THE JOURNAL OF BIOLOGICAL CHEMISTRY
© 1997 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 272, No. 13, Issue of March 28, pp. 8276 –8280, 1997
Printed in U.S.A.
Rapid Import of Cytosolic 5-Lipoxygenase into the Nucleus of
Neutrophils after in Vivo Recruitment and in Vitro Adherence*
(Received for publication, November 18, 1996, and in revised form, December 23, 1996)
Thomas G. Brock‡§, Robert W. McNish, Marc B. Bailie, and Marc Peters-Golden
From the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, University of Michigan,
Ann Arbor, Michigan 48109-0652
Leukotrienes (LTs)1 are potent lipid mediators derived from
arachidonic acid (AA). They have important roles in regulating
both normal cell functions, such as proliferation (1), and inflammatory processes implicated in disease states, such as
asthma (2). The synthesis of LTs is initiated by the enzyme
5-lipoxygenase (5-LO) (3). Recent studies have demonstrated
that the subcellular distribution of 5-LO differs among resting
* This work was supported by National Heart, Lung and Blood Institute Grant R01 HL47391, National Institutes of Health Training Grant
T32 HL07749, Specialized Center of Research Grant P50 HL46487 (to
M. P. G.), and an American Lung Association research grant (to
T. G. B.). The costs of publication of this article were defrayed in part by
the payment of page charges. This article must therefore be hereby
marked “advertisement” in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
‡ T. G. B. is a Parker B. Francis Fellow in Pulmonary Research, an
Edward Livingston Trudeau Scholar of the American Lung Association,
and Amgen’s Research Grant Awardee.
§ To whom correspondence should be addressed: M3317 Medical Sciences I, Division of Pulmonary and Critical Care Medicine, University
of Michigan, Ann Arbor, MI 48109-0652. Tel.: 313-936-2612; Fax:
313-764-4556.
1
The abbreviations used are: LT, leukotriene; 5-LO, 5-lipoxygenase;
AA, arachidonic acid; IFM, immunofluorescent microscopy; PMN, polymorphonuclear leukocyte.
cells of different types: it is predominantly cytosolic in peripheral blood PMNs (4) and peritoneal macrophages (5), but it can
be found in both the nucleus and cytosol of alveolar macrophages (6) as well as mast cells (7) and the mast cell-like rat
basophilic leukemia cell line (4). Upon stimulation, both cytosolic and nuclear 5-LO can translocate to the nuclear envelope
(6, 8, 9), where 5-LO can interact with the 5-lipoxygenaseactivating protein (10) to oxygenate AA.
The LT synthetic capacity of a cell can be elevated via an
increase in the amount of 5-LO (11, 12), 5-lipoxygenase-activating protein (13, 14), or phospholipase A2 (15, 16) protein.
However, increases in LT synthetic capacity can also occur
independent of changes in protein levels (17–19). This indicates
that other mechanisms for increasing LT synthetic capacity
must exist. The fact that 5-LO can exist in two distinct subcellular compartments (the cytoplasm and the nucleoplasm)
raises the possibility that changes in 5-LO distribution may be
subject to modulation. Such redistribution may, in turn, effect
a change in LT synthetic capacity.
PMNs, purified from peripheral blood, are known to synthesize and secrete large amounts of LTB4 after stimulation. Blood
PMNs are recruited rapidly and in large numbers into sites of
inflammation and play a critical role in the rapid resolution of
bacterial and fungal infections. Interestingly, PMNs recruited
to sites of inflammation in rats have been found to produce
even more LTB4 than blood PMNs (20). Because the recruitment of PMNs occurs rapidly, the enhanced LT synthetic capacity associated with recruitment may be expected to result
from changes other than increases in 5-LO or 5-lipoxygenaseactivating protein levels. In this study, we asked whether the
recruitment of PMNs into sites of inflammation causes a
change in the subcellular compartmentation of 5-LO protein.
EXPERIMENTAL PROCEDURES
Cells—Human PMNs were isolated from venous blood obtained from
normal volunteers. Purification involved the sequential steps of centrifugation through Ficoll-Paque (Pharmacia Biotech, Inc.), dextran sedimentation, and hypotonic lysis of erythrocytes (21). Rat peripheral
blood PMNs and elicited PMNs were from respiratory disease-free male
Sprague Dawley retired breeder rats (Charles River Laboratories, Portage, MI). Rat peripheral blood PMNs were purified according to the
human blood PMN purification protocol (purity, 90 –97%).
