From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
Posttranslational Processing of Defensins in Immature Human Myeloid Cells
By Erika V. Valore and Tomas Ganz
Human neutrophil promyelocytes synthesize, process, and
package several microbicidal proteins, including the highly
abundant defensins, into their azurophil granules. As deduced from their cDNA sequences, defensins are initially
synthesized as 94 amino acid (aa) precursors that must
undergo extensive processing.We performed metabolic labeling studies of defensin synthesis in the human promyelocytic
cell line HL-60 and in chronic myeloid leukemia cells, and
showed that preprodefensins are processed to mature 29 to
30 aa defensins over 4 to 24 hours via two major intermediates: a 75 aa prodefensin generated by the cleavage of the
signal sequence, and a 56 aa prodefensin that results from a
subsequent preaspartate proteolytic cleavage. Almost all of
the 75 aa form was found in the cytoplasmic/microsomal
fraction, whereas the 56 aa prodefensin and mature defensins predominatedin the granule-enrichedfraction. The 75 aa
prodefensin was also selectively released into the culture
supernatant. Treatment of HL-60 cells with monensin, chloroquin, or ammonium chloride, substances that neutralize
acidic subcellular compartments, partially blocked conversion of the 75 aa prodefensin into 56 aa prodefensin, but did
not increase the extracellular release of the 75 aa form.
Further studies will be required to determine the role of this
processing pathway in subcellular targeting to azurophil
granules and avoidance of autocytotoxicity.
0 1992 by The American Society of Hematology.
D
of N-linked oligosaccharide side chains that can interact
with the receptor system responsible for the targeting of
lysosomal enzymes. Defensins thus present a compelling
model for studying the processing and targeting of myeloid
granule proteins.
EFENSINS ARE 29 to 34 amino acid (aa) cysteineand arginine-rich peptides that have a broad spectrum of antimicrobial and cytotoxic properties in vitro.’ In
humans, the three major defensins, human neutrophil
peptides HNP-1 through -3, constitute more than 5% of the
total cellular protein in mature neutrophils, and at least
30% of the protein of their azurophil granules.’ Because
HNP-2 (29 aa), the smallest human defensin, differs from
HNP-1 and HNP-3 only by the absence of an aminoterminal alanine or aspartate respectively, the known human defensin cDNAs for HNP-1 and HNP-3 may in fact
encode all three major defensin peptides? The 94 aa
putative defensin precursors, deduced from the cDNAs,
contain a typical 19 aa signal sequence and a 45 aa propiece.
The 30 aa mature defensin peptide comprises the entire
carboxyl terminus of the preprodefensin?
The processing of myeloid granule proteins myeloperoxidase, elastase, cathepsin G, and lactoferrin has been
characterized in myeloid cell lines and bone marrow.””
Despite the common belief that azurophil granules are
homologous to lysosomes of nonmyeloid cells, there is
increasing evidence that azurophil granule proteins are
processed by pathways distinct from those responsible for
the processing of lysosomal enzymes in nonmyeloid cellsA6
Unlike other myeloid granule proteins, defensins and their
deduced precursors lack any consensus site for attachment
From the Will Rogers Institute Pulmonary Research Laboratory and
Department of Medicine, School of Medicine, University of California,
Los Angeles.
Submitted August 19, 1991; accepted October 15, 1991.
Supported in part by Grants No. HL-35640 and AI-22839from the
National Institutes of Health. Protein sequencing was peflomwd at the
UCLA Protein Microsequencing Facility, which is supported in pari by
a BRS Shared Instrumentation Grant lSlORR05554 from the National Institutes of Health.
Address reprint requests to Tomas Ganz, PhD, MD, UCLA School
of Medicine, Department of Medicine, The Center for the Health
Science, 37-055, Los Angeles, CA 90024-1736.
The publication costs of this article were defiayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C.section 1734 solely to
indicate this fact.
0 1992 by The American Society of Hematology.
0006-4971/92/ 7906-0016$3.00/0
1538
MATERIALS AND METHODS
Materialr. Chloroquin and monensin were from Sigma Chemical Company (St Louis, MO) and ammonium chloride was from
Fisher Scientific (Tustin, CA). Prestained electrophoresis molecular weight standards were from Bethesda Research Laboratories
(Bethesda, MD). Dog pancreas microsomes” were a generous gift
from Dr David Meyer (University of California, Los Angeles).
