally followed the procedure of Rubinstein et al. [19
]. Short-term (3 h) and long-term (4 days) BrdU experiments were conducted. For short-term BrdU administration, 2 mg BrdU in 200 µl PBS was injected via i.p. injection. Three hours after the injection, mice were killed, and BM cells and splenocytes were isolated for flow cytometry analysis. For long-term BrdU administration, each mouse was injected i.p. with 2 mg BrdU followed by administration of 0.8 mg/ml BrdU in the drinking fluid (water or alcohol) for 4 days. All BrdU solutions were prepared fresh daily, and the mice consumed equivalent amounts of the BrdU in alcohol or water. BM cells and splenocytes were isolated for flow cytometric analysis at the end of the BrdU administration period. Anti-CD16 was added, and the cells were stained with anti-CD3-PE and anti-NK1.1-PerCP. Stained cells were washed with FACS buffer and then fixed with Cytofix/CytopermTM at 4°C for 20 min. Fixed cells were washed twice with Perm/WashTM plus buffer before adding 30 µg DNase I at 37°C for 60 min. Cells were pelleted by centrifugation at 500 g for 10 min, washed twice with Perm/WashTM plus buffer, stained with anti-BrdU-FITC at room temperature for 30 min, washed once with FACS buffer, and then analyzed on the flow cytometer as described above. A total of 6 x 104–105 events was collected.
Intracellular staining of IFN- in activated splenic NK cells
Splenocytes were added to RPMI-1640 complete medium containing 10% FBS at the concentration of 2 x 106 cells/ml in a 24-well plate. Cells were activated with PMA (50 mg/ml) and ionomycin (100 mg/ml) and cultured with BD GolgiStop (BD PharMingen) at 37°C for 4 h. Cells were washed twice with 10 vol cold PBS + 0.1% NaN3. They were then blocked with anti-CD16 and stained with anti-CD3 and anti-NK1.1 for 20 min. They were washed twice with FACS buffer, fixed, and then permeabilized with Cytofix/CytoPerm reagent (BD PharMingen), following the manufacturer’s instruction. These cells were stained with anti- IFN- for 20 min, washed with buffer, and analyzed by flow cytometry as described above.
Intracellular staining of GATA3 in splenic NK cells
Splenocytes were blocked with anti-CD16 and stained with anti-CD3-FITC and anti-NK1.1-PerCP for 20 min. Cells were washed twice with FACS buffer, fixed, and permeabilized with Cytofix/CytoPerm reagent (BD PharMingen), following the manufacturer’s instruction. They were washed twice, stained with rabbit anti-mouse GATA3 for 30 min, washed twice, stained with donkey anti-rabbit IgG-PE (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and then analyzed with flow cytometry.
Statistical analysis
Differences between water-drinking and alcohol-consuming mice were determined with Microsoft Excel statistical computer program. The data are expressed as mean ± SD. Differences between groups were examined for significance by Student’s two-tailed t-test. Values were considered significant at P < 0.05.
RESULTS
Chronic alcohol consumption reduces the percentage and number of NK cells in the spleen
It is reported that the percentage and number of peripheral blood NK cells are reduced in alcoholics [20
]. Short-term (2 weeks) alcohol consumption in mice reduces the percentage and number of NK cells in peripheral blood and the spleen [21
, 22
]. In the present study, we found that the percentage and number of NK cells in the spleen were significantly lower in the mice consuming alcohol for 2 months and 6 months as compared with the water-drinking mice (Fig. 1A
and 1B
). The total number of NK cells in the spleen is lower at 6 months than at 2 months in water- and alcohol-fed groups. This most likely relates to the fact that NK cells decrease with age [23
].
 View larger version (19K): [in a new window] |
Figure 1. Effects of chronic alcohol consumption on NK cells in the spleen. Mice were given alcohol for 2–6 months. The percentage of CD3–NK1.1+ cells in the splenic lymphocytes (A) and the number of these cells in the spleen (B) were measured at the indicated time-points.  , Water-drinking mice;  , alcohol-consuming mice. Each group contained 10 mice. Values are mean ± SD; *, P < 0.05. The data are from one experiment, which was replicated once with similar findings. Results were obtained by two-color flow cytometric analysis.
|
 View larger version (26K): [in a new window] |
Figure 2. Chronic alcohol consumption decreases mature NK cells in the spleen. Mice were given alcohol for 2–3 months. Splenocytes were stained with anti-CD3/NK1.1/c-Kit, anti-CD3/NK1.1/CD43, or anti-CD3/NK1.1/Mac-1 and analyzed by three-color flow cytometry. (A) Percentage of c-Kit+, CD43hi, and Mac-1hi in the splenic CD3–NK1.1+ NK cells. (B) Numbers of Mac-1hi, CD43hi, Mac-1lo, and CD43lo NK cells in the spleen. , Water-drinking mice; , alcohol-consuming mice. Each group contained 10 mice. Values are mean ± SD. *, P < 0.05; **, P < 0.001. The data are from one experiment, which was replicated once with similar findings.