Glycogen-elicited rat peritoneal leukocytes were obtained by standard methods (22). Briefly, 30 ml of 1.0% glycogen (Sigma) in saline were
introduced into the peritoneum of diethyl ether-anesthetized rats; elicited leukocytes were recovered by peritoneal lavage, typically 4 h after
glycogen instillation. Rat peritoneal leukocytes were also obtained 4 h
after the introduction of heat-killed Cryptococcus neoformans (strain
52; American Type Culture Collection, Rockville, MD; 1 3 108 colonyforming unit equivalents in saline). Differential counts of cell populations from either procedure indicated that 85–95% of the cells were
PMNs.
Bleomycin was used to induce inflammation in the lungs of rats by
standard methods (23). Alveolar leukocytes were examined 2 days after
intratracheal injection of bleomycin (0.75 unit/100 g body weight in 300
ml of saline). Leukocytes recovered by bronchoalveolar lavage (24) were
8276
This paper is available on line at http://www-jbc.stanford.edu/jbc/
Downloaded from www.jbc.org at Michigan State University, on January 18, 2012
5-Lipoxygenase catalyzes the synthesis of leukotrienes from arachidonic acid. The subcellular distribution of 5-lipoxygenase is known to be cell type-dependent and is cytosolic in blood neutrophils. In this study,
we asked whether neutrophil recruitment into sites of
inflammation can alter the subcellular compartmentation of 5-lipoxygenase. In peripheral blood neutrophils
from rats, 5-lipoxygenase was exclusively cytosolic, as
expected. However, in glycogen-elicited peritoneal neutrophils, abundant soluble 5-lipoxygenase was in the
nucleus. Upon activation with calcium ionophore
A23187, intranuclear 5-lipoxygenase translocated to the
nuclear envelope. Elicited neutrophils required a
greater concentration of A23187 for activation than did
blood neutrophils (half-maximal response, 160 versus 52
nM, respectively) but generated greater amounts of leukotriene B4 upon maximal stimulation (26.6 versus 7.68
ng/106 cells, respectively). Intranuclear 5-lipoxygenase
was also evident in human blood neutrophils after adherence to a variety of surfaces, suggesting that adherence alone is sufficient to drive 5-lipoxygenase redistribution. These results demonstrate a physiologically
relevant circumstance in which the subcellular distribution of 5-lipoxygenase can be rapidly altered in resting cells, independent of 5-lipoxygenase activation. Nuclear import of 5-lipoxygenase may be a universal
accompaniment of neutrophil recruitment into sites of
inflammation, and this may be associated with alterations in enzymatic function.
Nuclear Import of 5-LO in PMNs
RESULTS
5-LO Distribution in Peripheral Blood and Glycogen-elicited
Peritoneal PMNs—The distribution of 5-LO was evaluated by
two complementary techniques: 1) IFM, and 2) cell fractionation combined with immunoblotting. By IFM, 5-LO was predominantly cytosolic in PMNs purified from peripheral blood
(Fig. 1A), as has been described for human blood PMNs (4).
This pattern was observed in essentially all blood PMNs. In
sharp contrast, most PMNs elicited into the peritoneum after
glycogen instillation showed the opposite pattern: the nucleus
stained strongly positive for 5-LO, whereas the cytosol had
negligible staining (Fig. 1B). This pattern was evident in approximately 80% of all PMNs obtained by peritoneal lavage
after elicitation. Other PMNs showed either predominantly
cytosolic 5-LO or a mix of both cytosolic and nuclear staining
(arrow, Fig. 1B). Peritoneal macrophages, a minor contaminant, had predominantly cytosolic 5-LO (data not shown), as
reported previously (5).
When rat peripheral blood PMNs were analyzed as a population by fractionation and immunoblotting, 5-LO was found to
be entirely soluble and exclusively in the cytosol (Fig. 1C),
FIG. 1. The subcellular localization of 5-LO in peripheral blood
PMNs versus elicited PMNs. Blood PMNs were purified by standard
methods; elicited PMNs were recruited by instillation of glycogen into
rat peritoneum followed by lavage 4 h later. Cells were prepared and
analyzed as described under “Experimental Procedures.” A, IFM analysis of rat blood PMNs. B, IFM analysis of elicited PMNs. Arrow
indicates a cell with both intranuclear and cytosolic 5-LO. Cells stained
with nonimmune IgG yielded completely negative results. C, immunoblot analysis of cell fractions, including cytosol (C), nonnuclear membrane (M), nuclear soluble proteins (Ns), and nuclear pelletable proteins (Np). Results are representative of at least three independent
experiments.
