Natural killer cells in the nonpregnant murine uterus.

Recent studies on the origin and function of the large granular cells that accumulate at implantation sites in the rodent uterus during pregnancy have shown that these cells are highly differentiated natural killer (NK) cells. These findings raise questions about the presence and regulation of NK cells in the normal, nonpregnant uterus. For example, do NK cells comprise a major or minor population of leukocytes in the uterus? Are these uterine NK cells (uNK) similar in phenotype and function to NK cells in other organs? Is the population of uNK cells maintained by local proliferation and/or by influx via the blood stream from the bone marrow? This brief review will examine our current understanding of those questions based on the experimental literature on rodents and our own recent studies.

Maturation of murine natural killer precursor cells in the absence of exogenous cytokines requires contact with bone marrow stroma.

To test whether differentiation of natural killer (NK) precursor cells in unsupplemented long-term bone marrow culture requires contact with stromal cells, bone marrow cells (BMC) that were depleted of NK 1.1+ and Thy-1+ cells were cultured either directly on established, irradiated bone marrow stroma or in cell culture inserts separated from the stroma by a 0.45-micron microporous membrane. No exogenous cytokines were added. Depleted BMC produced NK 1.1+ nonlytic cells by day 7 and lytic NK 1.1+ cells by day 14 only when grown directly on the stroma. NK precursor cells survived in the inserts, however, and could respond to recombinant interleukin-2 in secondary culture. Thus, stromal cell-dependent proliferation and differentiation of lytic NK 1.1+ cells from their nonlytic NK 1.1- precursors requires direct interaction between stromal cells and NK precursors.

Stage-specific expression of activation antigens on NK cells at uterine implantation sites in mice.

The differentiation of the cellular components of the uterine decidua, in particular the life history of NK cells, is poorly understood. With the use of two mAbs that recognize stage-specific activation Ags on NK cells, we investigated the development of NK cells known as granulated metrial gland (GMG) cells. Immunohistochemical analysis demonstrated that mAb 3C2, but not mAb 4H12, recognized numerous cells throughout the uterine decidua basalis during early gestation. Isolated (panned) 3C2+ decidual cells from day 7 of pregnancy co-expressed the NK1.1 Ag, displayed NK cytolytic activity, and proliferated in IL-2-containing media. A small percentage of those cells expressed the GMG-associated Ag 4H12. Immunohistochemical analysis of serial sections at midgestation demonstrated that most of the 3C2+ NK cells co-expressed 4H12 Ag. During the later part of pregnancy, however, 3C2 expression in the decidua was down-regulated, and the cells expressed high levels of 4H12 Ag. When 3C2+ NK cells were isolated from cell suspensions of decidua from 7-day pregnant mice, and cultured in IL-2-containing medium, the cells developed the large and granular morphology characteristics of GMG cells, and acquired 4H12 Ag. These results demonstrate that 4H12+ GMG cells differentiate from 3C2+, NK1.1+, cytolytic precursors that reside in the decidua during early gestation.

Induction of osteopontin mRNA expression during activation of murine NK cells.

Osteopontin (OPN) is an extracellular phosphorylated glycoprotein expressed in bone, kidney, nervous tissue, bone marrow, and granulated metrial gland (GMG) cells in murine decidua. We recently demonstrated that GMG cells are differentiated natural killer (NK) lineage cells that share phenotypic, functional, and morphologic characteristics with adherent interleukin-2 (IL-2)-activated NK cells. We now show that conditions that induce resting splenic NK cells to develop into adherent, activated cells induce the expression of opn mRNA. Nonstimulated NK cells did not express opn mRNA detectable by Northern analysis. However, expression was evident by day 1-2 of culture of NK cells with IL-2, increased to high levels by day 4, and was maintained at high levels thereafter. Thus, expression of mRNA for OPN, a secreted protein associated with cell adhesion, embryonic development, tissue remodeling, and immune regulation, is up-regulated during the activation of NK cells.

Hamster monoclonal antibodies to murine natural killer cells.