|
 View larger version (37K): [in a new window] |
Figure 3. Chronic alcohol consumption increases NK cells in the BM. Mice were given alcohol for 2–6 months. CD3–NK1.1+ cells were analyzed with two-color flow cytometry. (A) Representative dot-plot from water-drinking mouse. (B) Representative dot-plot from alcohol-consuming mouse. (C) Percentage of NK cells in the BM cells. (D) Numbers of NK cells in the BM. Cells from the 2-month groups were collected from one tibia and one femur. Cells from the 6-month groups were isolated from two femurs. (E) Numbers of Mac-1hi, CD43hi, Mac-1lo, and CD43lo NK cells in the BM as determined by three-color flow cytometric analyses. Data are from the 2-month groups. Cells were isolated from one tibia and one femur. , Water-drinking mice; , alcohol-consuming mice. Each group contained 10 mice. Values are mean ± SD. *, P < 0.05; **, P < 0.001. The data are from one experiment at each time period, and each experiment was repeated once with similar findings.
|
 View larger version (33K): [in a new window] |
Figure 4. In vivo BrdU incorporation in BM cells. Mice were given alcohol for 2 months and then treated with BrdU for 3 h or 4 days as outlined in Materials and Methods. BM cells were isolated and stained with anti-CD3/NK1.1/BrdU and analyzed by three-color flow cytometry. (A) Representative dot-plot of the gated NK cells. (B and C) Representative histograms of the BrdU+ cells in the gated BM NK cells from mice treated with BrdU for 3 h and 4 days, respectively. (D) Percentage of BrdU+ cells in the BM NK cells from mice treated with BrdU for 3 h and 4 days, respectively. Each group contained eight mice, and 6 x 104–105 total events were collected for each sample. , Water-drinking mice; , alcohol-consuming mice (ETOH). Values are mean ± SD. *, P < 0.05.
|
 View larger version (22K): [in a new window] |
Figure 5. Chronic alcohol consumption increases CD127+ NK cells in the spleen. Mice were given alcohol for 2–6 months. (A and B) Representative histograms of the CD127+ cells in the gated splenic CD3–NK1.1+ NK cells from water-drinking mice and alcohol-consuming mice, respectively. (C) Percentage of CD127+ cell in the splenic NK cells. , Water-drinking mice; , alcohol-consuming mice. Values are mean ± SD. **, P < 0.001. The data are from one experiment at each time period, and each experiment was repeated once with similar findings.
|
 View larger version (32K): [in a new window] |
Figure 6. Chronic alcohol consumption decreases Ly49D+ NK cells in the spleen but not in the BM. Mice were given alcohol for 2 months. BM cells and splenocytes were isolated and stained with anti-CD3/NK1.1/Ly49D and analyzed with three-color flow cytometry. (A) Representative, overlapping histogram of Ly49D+ cells in the gated splenic CD3–NK1.1+ population. Dashed line, Isotype control; thin line, water-drinking mice; bold line, alcohol-consuming mice. (B) Percentage of Ly49D+ cells in BM and splenic NK cells. Each group contained 8–10 mice. , Water-drinking mice; , alcohol-consuming mice. Values are mean ± SD. **, P < 0.001. The data are from one experiment at each time period, and each experiment was repeated once with similar findings.
|
 View larger version (29K): [in a new window] |
Figure 7. Expression of GATA3 in splenic NK cells. Mice were given alcohol for 2 months. Splenocytes were isolated and stained with anti-CD3/NK1.1/GATA3 and analyzed with three-color flow cytometry. (A) Representative histogram of GATA3 in splenic NK cells. Dashed line, Isotype control; thin line, water-drinking mice; bold line, alcohol-consuming mice. (B) Geometric mean fluorescence intensity (MFI) of GATA3 in splenic NK cells. Each group contained 10 mice. Values are mean ± SD. **, P < 0.001.
|
-producing NK cells in the spleen-producing NK cells in the total splenic NK population should be increased. To examine this, we determined IFN- expression in NK cells from spleen. After stimulation of splenocytes with PMA and ionomycin and analyzing IFN--producing cells by flow cytometry, we found that chronic alcohol consumption increased the percentage of these cells significantly in the splenic NK cell population (Fig. 8
).