consistent with conclusions based on IFM. By the same
method, 5-LO in lavaged peritoneal cells was divided between
the cytosolic and nuclear soluble fractions. Thus, two independent methods, one that evaluated individual cells and another
that analyzed cell populations, indicated that 5-LO was exclusively in the cytosol of peripheral blood PMNs, whereas glycogen-elicited PMNs contained abundant nuclear 5-LO. The apparent differences in the amount of cytosolic 5-LO in elicited
PMNs as judged by the two methods may be attributed to, on
the one hand, an underestimation due to loss during permeabilization for IFM and, on the other hand, an overestimation due
to unavoidable disruption of some nuclei combined with fractionation of a heterogeneous population (i.e. PMNs with nuclear 5-LO mixed with some peritoneal macrophages, as well as
PMNs, with cytosolic 5-LO).
The localization of 5-LO in elicited PMNs was further evaluated by optical sectioning of stained cells, using confocal microscopy (Fig. 2). An optical section through the nucleus of
elicited PMNs stained for both DNA (using acridine orange)
and 5-LO revealed colocalization of DNA and 5-LO in most
cells, although some cells retained cytosolic 5-LO (arrow, Fig.
2A). Serial sections through cells stained only for 5-LO demonstrated that 5-LO was dispersed homogenously throughout the
intranuclear volume rather than distributed peripherally near
the nuclear envelope (Fig. 2B).
5-LO Distribution in Other Types of Elicited PMNs—PMNs
can be recruited into a variety of tissue sites in response to a
number of different signals. We were interested in determining
if the nuclear import of 5-LO was unique to glycogen-elicited
peritoneal PMNs. 5-LO was also intranuclear in PMNs recovered from the rat peritoneum 4 h after instillation of heat-killed
C. neoformans (Fig. 3A), indicating that the result was not
glycogen-specific. 5-LO was also found to be intranuclear in
PMNs recruited into rat lung 2 days after intratracheal instillation of bleomycin as determined by either IFM of cells recovered from lung by lavage (Fig. 3B) or immunohistochemistry of
lung sections (Fig. 3C). The demonstration of intranuclear
5-LO in elicited PMNs in situ (by immunohistochemistry) indicated that the redistribution did not result from the lavage
process or the subsequent handling of PMNs. As an example of
elicited PMNs from humans, PMNs recovered by bronchoalveolar lavage of patients with pneumonia also exhibited intranuclear 5-LO (Fig. 3D). These results showed that 5-LO movement from the cytosol into the nucleus could occur in PMNs
recruited into different anatomic sites, in response to different
Downloaded from www.jbc.org at Michigan State University, on January 18, 2012
processed for indirect immunofluorescent microscopy (see below). Alternatively, lung sections (from different animals) were processed for
immunohistochemistry (see below). Human alveolar leukocytes were
obtained from patients with pneumonia undergoing diagnostic bronchoalveolar lavage, as described (25). The experimental protocol was
approved by the University of Michigan Medical Center Institutional
Review Board for Approval of Research Involving Human Subjects.
Indirect Immunofluorescence Microscopy (IFM)—Cells were prepared for IFM as described previously (4), using methanol (220 °C, 30
min) followed by acetone (220 °C, 3 min) to fix and permeabilize. They
were then probed with a rabbit polyclonal antibody raised against
purified human leukocyte 5-LO (a generous gift of Dr. J. Evans, Merck
Frosst Center for Therapeutic Research, Pointe Claire-Dorval, Quebec,
Canada; titer of 1:150) (8) followed by rhodamine-conjugated goat antirabbit antibody (1:200; Sigma). In some cases, cells were also stained for
DNA using acridine orange (in the presence of RNase). Preparations
were examined with a Nikon Labophot 2 microscope equipped for epifluorescence or imaged by confocal microscopy using a Bio-Rad MRC600 laser confocal microscope. The rhodamine signal was imaged using
a 560-nm-long pass filter followed by a 585-nm bandpass filter. Acridine
orange fluorescence was isolated using a 514-nm bandpass filter combined with a 526-nm-long pass dichroic reflector.
Cell Fractionation/Immunoblotting—As described previously (9),
cells were disrupted by nitrogen cavitation at 400 p.s.i. for 5 min at 4 °C.