Labeling of cells. HL-60 cells (passage 4 to 8 from stock CCL
240; American Type Culture Collection, Rockville, MD) were
maintained in RPMI 1640 (GIBCO-BRL, Grand Island, NY)
supplemented with 10% heat-inactivated fetal calf serum (FCS), 2
mmol/L L-glutamine, and 50 mg/L gentamicin at 37°C in humidified 5% CO,. Before metabolic labeling, 2 x lo6 cells/mL were
incubated in cysteine-free RPMI 1640 (Selectamine; GIBCOBRL) with 5% FCS for 60 minutes to allow depletion of the
intracellular cysteine pool. Cells were labeled by incubation with 50
kCi/mL ”S-cysteine (1,OOO Ci/mmol; NEN-Dupont, Boston, MA).
In “chase” experiments, the cells were washed after 1 hour of
labeling, suspended in RPMI 1640with 10% FCS at a density of lo6
cells/mL, and then incubated for the indicated chase times. When
used, the lysosomal acidification inhibitors were added 30 minutes
before addition of ”S-cysteine.
Chronic myelogenous leukemia (CML) cells were purified from
blood samples obtained from consenting patients. The cells were
sedimented in 3% dextran/normal saline, and depleted of residual
red blood cells by brief hypotonic lysis. Metabolic labeling of CML
cells with ”S-cysteine was performed for 24 hours as described for
HL-60 cells. Labeling with ’H-leucine (156 Ci/mmol; NENDuPont) was performed after 1 hour of leucine starvation, by
incubation with 10 pCi/mL of 3H-leucine in leucine-free RPMI
1640/5% FCS (Selectamine; BRL-GIBCO) for 24 hours.
In vitro translation of CML messenger RNA (“A).
Total and
polyA-selected mRNA from CML cells was purified as previously
described.” Either 10 kg of total RNA or 5 pg polyA-selected
mRNA was mixed with wheat germ extract (Promega, Madison,
WI) and 25 pCi ”S-cysteine in 50 kL total volume. For cotranslational processing experiments, 2 ILLdog pancreas microsomes (60
OD,,/mL) were also included in the reaction. The translation mix
was incubated at 25°C for 1hour and then subjected to immunoprecipitation as described below.
Subcellular fiactionntion. Labeled CML cells (2.5 X 10’) were
suspended in 1 mL 0.34 mol/L sucrose (pH 7.4), and disrupted by
Blood, Vol79, No6 (March 15). 1992: pp 1538-1544
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
POSTTRANSLATIONAL PROCESSING OF DEFENSINS
two 10-second bursts with a probe sonicator. Cell lysis was
monitored by light microscopy. Nuclei were collected by centrifugation at 20% for 10 minutes. The granule-rich fraction was pelleted
from the supernatant by centrifugation at 16,000g for 30 minutes at
4"C, leaving a supernatant enriched in microsomes and cytosol.
Immunoprecipitation of radioactive defensin peptides. Labeled
cells were extracted with radioimmunoprecipitation assay (RIPA)
buffer (100 pL/106 cells) containing 0.15 mol/L NaCl, 30 mmol/L
HEPES pH 7.3, 1% (vol/vol) Triton X-100, 1% (wt/vol) sodium
desoxycholate, 0.1% (wtlvol) sodium dodecyl sulfate (SDS), and 1
mmol/L phenylmethanesulphonylfluoride (Sigma; added just before use). Extracts were incubated on ice for l hour, centrifuged at
16,OOOg for 30 minutes at 4"C, and the supernatant was stored
frozen until used for immunoprecipitation. Subcellular fractions
were lysed by the addition of ?hvolume of a threefold concentrated
solution of RIPA buffer. Samples were incubated on ice for 30
minutes and then stored at -20°C until used for immunoprecipitation.
Cell extracts (300 pL) and culture supernatants (3 mL) were
mixed with 30 pL of antidefensin rabbit semm.I4After incubation
on ice for 1 hour, 60 pL of protein A-agarose (Biorad, Richmond,
CA; 1:l slurry with RIPA buffer) was added and the mixture
rotated at 4°C for 1hour. The immunoprecipitate was washed five
times in RIPA buffer and three times in phosphate-buffered saline
(PBS). The pellet was resuspended in 30 to 40 pL electrophoresis
sample buffer (0.17 mol/L Tris-HC1, pH 8.8, 21% [vol/vol] glycerol, 6% [wt/vol] SDS, 120 mmol/L dithiothreitol). Samples were
boiled for 10 minutes, briefly centrifuged, and the supernatant was
analyzed by electrophoresis.