 View larger version (28K): [in a new window] |
Figure 8. Chronic alcohol consumption increases the percentage of IFN--producing NK cells in the splenic NK cells. Mice were given alcohol for 2 months. Splenocytes were isolated and then stimulated with PMA/ionomycin. IFN- was detected with intracellular staining, and the percentage of IFN--producing cells in splenic NK cells measured by three-color flow cytometric analysis is indicated. , Water-drinking mice; , alcohol-consuming mice. Each group contained 10 mice. Values are mean ± SD. **, P < 0.001. The data are from one experiment at each time period, and each experiment was repeated once with similar findings.
|
Peripheral NK cells are often decreased in human alcoholics [20 , 24 ]. Experiments in mice show that high alcohol consumption can decrease the numbers of NK cells in the peripheral blood, spleen, and liver; however, these effects have not been well characterized [15 , 17 ]. This study provides further insight into the factors that affect the decrease in splenic NK cells during chronic alcohol consumption.
We found that the percentage and number of CD3–NK1.1+ are selectively depleted in the spleen of chronic alcohol-consuming mice (Fig. 1) . Further characterization of these cells revealed that the alcohol consumption differentially reduced NK cells, which exhibit a mature phenotype (Mac-1hi and CD43hi; Fig. 2 ). This finding led us to examine the effects of chronic alcohol consumption on NK cells in the BM and thymus, as they are the major sources of NK cells, which populate the spleen [11 ]. We hypothesized that chronic alcohol consumption would similarly deplete NK cells in these organs. However, we found no differences in the total numbers of NK cells in the thymus and increased numbers of immature (Mac-1lo and CD43lo) and mature (Mac-1hi and CD43hi) NK cells in the BM (Fig. 3) . These results suggested that the increase did not result from an influx of NK cells from the spleen. If this were the case, then we could have expected only the mature subtype of NK cells to increase in the BM. However, mature and immature NK cells increased in the BM during chronic alcohol consumption.
Another possibility, which could explain the increase in mature and immature NK cells in the BM, was that chronic alcohol consumption enhanced proliferation of these cells. Short-term BrdU incorporation into NK cells showed, however, that alcohol consumption did not affect the proliferation rate of NK cells in BM. Moreover, 4-day BrdU incorporation experiments showed that BrdU+ NK cells increased dramatically in the BM in alcohol-consuming mice as compared with water-drinking mice (Fig. 4) . These data suggest that release of NK cells from the BM is impaired.
With a decrease of BM-derived NK cells in the spleen, this should increase the percentage of thymus-derived NK cells. This was confirmed by the observations that the percentage of CD127+ NK cells (Fig. 5) was increased in the splenic NK cells of the alcohol-consuming mice along with the up-regulation of GATA-3 expression and decreased percentage of Ly49D+ NK cells (Figs. 6 and 7) .
Thymus-derived NK cells have lower cytolytic activity and higher cytokine production after activation than BM-derived NK cells [10
]. The fact that the numbers of BM-derived NK cells are decreased in the alcohol-consuming mice is consistent with previous reports that chronic alcohol consumption inhibits the cytolytic activity of NK cells [15
]. Moreover, the percentage of IFN--producing NK cells in the total population of NK cells in the spleen is higher in the alcohol-consuming mice than in the water-drinking mice (Fig. 8)
. This provides additional evidence that the spleen contains fewer BM-derived NK cells.
In summary, chronic alcohol consumption exhibits selective and differential responses on NK cells in the BM, spleen, and thymus. Chronic alcohol consumption, which depletes NK cells in the spleen, increases NK cell numbers in the BM. We show herein that one aspect associated with the decrease in splenic NK cells results from an inability of the BM to release sufficient NK cells to repopulate the spleen. This is supported by the data that chronic alcohol consumption decreases the phenotypically mature NK cells in the spleen and increases mature and immature NK cells in the BM. This suggests that the increase in BM-derived NK cells is not a result of an influx of NK cells from the spleen. The short-term (3 h) BrdU incorporation experiment indicated that chronic alcohol consumption does not alter the proliferation rate of NK cells in the BM. Thus, the increase is not a result of augmented production of NK cells. Last, the long-term (4 days) BrdU incorporation experiment demonstrated that BrdU+ NK cells significantly accumulate in the BM. Taken together, these data support the hypothesis that chronic alcohol consumption compromises the release of NK cells from the BM. The mechanism underlying the block in NK cell release from the BM is not known. Based on reports that chronic alcohol consumption inhibits NK cell migration to inflammatory sites [15 , 25 ], it is likely that the signaling mechanisms controlling NK cell migration to the spleen are compromised. We are currently studying this possibility along with other potential mechanisms.
ACKNOWLEDGEMENTS
This work was supported by grants R01AA07293 and R21AA014200 from the National Institute for Alcohol Abuse and Alcoholism. The authors thank Dr. Ya-Min Fu for his helpful discussions.
Received July 17, 2007; revised August 22, 2007; accepted September 5, 2007.
REFERENCES
} Proc. Natl. Acad. Sci. USA 103,9166-9171
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