Cavitate was centrifuged at 1000 3 g to pellet nuclei; the postnuclear
supernatant was centrifuged at 100,000 3 g for 30 min at 4 °C to
generate soluble (“cytosolic”) and insoluble (“membrane”) fractions. Nuclei were sonicated, and soluble and insoluble fractions were separated
by ultracentrifugation. Equivalent amounts of protein from each fraction were evaluated by immunoblot (5) using 5-LO antibody (1:5000)
and then peroxidase-conjugated goat anti-rabbit secondary (1:5000)
with chemiluminescent detection (Amersham).
Immunohistochemistry—Immunohistochemistry, as described (26),
utilized lung tissue specimens fixed in neutral-buffered formalin, embedded in paraffin, sliced, mounted, deparaffinized (Americlear), and
rehydrated in descending concentrations of ethanol. Tissue was blocked
(Power Block; Biogenics, San Ramon, CA), probed with 5-LO antibody
(1:500, 4 °C, 24 h), washed, and then incubated with biotinylated goat
anti-rabbit IgG (1:600). After washing, sections were incubated with
avidin-conjugated horseradish peroxidase reagent for 45 min, washed,
incubated with True-Blue peroxidase substrate (KPL Laboratories,
Gaithersburg, MD), and then counterstained with Contrast Red (KPL
Laboratories).
Cell Activation and Enzyme Immunoassay—PMNs in M199 at 106
cells/ml21 were activated with calcium ionophore A23187. After 15 min
at 37 °C, cells were pelleted, and the LTB4 concentration in the conditioned media was determined by enzyme immunoassay (Cayman
Chemical, Ann Arbor, MI). Average values of duplicate determinations
were obtained in each experiment.
Statistical Analyses—Statistical significance was evaluated by a
paired Student’s t test, using p , 0.05 as indicative of statistical
significance.
8277
8278
Nuclear Import of 5-LO in PMNs
FIG. 3. Nuclear localization of 5-LO in PMNs elicited by other
agonists and in other anatomic sites. A, PMNs elicited into rat
peritoneum by instillation of heat-killed C. neoformans. Elicited PMNs
were harvested by lavage 4 h after instillation; 5-LO was detected by
indirect immunofluorescent microscopy. B and C, PMNs elicited into rat
lung after administration of bleomycin by intratracheal instillation. 2
days after bleomycin administration, either lungs were lavaged, and
cells were probed for 5-LO and examined by IFM (B; AM, alveolar
macrophage) or lung tissue was fixed, sectioned, and stained for 5-LO
and examined by immunohistochemistry (C). In the latter, darker
staining results from the True-Blue signal localizing 5-LO, whereas the
lighter staining is from the Contrast Red counterstain. D, human lung
PMNs, derived by bronchoalveolar lavage of an individual with pneumonia, stained for 5-LO and examined by IFM. AM, alveolar
macrophage.
agents, and in humans as well as in rats.
The Effect of Subcellular Distribution on Function—How
does 5-LO redistribution affect 5-LO function? One hallmark of
5-LO action is its translocation to the nuclear envelope upon
cell stimulation, a process that colocalizes 5-LO with the AAbinding protein, 5-lipoxygenase-activating protein. As reported
for human blood PMNs (8, 9), after stimulation, 5-LO in rat
blood PMNs was found predominantly at the periphery of the
multilobed nuclei, apparently at the nuclear envelope (Fig. 4A).
The subcellular distribution of 5-LO in elicited PMNs after
activation was also consistent with translocation to the nuclear
envelope (Fig. 4B). However, the pattern of distribution was
subtly different from that seen in blood PMNs: in blood PMNs,
the nuclear lobes and connecting strands were clearly outlined
with fluorescence, whereas in elicited PMNs, only portions of
the nuclear lobes were labeled. These patterns would be consistent with translocation to different membrane faces of the
nuclear envelope, with cytosolic 5-LO decorating the outer
membrane, and intranuclear 5-LO binding to the inner membrane. Finally, the distribution of 5-LO in activated elicited
PMNs was distinctly different from that in resting elicited
PMNs (Fig. 1B), again indicating that 5-LO in resting elicited
PMNs was not simply associated with the nuclear envelope.