To further the study of natural killer (NK) cells we have produced hamster monoclonal antibodies (MAbs) with reactivities to mouse NK cells. MAbs 4A2 and 3C2 were obtained by fusing spleen cells from Syrian hamsters immunized with IL-2-activated NK cells with Fox-NY myeloma cells. 4A2 antigen was expressed by bone marrow (BM)-derived IL-2-responsive NK cell precursors, by mature NK cells of the BM, and by a highly lytic subset of splenic NK cells, in addition to IL-2-activated NK cells. 3C2 antigen was also expressed by BM-derived NK cell precursors, by mature NK cells in the BM, at low levels by splenic NK cells, and at high levels by IL-2-activated NK cells. These MAbs are likely to provide useful reagents for the study of the life history and functional significance of NK cells.

Effects of rIL-7 on murine bone marrow NK precursor cells.

We recently demonstrated that functional NK cells are produced from their precursors in murine long-term bone marrow cultures without the addition of exogenous growth factors. Because IL-7 is known to be produced by bone marrow stromal cells and is a proliferation factor for some immature cells of the B and T cell lineages, we tested whether rIL-7 could support the proliferation or maturation of NK precursor cells. By itself, rIL-7 did not induce NK lytic activity in cultures of unseparated bone marrow cells (BMC), but it did augment the response of unseparated BMC to 50 U rIL-2/ml, with a maximal enhancement at 10 ng IL-7/ml. Depletion experiments demonstrated that the IL-7-induced increase in cytotoxicity was not due to NK precursors, however, but to mature T and NK cells. IL-7 preferentially increased the number of CD8+ cells. Preculture of NK 1.1-depleted BMC with IL-7 did not increase the total number of NK 1.1+ cells or the lytic activity generated from NK precursor cells. However, IL-2-responsive NK lineage cells survived better in IL-7-supplemented medium than in medium alone. Thus, soluble rIL-7 did not expand the NK precursor cell population in vitro but it maintained the viability and subsequent responsiveness to IL-2 of NK precursors.

Production and differentiation of NK linkage cells in bone marrow.

The production of natural killer (NK) cells from their bone marrow progenitors is a multistep developmental process requiring interaction of hemopoietic NK lineage cells, stromal cells and growth factors. In this review, recent studies on the generation of NK cells provide the basis for a model of NK cell production. The development of NK cells is suggested to be similar to that of B cells, the other major lymphocyte population produced in the bone marrow. This scheme, for which substantial data already exist and which is testable vis-à-vis other points, can be summarized as follows: large, proliferating progenitor NK cells (pro-NK) give rise, under the influence of stromal cells and/or growth factors to less rapidly proliferating NK precursor cells (pre-NK) which exhibit some but not all of the phenotypic and morphologic characteristics of NK cells. Pre-NK cells divided and differentiate into phenotypically recognizable primary NK cells in the bone marrow. Primary NK cells are not in cell cycle, have low but measurable lytic activity, and express the phenotype of resting NK cells. They exit the bone marrow to populate peripheral sites where they undergo further maturation, become activated by such agents as interferon and may undergo a new cycle of proliferation in response to viral infections or other stimuli.

Prostaglandin E2-induced changes in the phenotype, morphology, and lytic activity of IL-2-activated natural killer cells.

During pregnancy, many large granular NK lineage cells that are recognized by mAb 4H12 accumulate at uterine implantation sites. Similar large, granular, 4H12+ NK cells accumulate during the 2nd wk of culture of IL-2-activated NK cells. Inasmuch as uterine NK cell lytic activity is suppressed during pregnancy, apparently due to the effects of PGE2, we have analyzed the effect of PGE2 on the phenotype, morphology, and lytic activity of IL-2-activated NK cells as a model for NK cell differentiation at uterine implantation sites. When cultures of IL-2-activated NK cells were supplemented with 1 microM PGE2, the cells increased in size and granularity, accompanied by an increase in the number of cells expressing the 4H12 Ag. This effect was specific to PGE2 because PGF2 alpha at the same concentration had no effect. In addition, changes in morphology and up-regulation of 4H12 Ag expression could be augmented by 100 mM dibutyryl cAMP, but not 100 mM dibutyryl cGMP, suggesting that the PGE2 effects are mediated by changes in cytoplasmic cAMP levels. When indomethacin was added to cultures of IL-2-activated NK cells at the beginning of the culture period, 4H12 expression was markedly reduced. Indomethacin also increased the cytolytic capacity of NK cells and reduced the number of cells that developed the large granular morphology characteristic of 4H12+ cells. When activated NK cells were sorted into 4H12+ and 4H12- populations and tested for their ability to kill YAC-1 cells, we found that 4H12+ NK cells have lower lytic activity as compared to the 4H12- subset. Furthermore, when activated NK cells were labeled with [3H]TdR, combined with 4H12 labeling, we found that the 4H12+ subset was not proliferative. These results are likely to have direct relevance to NK cell differentiation at uterine implantation sites where NK cell activity is suppressed by PGE2, 4H12 expression is up-regulated and NK cells differentiate into large, granular and relatively nonlytic cells termed granulated metrial gland cells.