SDS-Tricinepolyacrylamide gel electrophoresis (PAGE) and P o rography. SDS-Tricine PAGE" was performed on 0.75"
thick
minigels containing 16.5% acrylamide in the running layer and 4%
polyacrylamide in the stacking layer (Hoefer Scientific Instruments, San Francisco, CA). Electrophoresis was conducted at 30
mA/gel without cooling. Gels were stained with Coomassie blue,
destained in 25% methanol/0.4% formaldehyde, washed in water
for 10 minutes, and treated with 1 mol/L sodium salicylate/2%
glycerol for 20 minutes. After drying, the gels were exposed to
Kodak XAR 5 X-ray film (Eastman Kodak, Rochester, NY) at
-80°C.
For Edman degradation analysis, samples were subjected to
electrophoresis as above and then electroblotted to polyvinylidene
difluoride (PVDF; Biorad) membranes. Transfer time (Transblot;
Biorad) was 2 hours in 0.05 mol/L sodium borate, pH 9, 20%
methanol, 0.05% SDS at 20 V (0.2 A). The membranes were
exposed to Kodak XAR 5 film to identify the appropriate bands for
radioactive amino acid sequencing.
Amino acid sequence alignment. To characterize the dkfensin
precursors, "S-cysteine- or 'H-leucine-labeled proteins bound to
PVDF membranes were subjected to Edman degradation on an
AB1 475 (Applied Biosystems, Foster City, CA) automated amino
acid sequencer. Fractions were collected directly from the cartridge and subjected to liquid scintillation counting. The radioactivity (cpm) in each cycle versus the cycle number were plotted, and
the peaks were aligned with the known cysteines or leucines in the
preprodefensin amino acid sequence.'
RESULTS
Biosynthesis. Although mature polymorphonuclear leukocytes (PMN) contain about 5 Fg of defensins HNP-1
through -3 per lo6 cells, they do not actively synthesize
these proteins. Accordingly, biosynthetic studies were performed in HL-60 cells or CML cells. In acid-urea polyacrylamide gels or their Western blots, the defensins produced
1539
by HL-60 cells comigrated with those found in human PMN
(data not shown). As judged by microbicidal assays and
analysis in acid-urea PAGE or SDS-PAGE, mature defensins prepared from leukophoresed CML cells were identical to those prepared from mature PMN (unpublished
data). Because cysteines represent 6 of the 29 or 30 aa
residues in human defensins,'6 "S-cysteine was used to
metabolically label defensin precursors. The 94 aa preprodefensins' deduced from the cDNA sequence do not
contain additional cysteines outside the mature defensin
region. Consequently, the intensity of labeling of each
defensin precursor should be proportional to its relative
abundance in the extract, and independent of the size of the
precursor.
An autoradiogram of a representative pulse chase experiment in HL-60 cells is shown in Fig 1.The upper and lower
panels show labeled defensin proteins immunoprecipitated
from cell lysates and the culture medium, respectively. In
cells labeled for 24 hours (lane EQ), three major protein
species could be identified, with apparent molecular weights
(MW) of 7.1 Kd, 4.9 Kd, and 3.1 Kd. The earliest form (7.1
Kd, form 1) was seen after 15 to 30 minutes of chase. In
some experiments, this band appeared to be a closely
spaced doublet at 15 and 30 minutes. Between 1 and 4
hours, form 1was gradually converted to a 4.9-Kd protein
(form 2). From 4 to 24 hours, most of the form 2 precursor
was cleaved to the 3.1-Kd species (form 3) that comigrated
with mature defensins HNP-1 through -3. As shown in the
lower panel, a protein with the same MW as form 1
accumulated in the medium. A second extracellular protein
with an approximate MW of 15 Kd could also be seen in
some experiments (see 4h and E Q lanes), but was not
further investigated in these studies.