Elicited PMNs, like blood PMNs, generated little LTB4 without exogenous stimulation (7.3 6 2.3 versus 12.9 6 2.4 pg/106
cells, respectively; p 5 0.21; n 5 3), indicating that elicitation
itself did not activate the 5-LO pathway. We hypothesized that
a consequence of the shift in 5-LO localization seen with elicitation might be a change in the dose of ionophore required for
LT synthesis. Specifically, we reasoned that more ionophore
would be needed to cause an increase in nuclear calcium and
thus activate nuclear 5-LO than would be needed to have the
same result within the cytosol. Indeed, significantly more ionophore was needed to stimulate LTB4 synthesis in elicited
PMNs than in peripheral blood PMNs (half-maximal stimulation at 160 6 16 versus 52 6 4.6 nM A23187, respectively; p 5
0.0027; n 5 4). However, maximal LT synthetic capacity for
elicited PMNs (26.6 6 7.4 ng/106 cells at 407 6 210 nM A23187)
exceeded that for blood PMNs (7.68 6 0.84 ng/106 cells at 165 6
78 nM A23187) significantly (p 5 0.044; n 5 4). Dose-response
curves from a representative experiment are shown in Fig. 5.
Thus, a greater level of stimulation was needed to initiate LT
synthesis in elicited PMNs, but the maximal LT synthetic
capacity of elicited PMNs significantly surpassed that of blood
PMNs.
Possible Role for Adherence in Redistribution of 5-LO—PMN
recruitment is a multistep process, but one prerequisite for
recruitment to all anatomic sites is the adherence of PMNs to
the endothelium. If adherence triggers the redistribution of
5-LO from the cytosol to the nucleus of PMNs, then PMNs
examined at earlier time points after recruitment should also
show intranuclear 5-LO. Indeed, PMNs recovered by peritoneal
lavage at either 1 or 2 h after glycogen instillation (in separate
animals) exhibited intranuclear 5-LO (Fig. 6). This indicated
that the nuclear import of 5-LO occurred relatively early in
recruitment, resulting from an early event, such as adherence,
and was not necessarily a function of prolonged residence in the
inflammatory site. Furthermore, because elicited PMNs from
both human and rat sources demonstrated intranuclear 5-LO
(Fig. 3), it seemed likely that a common process, like adherence,
could cause the nuclear import of 5-LO in PMNs from either
source.
As an initial evaluation of the role of adherence in 5-LO
redistribution in human PMNs, peripheral blood PMNs were
Downloaded from www.jbc.org at Michigan State University, on January 18, 2012
FIG. 2. The subcellular localization of 5-LO in elicited PMNs by
confocal microscopy. A, colocalization of 5-LO and DNA. Rat peritoneal PMNs, elicited by glycogen instillation, were fixed and dualstained for DNA (left) and 5-LO (right), and dual images were collected
simultaneously. Arrow (right) indicates a PMN with both intranuclear
and cytosolic 5-LO. B, intranuclear localization of 5-LO demonstrated
by optical serial sectioning of two cells. Transverse optical sections were
obtained at 1-mm steps, beginning (a) close to the slide and proceeding
upward (b–f). An image of each optical section was created by averaging
six 1-s scans.
FIG. 4. Translocation of intranuclear 5-LO to the nuclear envelope of elicited PMNs. Rat peripheral blood PMNs and elicited
PMNs were treated with 1 mM A23187 for 15 min at 37 °C and then
probed for 5-LO. Cells were prepared and analyzed as described under
“Experimental Procedures.” A, IFM analysis of activated rat blood
PMNs. B, IFM analysis of activated, elicited PMNs. Results are representative of at least three independent experiments.
Nuclear Import of 5-LO in PMNs
FIG. 5. Comparison of the LTB4 synthetic capacity of PMNs
from rat peripheral blood and glycogen-elicited peritoneal
PMNs. Cells (0.5 3 106 ml21) were activated with A23187 for 15 min at
37 °C. LTB4 in the conditioned media was determined by enzyme immunoassay. Data are from one experiment (two replicates) and are
representative of four independent experiments. Bars, S.E.
purified and examined before and after adherence to glass. By
IFM, 5-LO was cytosolic in freshly purified human blood PMNs
maintained in suspension in Teflon cups (Fig. 7A), as expected.
After adherence to glass for 1 h, 5-LO was predominantly
intranuclear (Fig. 7C). A shift of 5-LO into the nucleus was
evident as early as 15 min after adherence (Fig. 7B). Intranuclear localization of 5-LO was also prominent in PMNs adhered
for 1 h to albumin-coated glass or plastic (data not shown).