Renal cell carcinoma and natural killer cells: studies in a novel rat model in vitro and in vivo.

The transplantable rat kidney carcinoma (RKC) provides an excellent experimental model for immunological and therapeutic studies of renal cell carcinoma. In this report, we define the biological characteristics of RKC and explore the interactions between RKC and natural killer (NK) cells. RKC, a transplantable tumor of spontaneous origin, grows progressively over a 12-week period and metastasizes to the lung when implanted orthotopically in the kidneys of female Lewis rats. Rats bearing RKC survived for an average of 10.5 +/- 1.5 (SD) weeks postimplantation. Lung metastases were visible between 7.5 and 8.5 weeks postimplantation, and by 9 to 10 weeks the incidence of metastases reached approximately 67%. Injection of the NK cell-specific monoclonal antibody 3.2.3 depleted Lewis rats of their NK activity for up to 14 days. Adherent lymphokine-activated killer cells generated from the spleens of 3.2.3-injected rats were significantly less lytic than those from control rats and contained a significantly lower percentage of 3.2.3+ cells when analyzed by flow cytometry. Groups of rats were implanted with RKC and received injections of 3.2.3 biweekly to maintain depletion of NK cells or of a control antibody, NK1.1, specific for mouse NK cells. At 10 weeks postimplantation, 3.2.3-injected rats had significantly (P < or = 0.005) larger tumors (104.4 +/- 20.1 g) than NK1.1-injected rats (75.4 +/- 13.9 g). Spleen cells and peripheral blood cells from uninjected, tumor-bearing rats had a slight but nonsignificant decrease in NK activity against 51Cr-labeled YAC-1 targets over the course of RKC progression. The activity of adherent lymphokine-activated killer cells from tumor-bearing rats was lower than that from normal rats, but not significantly. Cultured RKC cells were killed by both splenic NK cells and adherent lymphokine-activated killer cells. These data demonstrate that RKC is NK sensitive and that tumor growth does not abrogate NK activity. The RKC tumor provides a model system for the analysis of immunological factors in renal cell carcinoma growth and presents opportunities for testing therapeutic interventions in a system that closely mimics the human disease.

Regulation of natural killer cell production in bone marrow of mice: no evidence for negative feedback control.

The possible presence of a negative feedback control mechanism regulating natural killer (NK) cell production in the bone marrow of B6 mice was investigated by depleting NK cells with a single intravenous injection of anti-ASGM1 antibody. At times ranging from 1 to 21 days following injection, the response of the bone marrow cells to this depletion was assessed by measuring both NK activity against YAC-1 target cells in a 51Cr assay and the frequency of NK precursors by limiting dilution analysis. For comparison, measurements of lytic activity were also done in the peripheral blood and spleen. Anti-ASGM1 injection resulted in a depletion of mature NK cell activity in all compartments tested as expected, as well as the depletion of NK precursors from the bone marrow. Clearance of the anti-ASGM1 from the circulation of recipient mice was biphasic. More than 99% of the antibody was eliminated in the first 4 min, while the remaining 0.24% had a half-life in the serum of 6 days. Bone marrow was able to produce new lytic NK cells which were detectable between days 6 and 10 after depletion. The bone marrow of the NK cell-depleted mice did not show an 'overshoot' (compensatory increase) in NK cell production through day 21 after depletion. It therefore appears that the NK cell production in the bone marrow is independent of the activity of the peripheral NK cell pool.

Production and differentiation of NK lineage cells in long-term bone marrow cultures in the absence of exogenous growth factors.