When we compared labeled forms 1through 3 biosynthesized by myeloid cells to the immunoprecipitated products
of in vitro translation of CML mRNA, the translated
product migrated with an apparent MW somewhat higher
than form 1 (data not shown). If dog pancreas microsomes
were present during in vitro translation, a second band
appeared that comigrated with biosynthetically labeled
form 1. These observations are consistent with the initial
synthesis of defensins as 94 aa preproproteins that are
rapidly processed to a lower MW precursor form 1. Further
proteolytic events occur over a 24-hour period to give form
2 (4.9 Kd) and finally the 3.1-Kd mature defensin peptide
(form 3).
Defensin precursor identification. In preliminary experiments in CML cells, we found that they contained the same
pattern of defensin precursor forms as HL-60 cells (data
not shown) but labeled much more intensely. To determine
the exact cleavage pattern that generates defensin precursors, CML cells were allowed to incorporate 3H-leucine for
24 hours. Preprodefensins contain 9 leucine residues: 5 in
the putative signal sequence, 3 in the propiece, and 1in the
mature defensin sequence. Alternatively, the cells were
labeled with "S-cysteine for 24 hours as before. The
immunoprecipitated proteins were blotted to PVDF membranes and subjected to 30 cycles of N-terminal Edman
degradation. Each fraction was subjected to scintillation
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
VALORE AND GANZ
1540
15' 30' I h 2h
4h
7h 24h EQ
3-
7.1 kD
4.9 kD
3.1 kD
S-c
7.1 kD
I*
2*
Fig 1. Pulse-chase study of defensin processing in HL-60 cells. Each aliquot of cells was labeled with %-cysteine for 1 hour and subjected to
cold chase for the time indicated. Lane EQ was labeled for 24 hours without cold chase. The lysates of cells (top panel) or the corresponding
culture supernatants (bottom panel)were immunoprecipitatedwith antidefensin serum and analyzed by Tricine SDS-PAGE.The autoradiograms
of the gels are shown with time of cold chase indicated above each lane (15 minutes to 24 hours). The MW of the three major cellular forms 1,2,
and 3, and the major culture supernatant form S were calculated from their electrophoretic mobilitiesrelative to MW standards.
counting and the radioactive peaks were aligned with the
leucines or cysteines in the known preprodefensin amino
acid sequence (Fig 2). Forms 2 and 3 could be aligned
unequivocally (Fig 2C through E) because at least two
distinct peaks with characteristic spacing were detected in
each case. Only one large leucine peak was detected in form
S (Fig 2A), and the proposed alignment is based on the
assumption that any additional leucine before the tenth
N-terminal residue would be detectable if it were present.
With an estimated repetitive Edman degradation yield of
0.9, this is a realistic assumption. Form 1, which comigrated
with form S in SDS-PAGE,yielded the same peak as form S
but also contained two additional small peaks, presumably
due to contamination with small amounts of form 2. Based
on these observations, we conclude that the form 1 precursor is a 75 aa proprotein that is formed after cleavage of the
signal sequence. The radiosequencing confirmed that the
same form 1 precursor was also released into the culture
medium as the predominant defensin species. Form 2 is a
56 aa product resulting from cleavage between an alanine
and an aspartic acid residue (residues 38 and 39). The
mature defensin peptides HNP-1, -2, and -3 correspond to
form 3. Given the approximate nature of protein MW
derived from electrophoretic mobility in SDS-PAGE, there
is reasonable agreement between the MW calculated from
the proposed amino acid sequences of the three forms and
the MW derived from their electrophoreticmobilities (form
1: 8.2 Kd calculated, 7.1 Kd electrophoretic; form 2: 6.3 Kd
calculated, 4.9 Kd electrophoretic; form 3: 3.4 Kd calculated, 3.1 Kd electrophoretic).
Subcellular fractionation. After labeling CML cells for
24 hours with "S-cysteine, crude nuclear, cytoplasmic/
microsomal, and granule fractions were isolated as described in Materials and Methods, and were subjected to
immunoprecipitation.As seen in Fig 3, the nuclear fraction
was devoid of defensin precursors. Form 1 was found only
in the conditioned medium and the cytoplasmic/microsomal fraction. Forms 2 and 3 were detected predominantly in
the granule-enriched fraction but were present also in the
cytoplasmic/microsomal fraction. Although the results suggest that the sequential proteolytic cleavages may take
place in distinct compartments, more detailed studies will
be required to ascertain the subcellular location of these
processing steps precisely.