When adherent PMNs were detached by trypsinization, fractionated, and analyzed by immunoblotting, abundant 5-LO was
associated with the nuclear soluble fraction, whereas 5-LO was
solely cytosolic in PMNs in suspension (Fig. 7C). Thus, adherence alone is sufficient to induce the redistribution of 5-LO
from the cytosol into the nucleus.
DISCUSSION
It is well established that upon activation, 5-LO undergoes
rapid translocation to the nuclear envelope. However, the subcellular distribution of 5-LO in resting cells has previously been
viewed as static, although it was known to vary among cell
types (27). The principle finding of this study is that the distribution of 5-LO can be dynamic, even in the resting cell:
unstimulated PMNs in the peripheral blood have none of the
enzyme associated with the nucleus, whereas elicited or adherent PMNs have abundant 5-LO within the nucleus. It is important to distinguish between the import into the nucleoplasm
seen here and the translocation to the nuclear envelope associated with activation. The former, unlike the latter, was not
associated with LT synthesis.
This study also demonstrated several important aspects of
the shift in 5-LO distribution: 1) this redistribution can be
relatively rapid. 2) The redistribution of 5-LO into the nucleus
is a general feature of PMN recruitment to sites of inflammation: it can occur during elicitation into different sites, in response to different agents, and in both rats and humans. 3)
5-LO within the nucleus, like 5-LO in the cytosol, will translo-
FIG. 7. Redistribution of 5-LO from the cytosol to the nucleus
in human PMNs after adherence. Human peripheral blood PMNs
were fixed and stained for 5-LO either after 1 h suspension in Teflon
cups (A) or at 15 min (B) or 1 h (C) after adherence to glass. Images are
representative of at least five independent experiments. D, immunoblot
analysis of cell fractions from adherent PMNs. After suspension or after
adherence for 1 h followed by release by trypsinization, cells were
fractionated into cytosol (C), nonnuclear membrane (M), nuclear soluble
protein (Ns), and nuclear pelletable protein (Np) fractions and probed
for 5-LO. Results are representative of three independent experiments.
cate to the nuclear envelope upon cell stimulation. 4) Elicited
PMNs, with predominantly intranuclear 5-LO, can not only
synthesize LTs but can generate significantly greater amounts
of LTs than blood PMNs, which have exclusively cytosolic 5-LO.
5) Nuclear sequestration seems to buffer the 5-LO pathway
from activation. 6) Adherence alone is sufficient to induce redistribution of 5-LO.
The actual trigger for import is not known. Although adherence seems to rapidly promote nuclear import, different types
of cell surfaces, adhesion molecules, and extracellular matrices
may have different effects on the import of 5-LO. These have
not been examined here. Also, the mechanism of 5-LO import
into the nucleus is not known. Large molecules require either a
nuclear localization sequence or a chaperone with a nuclear
localization sequence to traverse nuclear pores (28). The 5-LO
protein has a region that is rich in basic amino acids (amino
acids 650 – 657) that might serve as a nuclear localization sequence. Furthermore, the dynamic nature of 5-LO import suggests that mechanisms for regulating 5-LO redistribution must
exist. Likely mechanisms include direct modification [e.g.,
phosphorylation, as has recently been described (29)] of the
5-LO protein itself or a change in an unidentified accessory
molecule, such as an import-chaperone protein or a cytosolic
tethering protein. These possibilities await clarification.
The functional importance of the 5-LO redistribution that we
have described is unclear, but it could have ramifications for
enzyme activation. 5-LO activation is calcium-dependent (30).
In some cell types, changes in intranuclear calcium levels seem
to be independent of changes in cytosolic calcium (31), whereas
in other cell types, the two apparently are linked (32). Theoretically, an agonist that increased calcium exclusively in the
cytosol would activate 5-LO (and initiate LT synthesis) in blood
PMNs but not in elicited PMNs. In such a scenario, the sequestration of 5-LO within the nucleus would thereby serve to
decrease the susceptibility of the enzyme to activation. Consistent with this possibility, our data using ionophore demonstrated that more agonist was needed to drive LT synthesis in
elicited PMNs than in blood PMNs. A similar pattern exists
among macrophages: the threshold for initiation of LT synthesis is higher in alveolar macrophages, which have intranuclear
5-LO (6), than in peritoneal macrophages (24), which have
cytosolic 5-LO (5). However, once triggered, the level of LTB4
production by nuclear 5-LO in elicited PMNs greatly exceeds
that produced by cytosolic 5-LO in blood PMNs. It is interesting
that just as alveolar macrophages have a higher activation
threshold than peritoneal macrophages, they also have a
greater maximal LTB4 synthetic capacity (24). The identical
Downloaded from www.jbc.org at Michigan State University, on January 18, 2012
FIG. 6. Intranuclear 5-LO in elicited PMNs shortly after elicitation. Peritoneal cells were harvested by lavage at either 1 (A) or 2 (B)
h after glycogen instillation and enriched for PMNs by rapid differential
centrifugation. Cells were mounted and fixed immediately and then
probed for 5-LO. Images are representative of two independent
experiments.