Neither lytic NK cells nor IL-2-responsive NK precursors were produced in myeloid (Dexter) long-term bone marrow cultures (LTBMC). However, when myeloid LTBMC were switched to lymphoid (Whitlock-Witte) conditions and reseeded ("recharged") with fresh bone marrow cells (BMC), nonadherent cells with NK lytic activity and NK 1.1+ phenotype were produced within 1-2 weeks without the addition of exogenous IL-2 to the cultures. NK- and T cell-depleted BMC proliferated extensively in switched cultures and in 2 weeks generated cells that lysed the NK target YAC-1 but not the LAK target P815. The presence of NK precursors in the cultures was confirmed by reculturing nonadherent cells harvested from recharged LTBMC in fresh medium containing 50 U rIL-2/ml. High levels of NK lytic activity were generated. Sequential expression of NK 1.1 and IL-2 responsiveness followed by lytic activity was demonstrated by harvesting cells early after recharge, prior to the appearance of lytic cells. Elimination of NK 1.1+ cells depleted the ability to respond to IL-2 in secondary culture. Our studies demonstrate that myeloid-to-lymphoid switched LTBMC support the proliferation and differentiation of NK lineage cells from their NK 1.1-, nonlytic progenitors in the absence of an exogenous source of growth factors.

Murine granulated metrial gland cells at uterine implantation sites are natural killer lineage cells.

Granulated metrial gland (GMG) cells, a population of morphologically distinct, bone marrow-derived cells in murine decidua that react with mAb 4H12, are shown in this report to express NK-specific Ag and to become cytolytic to the NK cell target YAC-1 when cultured in the lymphokine IL-2. When 1-mm3 explants of 8-day decidual tissue were cultured with IL-2, large numbers of 4H12+ GMG cells migrated out of the tissue. Migration was dependent on the amount of IL-2 used. This explant technique was used to isolate a pure population of GMG cells. The migratory activated GMG cells were phenotypically 4H12+, NK1.1+, LGL-1+/-, CD3-, and MAC-1-. Furthermore, the IL-2-activated GMG cells killed YAC-1 but not P815 cells in a 4-h 51Cr-release cytotoxicity assay. 4H12+ GMG cells from collagenase-digested decidual tissue also were analyzed for the presence of NK lineage Ag by flow cytometry and shown to coexpress the NK-associated Ag NK1.1 and ASGM1 but not the T cell Ag CD3 or macrophage Ag MAC-1 or F4/80. GMG cells isolated by collagenase digestion did not express LGL-1, an Ag associated with lytic NK cells. Our results demonstrate that GMG cells express Ag and functions characteristic of NK cells, and thus GMG cells can be assigned to the NK lineage. The possible relevance of NK cells at implantation sites is discussed.

Monoclonal antibody 4H12 recognizes subsets of adherent-lymphokine activated killer cells and splenic natural killer cells from pregnant and neonatal mice.

Using a new mAb, 4H12, that recognizes a plasma membrane-associated Ag on granulated metrial gland cells, we identified subsets of murine NK cells in spleen cell-derived adherent lymphokine activated killer cells and in the spleens of neonatal and pregnant mice. In spleen cell adherent-lymphokine-activated killer cultures, 4H12 Ag was detected on a small subset of cells after 7 days culture and expression increased with time to 70% of the cells after 21 days culture. 4H12+ cells were large (up to 70 microns) and granular. The Ag was also detected in the cytoplasmic granules of some, but not all 4H12 surface positive cells. Coexpression studies indicated 4H12+ cells were predominantly positive for the NK1.1 and ASGM1 Ag, negative for the MAC-1 and F4/80 Ag, and +/- for the LGL-1 and CD3 Ag. Subsets of 4H12+/-, LGL-1+/- exhibited morphologic characteristics restricted to specific phenotypic subsets. The 4H12-/LGL-1+ subset was shown to contain the smallest, least granular cells, whereas the 4H12+/LGL-1+/- subsets contained the largest and most granular cells. Although 4H12 expression was negligible in the spleens of normal adult mice, spleen cells of neonatal and pregnant mice contained subsets of NK1.1+ cells that coexpressed 4H12. The 4H12+/NK1.1+ and 4H12-/NK1.1+ subsets displayed differential levels of NK1.1 expression. 4H12+ NK cells were NK1.1 low to high, whereas 4H12- NK cells were NK1.1 high only. The functional significance of subsets of NK cells in IL-2 culture and in the spleens of neonatal and pregnant mice remains to be elucidated.