Effects of monensin and weak bases on processing. To
define the role of acid compartments in the processing of
defensin precursors, pulse chase experiments in HL-60 cells
were performed in the presence of monensin and the weak
bases chloroquin and ammonium chloride. The ionophore
monensin exchanges protons for Na' ions leading to an
increase in the pH of acidic organelle^.^' Figure 4 shows the
effects of 0.5 pmol/L and 1 pmol/L monensin on defensin
processing. Cells were labeled for 1 hour and chased for 1,
7, and 24 hours. Another set of cells was labeled for 24
hours with no chase (lane E). After 7 hours of chase, there
was accumulation of form 1 in monensin-treated cells
compared with the control cells. Form 2 wasvirtually absent
from the monensin-treated 24-hour labeled cells. These
data suggest that monensin inhibits the conversion of form
1 to form 2, but does not interfere with the conversion of
form 2 to the mature defensin peptide. Chloroquin and
ammonium chloride, weak bases that act by a different
mechanism to increase the pH of acid compartments," had
a similar effect to that of monensin (Fig 5). Cells cultured in
the presence of 100 pmol/L chloroquin or 50 mmol/L
ammonium chloride also accumulated form 1 at all chase
timcs. A similar effcct was sccn with cclls culturcd in 10
mmol/L ammonium chloride (data not shown). However,
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
POSTTRANSLATIONAL PROCESSING OF DEFENSINS
1541
A
60
70
c
B
FORM 1
FORM S
2 50
a
0 40
30
20
10 BBEPLQARADEVAAAPEQIAADIPEVVVSLAW
PRO-HNP (75aa)
80 r
701 K
60
Fig 2. Alignment of defensin
precursor forms 1, 2, 3, and S
with the amino acid sequence of
preprodefensin. CML cells were
labeled for 24 hours with 'Hleucine (A through D) or 'Scysteine [E). Precursor forms
were purified from cell lysates (B
through E) or culture supernatant (A) by immunoprecipitation,
Tricine SDS-PAGE, and blotting
to PVDF membrane. The individual bands were cut out from the
membrane and subjected to Edman degradation and scintillation counting of each cycle. Each
plot displays the cpm on the vertical axis and cycle number on
the horizontal axis. The first two
cycles are blank (B, wash without Edman reagents). Subsequent cycles are labeled with the
one letter amino acid code of the
optimal alignment with the preprodefensin sequence. Arrows
show the positions of labeled
amino acids in the proposed
alignment.
C
FORM 2
PRO-HNP (56aa)
MATURE HNP-2
BBCYCRIPACIAGERRYGTCIYQGRLWAFCC
MATURE HNP-2
BBCYCRIPACIAGERRYGTCIYQGRLWAFCC
35
30
h
25
20
15
l o BBACYCR
I PAC I AGERRYGTC IYQGRLWAFCC
D
MATURE HNP-1/3
depletion of form 2 in 24-hour labeled cells was much more
prominent with ammonium chloride than with chloroquin.
These three inhibitors had no detectable effect on the
release of the 75 aa prodefensin into the culture medium
(data not shown). The pH of the culture media was not
significantly affected by the addition of ammonium chloride
or chloroquin. Taken together, these data suggest that
processing of form 1to form 2 involves a pH-sensitive step,
2o BBACYCR IPAC IACERRYGTC I YCGRLWAFCC
D
MATURE HNP-1/3
whereas the conversion of form 2 to form 3 is resistant to
neutralization of acid compartments.
DISCUSSION
The azurophil granules of PMN encompass a variety of
proteolytic, membrane-active, and cytotoxic proteins that
participate in the destruction of microbes in the phagolysosome and in tissue injury during inflammation." These
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
1542
VALORE AND GANZ
N
C
G
M
1,
2,
3,
Fig 3. Subcellular fractionation of CML cells labeled for 24 hours
with %cysteine. The cells were disrupted by sonication in 0.34
mol/L sucrose, pH 7.4, and subjected t o differential centrifugation t o
collect nuclei (fraction N, 2OOg sediment), granule-enriched fraction
(fraction G, 16,OOOg sediment), and cytoplasmic/microsomal fraction
(fraction C, 16,OOOg supernatant). For comparison, an equivalent
amount of the CML culture medium (M) was also analyzed. Each
fraction (5 x lo6cell equivalents) was subjected t o lysis, immunoprecipitation, and analyzed by Tricine SDS-PAGE. An autoradiogram of
the gel is shown, with the fraction indicated above each lane. Forms 1,
2, and 3 correspond to those shown in Fig 1.