8279
8280
Nuclear Import of 5-LO in PMNs
correlates the import of 5-LO into the nucleus with an increase
in cellular LT synthetic capacity. The molecular mechanisms of
both 5-LO redistribution and altered LT synthetic capacity
remain to be determined.
REFERENCES
1. Bortuzzo, C., Hanif, R., Kashfi, K., Staiano-Coico, L., Shiff, S. J., and Rigas, B.
(1996) Biochim. Biophys. Acta 1300, 240 –246
2. Drazen, J. M., Lilly, C. M., Sperling, R., Rubin, P., and Israel, E. (1994) Adv.
Prostaglandin Thromboxane Leukotriene Res. 22, 251–262
3. Rouzer, C. A., Matsumoto, T., and Samuelsson, B. (1986) Proc. Natl. Acad. Sci.
U. S. A. 83, 857– 861
4. Brock, T. G., Paine, R., III, and Peters-Golden, M. (1994) J. Biol. Chem. 269,
22059 –22066
5. Peters-Golden, M., and McNish, R. (1993) Biochem. Biophys. Res. Commun.
196, 147–153
6. Woods, J., Coffey, M., Brock, T., Singer, I., and Peters-Golden, M. (1995)
J. Clin. Invest. 95, 2035–2040
7. Chen, X.-S., Naumann, T. A., Kurre, U., Jenkins, N. A., Copeland, N. G., and
Funk, C. D. (1995) J. Biol. Chem. 270, 17993–17999
8. Woods, J., Evans, J., Ethier, D., Scott, S., Vickers, P., Hearn, L., Charleson, S.,
Heibein, J., and Singer, I. (1993) J. Exp. Med. 178, 1935–1946
9. Brock, T. G., McNish, R. W., and Peters-Golden, M. (1995) J. Biol. Chem. 270,
21652–21658
10. Abramovitz, M., Wong, E., Cox, M., Richardson, C., Li, C., and Vickers, P.
(1993) Eur. J. Biochem. 215, 105–111
11. Pueringer, R., Bahns, C., and Hunninghake, G. (1992) J. Appl. Physiol. 73,
781–786
12. Stankova, J., Rola-Pleszczynski, M., and Dubois, C. (1995) Blood 85,
3719 –3726
13. Coffey, M. J., Wilcoxen, S. E., and Peters-Golden, M. (1994) Am. J. Respir. Cell
Mol. Biol. 11, 153–158
14. Pouliot, M., McDonald, P., Borgeat, P., and McColl, S. (1994) J. Exp. Med. 179,
1225–1232
15. Brock, T. G., McNish, R. W., Coffey, M. J., Ojo, T. C., Phare, S. M., and
Peters-Golden, M. (1996) J. Immunol. 156, 2522–2527
16. Nakamura, T., Lin, L., Kharbanda, S., Knopf, J., and Kufe, D. (1992) EMBO J.
11, 4917– 4922
17. Petersen, M., Steadman, R., and Willieams, J. (1992) Immunology 75, 275–280
18. Suzuki, K., Yamamoto, T., Sato, A., Murayama, T., Amitani, R., Yamamoto, K.,
and Kuze, F. (1993) Am. J. Respir. Cell Mol. Biol. 8, 500 –508
19. Steinhilber, D., Hoshiko, S., Grunewald, J., Radmark, O., and Samuelsson, B.
(1993) Biochim. Biophys. Acta 1178, 1– 8
20. Powell, W. S., and Gravelle, F. (1990) J. Biol. Chem. 265, 9131–9139
21. Coffey, M., Peters-Golden, M., Fantone, J. C., III, and Sporn, P. H. S. (1992)
J. Biol. Chem. 267, 570 –576
22. Tithof, P., Schiamberg, E., Peters-Golden, M., and Garney, P. (1996) Environ.
Health Perspect. 104, 52–58
23. Phan, S., and Kunkel, S. (1986) Am. J. Pathol. 124, 343–352
24. Peters-Golden, M., McNish, R. W., Hyzy, R., Shelly, C., and Toews, G. B. (1990)
J. Immunol. 144, 263–270
25. Balter, M., Toews, G., and Peters-Golden, M. (1989) J. Immunol. 142, 602– 608
26. Wilborn, J., Bailie, M., Coffey, M., Burdick, M., Strieter, R., and PetersGolden, M. (1996) J. Clin. Invest. 97, 1827–1836
27. Serhan, C., Haeggston, J., and Leslie, C. (1996) FASEB J. 10, 1147–1158
28. Silver, P. (1991) Cell 64, 489 – 497
29. Lepley, R. A., Muskardin, D. T., and Fitzpatrick, F. A. (1996) J. Biol. Chem.
271, 6179 – 6184
30. Schatz-Munding, M., Hatzelmann, A., and Ullrich, V. (1991) Eur. J. Biochem.
197, 487– 493
31. Minamikawa, T., Takahashi, A., and Fujita, S. (1995) Cell Calcium 17,