Trends in hospital deductions from revenue. Increases in uncompensated care and payment adjustments.

The 1980s produced tighter restrictions on Medicaid eligibility, greater reluctance among insurers to cover small groups, and increased price competition, resulting in a larger percentage of unreimbursed charges. Here the authors use financial data from a cross-section of California hospitals to explore the extent and variation of such deductions from revenue.

Injury Severity Scores in desert recreational all-terrain vehicle trauma.

A retrospective analysis of 1,259 patients injured in all-terrain vehicle accidents over a 2-year period was performed. All patients presented to the same community hospital Emergency Department, located near a major desert off-road vehicle recreational area. The patients were characterized by demographics and Injury Severity Scores. There were 2,149 recorded injuries. Mean ISS +/- standard deviation for all patients was 5.07 +/- 4.93; among patients admitted for inpatient care, 9.83 +/- 6.46. Injury Severity Scores were independent of vehicle type (two, three, or four wheels). In this setting, injuries sustained by children were not disproportionately more severe than those sustained by adults. Where documented, alcohol use was associated with higher ISS, and the use of helmets was associated with less severe head and facial injuries. The most common diagnoses in this series were integumental injuries and fractures, frequently involving the face and extremities.

Modulation of murine natural killer cells by a granulocytosis-inducing tumor.

The nonmetastatic neutrophilia-inducing murine mammary carcinoma CE1460 has been shown previously to have profound effects on hemopoiesis and lymphopoiesis. In this report we examined the effects of progressive growth of CE1460 on natural killer (NK) cell activity both in the bone marrow, the site of primary NK cell production, and in a peripheral site, the spleen. (BALB/c x CE)F1 mice were injected subcutaneously with trypsinized cells from in vivo passaged CE1460 or from B66, a BALB/c mammary carcinoma that does not induce neutrophilia. 3 days posttumor implantation, NK activity in bone marrow cells or spleen cells was greatly enhanced compared to normal controls. In B66 tumor-bearing mice, NK activity returned to normal by Day 7 and remained there through Day 14. In contrast, however, NK activity in CE1460 tumor-bearing mice decreased to only 10-20% of normal by Day 14. Excision of the tumor on Day 14, when WBC counts were three times normal, was followed by a rapid return of the WBC count to the normal range. NK activity in bone marrow and in spleen cells recovered somewhat but was still significantly suppressed 7 days after tumor excision. Limiting dilution analysis revealed a 3-5-fold decrease in frequency of NK precursors in bone marrow cells of mice bearing CE1460 for 7 or 14 days. The dramatic changes in NK activity observed in these experiments may reflect perturbation in production as well as an initial activation and subsequent suppression of mature NK cells.

Asialo-GM1+ natural killer cells directly suppress antibody-producing B cells.

Natural killer (NK) cells were tested for their ability to suppress antigen-induced antibody responses in vitro. Asialo-GM1+ (ASGM1+) cells were prepared from nylon-wool-nonadherent spleen cells obtained from normal mice. After depletion of Ig+, L3T4+ and Lyt-2+ cells, the ASGM1+-enriched cell population had high NK activity which was abrogated by treatment with anti-ASGM1 and C'. This NK-enriched ASGM1+ cell fraction significantly suppressed the generation of antibody-producing cells when added to in vitro immunization cultures of primed spleen cells. Treatment of the NK-enriched cell population with anti-ASGM1 and C' abrogated the ability of these cells to suppress antibody responses. In vitro antibody production by purified B cells was also suppressed in the presence of the NK-enriched cell population, although the kinetics of the suppression differed from that observed with unfractionated spleen cells. In addition, the NK-enriched cell population suppressed the proliferation of the B cell line WEHI-279.1. Suppression of WEHI-279.1 cells was abrogated when the NK-enriched cell population was treated with anti-ASGM1 and C'. These results suggest that normal NK cells suppress the generation of antibody-producing B cells and that this occurs, at least in part, through a direct regulation of the B cell.

Self-generating density gradients of Percoll provide a simple and rapid method that consistently enriches natural killer cells.