proteins, synthesized during the promyelocyte stage of
neutrophil maturation in the bone marrow, must undergo
posttranslational processing and transport to azurophil
granules without incurring autocytotoxicity. Although similar constraints also govern the synthesis and intracellular
transport of lysosomal enzymes in nonmyeloid cells, there is
increasing evidence that the processing and intracellular
targeting of lysosomal enzymes versus azurophil granule
proteins involves separate and distinct mechanisms. Lysosomal enzymes'' undergo N-linked glycosylation in the endoplasmic reticulum (ER), followed by modification of the
oligosaccharide side chains in the ER and Golgi. The
resulting mannose-6-phosphate-containing oligosaccharides are recognized by specific receptors that mediate the
l h 7h 24h E
intracellular transport and targeting of lysosomal enzymes.
Upon reaching the acid pH of (pre)lysosomal compartments, the receptor-enzyme complexes dissociate and the
receptors are recycled. Interventions that interfere with
glycosylation, oligosaccharide processing, or the maintenance of acid pH in these compartments result in incomplete processing and intracellular accumulation or extracellular release of proenzymes. Like lysosomal enzymes,
myeloperoxidase (MPO) is N-glycosylated and normally
contains mannose-6-phosphate residues that can function
to mediate the endocytosis of the protein by an M6Preceptor-dependent pathway." Unlike the processing of
lysosomal enzymes in fibroblasts, MPO processing is only
modestly affected by agents that increase the pH of acid
compartments, and not affected by substances that inhibit
the sequential trimming of oligosaccharides to structures
recognized by the M6P receptors. Based on these observations, it appears that the M6P receptor is not required for
MPO processing or targeting: The case for M6P-independent processing of myeloid granule proteins is made even
more clearly by defensins because their precursors lack the
consensus sequence for N-glycosylation and are thus unlikely to form the oligosaccharide side chains required for
recognition by the M6P receptor.
We have shown that the processing of human defensins
involves the sequential cleavage of the 94 aa preprodefensin
to 75 aa prodefensin form 1, then 56 aa prodefensin form 2,
and subsequently to the 29 to 30 aa mature defensin (Fig 6).
However, the pulse-chase method preferentially detects the
major longer-lived intermediates, and the existence of
additional short-lived intermediates cannot be excluded.
Some of these potential intermediates are described by
Harwig et a1."
Of the three major sites of preprodefensin cleavage, the
signal sequence cleavage site that generates form 1 closely
follows the consensus for similar sequences?' The cleavage
that generates form 2 takes place on the amino side of an
aspartate residue, reminiscent of the preaspartate endopro-
l h 7h 24h E
l h 7h 24h E
monensin
control
1,
2,
3,
monensin
0.5 pM
1.O pM
Fig 4. The effect of monensin on defensin processing in HL-60 cells. The autoradiogram of a Tricine SDS-PAGE is shown. Each group of four
lanes (monensin 0.5 Fmol/L, monensin 1.0 pmol/L, and control) represents the immunoprecipitates of cell lysates from a single pulse-chase
study. The cells were preincubated for 30 minutes with the concentration of monensin indicated before labeling for 1 hour with "S-cysteine. In
control, the monensin was omitted and only its solvent added. Each aliquot of cells was then subjected t o cold chase for the times shown above
each lane (1.7, or 24 hours). Lanes E were labeled for 24 hours without cold chase.
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
POSTTRANSLATIONAL PROCESSING OF DEFENSINS
1543
Fig 5. The effects of chloroquin and ammonium chloride on defensin processing in HL-60 cells. The autoradiogramof a Tricine SDS-PAGE is
shown. Each group of four lanes (chloroquin 100 KmollL. ammonium chloride 50 mmol/L, and control)represents the immunoprecipitates of cell
lysates from a single pulse-chase study. The cells were preincubated for 30 minutes with the concentration of chloroquin or ammonium chloride
indicated before labeling for 1 hour with %cysteine. In control, the chloroquin and ammonium chloride were omitted. Each aliquot of cells was
then subjected to cold chase for the times shown above each lane (1.7, or 24 hours). Lanes E were labeledfor 24 hours without cold chase.