167–176
32. Munzenmaier, D., and Greene, A. (1995) Am. J. Physiol. 269, H565–H570
33. Sierra-Honigmann, M., Bradley, J., and Pober, J. (1996) Lab. Invest. 74,
684 – 695
34. Devchand, P. R., Keller, H., Peters, J. M., Vazquez, M., Gonzalez, F. J., and
Wahli, W. (1996) Nature 384, 39 – 43
35. Lepley, R. A., and Fitzpatrick, F. A. (1994) J. Biol. Chem. 269, 24163–24168
Downloaded from www.jbc.org at Michigan State University, on January 18, 2012
patterns of 5-LO distribution and action in macrophages and
PMNs suggest a causal relationship between changes in 5-LO
localization and altered AA metabolism. Although the basis for
the greater maximal LT synthetic capacity in elicited PMNs
and alveolar macrophages is not known, these results underscore the importance of intranuclear 5-LO as an active pool in
LT generation.
Within the activated cell, cytosolic 5-LO will move to the
outer membrane of the nuclear envelope, whereas intranuclear
5-LO will move to the inner membrane. These sites of 5-LO
action are spatially distinct, being separated by a lumen. The
function of activated 5-LO may therefore be influenced by its
topographic proximity to substrate and cofactors, as well as
other enzymes that act either upstream or downstream from it.
In this manner, the disparate distributions of 5-LO in resting
and elicited PMNs would also be expected to have further
ramifications on molecular aspects of enzyme action. Furthermore, the subcellular localization of these other elements may
also be dynamic rather than static. For example, the “cytosolic”
phospholipase A2 is indeed cytosolic in quiescent endothelial
cells but becomes intranuclear when the same cells are growing
rapidly (33). Thus, the element of dynamic redistribution adds
a new level of complexity to the process of AA liberation and
5-LO action.
The changeable compartmentation of 5-LO suggests novel
roles for 5-LO products and also for 5-LO itself. The rapid
redistribution of 5-LO may serve to regulate the site of 5-LO
enzymatic action. This would be particularly interesting if
5-LO products acted at the immediate site of synthesis. For
example, if the products of cytosolic 5-LO affected cytosolic
processes (e.g., cytoskeletal functioning), and the products of
nuclear 5-LO affected nuclear processes (e.g., transcription),
then 5-LO compartmentation could be a way to regulate cell
function. Such localized effects could be achieved by much
lower concentrations of LTs than are needed for extracellular
release and paracrine action. These effects might be mediated
by soluble intracellular receptors like the peroxisome proliferator-activated receptor a, which can bind LTB4 in the nucleus
(34). Alternatively, the 5-LO protein itself has sites for interacting with other proteins, including a Src homology 3-binding
domain (35). Such domains mediate the interaction of 5-LO
with other proteins (35), suggesting the possibility of function(s) for 5-LO independent of its enzymatic activity. The
rapid redistribution of 5-LO, then, could serve as a mechanism
for regulating the accessibility of 5-LO to other proteins and
thus controlling such (hypothetical) effects.
The potential for rapid movement of 5-LO between the cytosol and the nucleus in a single cell type, independent of
activation, indicates a level of complexity in 5-LO action that
has heretofore been unappreciated. Furthermore, this study