The separation or enrichment of natural killer (NK) cells from the heterogeneous cell populations in murine spleen or bone marrow is a vital step for the study of NK cells. We report in this study a simple and rapid method for the enrichment of NK cells from B cell-depleted spleen cells, using a self-generating density gradient of polyvinyl pyrrolidone-coated silica (Percoll). Nylon wool-passed spleen cells are suspended in Percoll that is isotonic and isosmotic with mouse blood at a density of 1.087 g/ml and ultracentrifuged at 30,000 x g for 10 min. This method consistently enriches for NK cell cytotoxic activity, in spleen cells of both unstimulated and interferon-stimulated mice, as measured in the chromium release assay. There is a concomitant enrichment for cells bearing the NK marker asialo GM-1 and depletion of L3T4 or Lyt-2-bearing T cells. In contrast to discontinuous, step-wise gradients, the self-generating Percoll gradient, which relies on the intrinsic property of Percoll to form a continuous density gradient, appears to provide the cells with a physiological environment both before and during the centrifugation step.

The primary role of murine bone marrow in the production of natural killer cells. A cytokinetic study.

The normal steady state production of natural killer (NK) cells in the bone marrow and spleen was characterized with cytokinetic technics. We developed a protocol to enrich for NK cells in bone marrow and demonstrate that target binding can be used as a criterion for marrow NK cells if nonspecifically "sticky" cells are eliminated. The selected population of B cell-depleted bone marrow lymphoid cells was comprised mainly of lymphocytes, of which 80% were NK-1.1+. B cell-depleted bone marrow lymphocytes that bound to YAC-1 could be characterized as two populations on the basis of morphology and proliferative status: large, proliferating target-binding cells (TBC), of which 25% were in S phase of the mitotic cycle, and small postmitotic TBC. Pulse and chase studies indicated that the small TBC in bone marrow were derived from an immediate proliferating precursor, presumably the large TBC, which were, in turn, derived from a precursor population that was more rapidly proliferating. In contrast, few if any splenic TBC were labeled after a 30-min pulse with [3H]TdR and significant numbers of labeled TBC did not appear in the spleen until 2 or more days after the pulse label. Surprisingly, some of the splenic TBC were relatively long lived and survived 2 mo or longer. These studies are the first to directly characterize the production of NK cells in situ in normal marrow. We demonstrate that the marrow is the primary site of production of NK cells and that little, if any, proliferation of NK cells occurs in the periphery of unstimulated mice. The data suggest the existence in the bone marrow of at least three compartments in the NK lineage: a rapidly proliferating NK precursor population, a less rapidly proliferating population of large TBC, and a population of small postmitotic TBC.

OK-432 stimulates primary production and activity of murine natural killer cells.

Although many immunostimulants have been shown to increase the lytic activity of natural killer (NK) cells in the periphery, little is known about their effects on NK cells in the bone marrow, the primary site of NK production. In the experiments reported here, we tested OK-432, a pharmaceutical preparation of Streptococcus pyogenes, for its effects on both the primary production and lytic activity of NK cells in C57BL/6J mice. NK activity in bone marrow cells (BMC) and spleen cells (SC) was significantly increased following intravenous administration of OK-432, peaking on day 2 in BMC and on day 3 in SC. Concomitantly, there were marked changes in the cellularity in the two compartments. Bone marrow cellularity fell significantly on day 1 post-OK-432 and then gradually returned to normal, whereas spleen cellularity rose rapidly and remained elevated. As a consequence, the total NK activity (per femur or per spleen) was significantly increased at 48-96 h after administration of OK-432. The target specificity was unchanged. The phenotype of NK cells in BMC as determined by cytotoxic depletion was unchanged by OK-432, but splenic NK activity shifted to a 'less mature' phenotype, intermediate between that of normal BMC and SC. Cytokinetic studies using 3H-TdR revealed an increase in the production of NK cells in the bone marrow following administration of OK-432. Proliferating NK cells also appeared in the spleen. Whether these were recently produced NK cells from the bone marrow that still retained the ability to proliferate or mature NK cells that were stimulated into cell cycle cannot be determined from these experiments. These data are the first to directly demonstrate the modulation of the primary production of NK cells by an immunologically active drug.