tease activity detected in media that had been conditioned
by phytohemagglutinin-stimulated peripheral blood mononuclear cells.” Although we found form 1 (75 aa prodefensin) as the predominant prodefensin in fresh HL-60 conditioned media, we also noted that form 2 became detectable
after storage of conditioned media at -20°C for several
weeks (data not shown). The rccovery of a prodefensin
identical to our form 2 (56 aa prodefensin) from media
conditioned by HL-60 cells has been previously reported,:‘
and may be explained perhaps by the presence of preaspartate endoprotease activity released by HL-60 cells into the
media. The final maturation cleavage that generates form 3
(mature defensin) takes place in the vicinity of a dibasic
sequence with a nearby upstream monobasic residue, as has
It remains to
been described for prohormone proces~ing.~~
be seen whether defensins are also processed by one of the
members of the recently described kex family of dibasic
site-specific endoproteases that mediate prohormone processing.” The possibility that several sequential proteolytic
steps are involved in the conversion of form 2 to form 3 is
discussed in the accompanying manuscript.”
The effects of inhibitors of acidification on defensin
processing were similar to their effects on MPO processing.’” Incubation with these inhibitors resulted in a partial
Fig 6. The major proteolytic cleavage sites in preprodefensin
processing. Arrows indicate the sites of proteolytic cleavage. The
hydrophobic stretch of the signal sequence is underlined. The disulfide linkage pattern of the mature defensin protein is also shown.
block of conversion of prodefensin form 1 to prodefensin
form 2, suggesting that the activity of intracellular transporters or enzymes involved in this step takes place in an acid
compartmcnt and is somewhat affected by pH. However,
the influence of these inhibitors on defensin processing in
HL-60 cells was modest comparcd with their effects on the
sorting of lysosomal enzymes in fibroblasts.” Because the
ability of monensin, ammonium chloride, and chloroquin to
neutralize vacuolar pH in HL-60 cells is well documcntcd,“’
these observations suggest that defensin processing and
subcellular transport are not critically dependent on an
acid-dissociable carrier analogous to the mannosc-6phosphate receptor.
Why are preprodefensins processed in this relatively
complex manner? During defensin synthesis, processing,
and intracellular transport, the prepropiece and its fragments could subserve several functions that are no longer
required once the protein is packaged into the azurophil
granule. In its intact initial form the prepropiece may assist
in the folding of the defensin molecule into its correct
shape. probably by electrostatic interactions of the arginines of the mature defensin with the negatively charged
glutamate and aspartate residues in the propiece. After
removal of the signal sequence that functions to anchor the
nascent peptide in the membrane of the endoplasmic
reticulum, the anionicity of the propiece balances the
cationicity of the mature protein, and may act to neutralize
the latter’s cytotoxicitJ” or facilitate its transit through
organellar membranes. Alternatively, the propiece may
contain addressing information that sequentially directs the
nascent defensin into the appropriate subcellular compartments. Finally, the endoproteolytic fragments generated
during defensin proccssing may exert as yet unknown
intraccllular or extracellular functions. In the future. the
structure of preprodefensin can be subjected to selective
mutagenesis to test these conjectures.
In summary. the processing of defensins from the 94 aa
preprodefensins to mature 29 to 30 aa defcnsins takes place
over 4 to 24 hours and involves at least three endoproteolytic cleavages. In the process, two major transient
intermediates, a 75 and a 56 aa prodefensin, arc generated,
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
1544
VALORE AND GANZ
and the larger intermediate is released extracellularly.
Additional studies will be required to determine the subcellular location Of the processing steps, the nature Of the
enzymes involved, and the targeting mechanism responsible
for the selective delivery of defensins to azurophil granules.
ACKNOWLEDGMENT
We thank Dr Elizabeth Neufeld for several helpful discussions,
and Dr Robert I. Lehrer for his critical reading of the manuscript.
We acknowledge the technical assistance of Dr Audree Fowler and
the U C U Protein MicrosequencingFacility.
REFERENCES
1. Ganz T, Selsted ME, Lehrer RI: Defensins. Eur J Haematol
44:1,1990
2. Rice WG, Ganz T, Kinkade JMJ, Selsted ME, Lehrer RI,
Parmley R T Defensin-rich dense granules of human neutrophils.
Blood 70:757,1987
3. Daher m,Lehrer RI, Ganz T, &onenberg M: Isolation and
characterization of human defensin cDNA clones. Proc Natl Acad
Sci USA 857327,1988
4. Nauseef WM, Olsson I, Arnljots K Biosynthesis and processing of myeloperoxi,jase-A marker for myeloid cell differentiation.
Eur J Haematol4097,1988
5. 0lsson I, L~~~ M, penson AM, Arnljots K: Biosynthesis and
processing of ]actoferrin in bone mamow cells, a comparison with
processingof myeloperoxidase.Blood 7141,1988
6. Stromberg K, Persson AM, Olsson I: The processing and
intracellular transport of myeloperoxidase. Modulation by lysosomotropic agents and monensin. Eur J Cell Biol39:424,1986
7. Olsson I, Persson AM, Stromberg K Biosynthesis, transport
and processing of myeloperoxidasein the human leukaemic promyelocytic cell line HL-60 and normal marrow cells. Biochem J
223:911,1984
8. Lindmark A, Persson AM, Olsson I: Biosynthesis and processing of cathepsin G and neutrophil elastase in the leukemic myeloid
cell line U-937. Blood 76:2374,1990
9. Nauseef WM: Myeloperoxidase biosynthesis by a human
promyelocytic leukemia cell line: Insight into myeloperoxidase
deficiency. Blood 67:865,1986
10. Nauseef WM: Postranslational processing of a human myeloid lysosomal protein, myeloperoxidase.Blood 701143,1987
11.
A, Pohlmann R, Olsen RL, von Figura K Myeloperoxidase is synthesized as larger phosphorylated precursor. EMBO
J 3:2671,1984
12. Meyer DI, Dobberstein B: A membrane component w e n tial for vectorial translocation of nascent proteins across endoplasmic reticulum: Requirements for its extraction and reassociation
with the membrane. J Cell Biol87:498,1980
13. Sambrook J, Fritsch EF, Maniatis T Molecular Cloning (ed
2). Cold Spring Harbor, NY,Cold Spring Harbor Laboratory, 1989,
p 7.19
14. Ganz T, Selsted ME, Szklarek D, Hanvig SS, Daher K,
Bainton DF, Lehrer RI: Defensins. Natural peptide antibiotics of
human neutrophils. J Clin Invest 76:1427,1985
15. Schaegger H, von Jagow G: Tricine-sodium dodecyl sulfatePlYacrylamidegel electrophoresisfor the separation of Proteins in
the range from 1to 100 kDa. Anal Biochem 166:368,1987
16. Selsted ME, Harwig S S , Ganz T, Schilling JW, Lehrer RI:
Primary structures of three human neutrophil defensins. J Clin
Invest 76:1436,1985
17. Braulke T, Geuze HJ, Slot JW, Hasilik A, von Figura K: On
and processing
the effectsof weak bases and monensin On
of lysosomal enzymes in human cells. Eur J Cell Biol43:316,1987
18. Lehrer RI, Ganz T Review: Antimicrobial polypeptides of
human
76:21699 1990
19. Kornfeld S: Trafficking of lysosomal enzymes. FASEB J
1:462,1987
20. Harwig SSL, Park ASK, Lehrer RI: Characterization of
defensin precursors in mature human neutrophils. Blood 79:1532,
1992
21. von Heine G: The signal peptide. J Membr Biol 115:195,
1990
22. Black RA, Kronheim SR, Merriam JE, March CJ,Hopp Tp:
A pre-aspartate-specific protease from human leukocytes that
cleaves pro-interleukin-1. J Bioi Chem 2@:5323,1989
23. Bateman A,
A, Shustik c,Mars WM, Solomon s:ne
isolation and identification of multiple forms of the neutrophil
granule peptides from human leukemic cells. J Biol Chem 266:
7524,1991
24. Harris RB: Processing of pro-hormone precursor proteins.
Biochem Biophys 275:315, 1989
25. Fuller RS, Brake AJ, Thomer J: Intracellular targeting and
structural conservation of a prohormone-processing endoprotease.
Science 246:482,1989
26. Barker RL, Gleich GJ, Pease LR:Acidic precursor revealed
in human eosinophil granule major basic protein cDNA. J Exp
Med 168:1493,1988
From www.bloodjournal.org by guest on June 14, 2017. For personal use only.
1992 79: 1538-1544
Posttranslational processing of defensins in immature human myeloid
cells
EV Valore and T Ganz
Updated information and services can be found at:
http://www.bloodjournal.org/content/79/6/1538.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.