* Department of Hematology, Lund University, Sweden;
Inflammation Program and Department of Medicine, Veterans Administration Medical Center and University of Iowa, Iowa City; and
The Netherlands Cancer Institute, Amsterdam
Correspondence: Elinor Bülow, M.D., Dept. of Hematology, C14, BMC, S-221 84 Lund, Sweden. E-mail: elinor.bulow{at}hematologi.lu.se
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-microglobulin (1-m)
was targeted to storage organelles in 32D cells and colocalized
with the lysosomal marker LAMP-1, whereas soluble TNF receptor 1
(sTNFR1) was secreted without granule targeting. Fusion of MPOpro to
1-m delayed exit from endoplasmic reticulum (ER), but
subsequent targeting to dense organelles was indistinguishable from
that of 1-m alone. Fusion proteins between MPOpro and
sTNFR1 or green fluorescent protein expressed in myeloid 32D, K562, or
PLB-985 cells did not associate stably with calreticulin or calnexin,
molecular chaperones that normally interact transiently with the MPO
precursor, but were still efficiently retained in the ER followed by
degradation. We conclude that normally secreted, nonmyeloid proteins
can be targeted efficiently to storage organelles in myeloid cells,
that myeloid cells selectively target some proteins for storage but not
others, and that MPOpro may contribute to the prolonged ER retention of
the MPO precursor independent of the ER-molecular chaperones
calreticulin and calnexin.
Key Words: neutrophil • granule • secretory pathway • 1-microglobulin • secretion
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Optimal microbicidal activity of human PMNs requires the MPO-hydrogen peroxide-halide system. Upon stimulation, PMNs assemble and activate the reduced nicotinamide adenine dinucleotide phosphate (NADPH)-dependent oxidase, thus generating hydrogen peroxide from molecular oxygen, and degranulate, thereby releasing the azurophilic granule protein MPO into the phagolysosome. MPO catalyzes the formation of HOCl, which together with other highly reactive species generated within the phagosome constitutes a potent cytocidal system. Synthesized during the promyelocytic stage of myeloid development, the MPO precursor apoproMPO interacts transiently and reversibly with molecular chaperones in the endoplasmic reticulum (ER), including calreticulin (CRT) and calnexin (CLN), and subsequent incorporation of heme generates the enzymatically active 90-kDa proMPO (reviewed in [3 , 6 7 8 ]). Heme acquisition is required for proteolytic maturation and export from the ER, indicating that heme incorporation induces conformational changes necessary for successful maturation and intracellular targeting. Although a minor portion of newly synthesized proMPO is secreted constitutively from the cell, most of the precursor undergoes dimerization, modification of oligosaccharide sidechains, and proteolytic processing with final compartmentalization in the azurophilic granule. Among these proteolytic events is the cleavage of the amino terminal 125 amino acid propeptide [9 , 10 ], a modification necessary for the normal maturation of the precursor to the native form of MPO.
To investigate further the specificity of sorting for storage in myeloid cells, we stably expressed the genes for several nonmyeloid proteins in several myeloid cell lines. In addition, we assessed a possible role of the amino-terminal MPO propeptide in targeting by examining the impact of propeptide deletion and the intracellular trafficking of chimeric constructs containing the MPO propeptide (MPOpro) coupled with the amino terminus of nonmyeloid proteins. Our results indicate that normally secreted, nonmyeloid proteins can be targeted for storage granules of myeloid cells. We also observed that the chimeric proteins were retained in the ER to a greater degree than were native proteins. Although the ER retention observed may also result from slow or abnormal folding of the chimeric proteins, the propeptide may play an important role in the long retention time normally observed in MPO precursors in the ER. Prolonged retention may be required for the additional time needed for heme acquisition, a modification necessary for exit from the ER and functional maturation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-microglobulin
(1-m) was a gift from Dr. Bo Åkerström (Lund,
Sweden). A rabbit antiserum directed against the amino-terminal
propeptide of MPO was raised against two peptides derived from the MPO
propeptide (42TPQPSEGAAPAVLG56E and
72LVDKAYKERRESIKQRLRS91G) coupled via a
C-terminal cysteine residue to maleimide-activated keyhole limpet
hemocyanin, according to the manufacturer’s instructions (Pierce
Chemical Company, Rockford, IL).
cDNA, mutagenesis, and construction of expression vectors
A deletion mutant, -pro, was constructed from cDNA encoding
full-length MPO and inserted into the pREP vector for stable expression
in K562 cells [11
].
cDNA encoding human 1-m was constructed by polymerase
chain reactions (PCR), using the cDNA for human
1-m-bikunin (HAM-pCRscript), generously provided by Dr.
Bo Åkerström, as a template in a 20-cycle reaction. By design of
the primers, the Kozak consensus leader sequence for maximal
translational efficiency [16
] was introduced 5' to the
ATG initiation codon, and the flanking restriction enzyme sites
EcoRI and XbaI were added for subsequent cloning
into the plasmid. The primers were upstream
5'-GACTCCGAATTCGCCACCATGAGGAGCCTCGGGGCCCTGC (no.
5684) and downstream
5'-CTTCAGTCTAGATCATCTCGGGATTAAGATGGGC (no. 5215;
start and stop codons in boldface and restriction enzyme sites
underlined).
A cDNA for the chimeric protein, MPOpro/1-m, the
propeptide of MPO joined to 1-m, was constructed in a
two-step "spliced overhang extension" PCR. In the primary PCR
reactions, two separate amplifications with 100 ng DNA template, cDNA
of full-length MPO [17
], and 1-m-bikunin
(HAM-pCRscript), respectively, produced two fragments, MPOpro (10 PCR
cycles) and 1-m without the signal peptide (20 cycles),
respectively. The Kozak sequence, start and stop codon and flanking
sites for HindIII and BamHI, were used in
designing the primers as described above. The PCR primers in the two
amplifications were upstream
5'-GACTTCAAGCTTGCCACCATGGGGGTTCCCTTCTTCTCT (no.
3720) plus downstream 5'-CCCCACGTCCTGGTAGGCGCA (no. 3588) and upstream
5'-TGCGCCTACCAGGACGTGGGGGGCCCTGTGCCAACGCCGCC (no. 5311) plus downstream
5'-CTTCAGGGATCCTCATCTCGGGATTAAGATGGGC (no.
5310), respectively. The PCR products were isolated on agarose gel,
mixed, and subjected to a second 20-cycle splicing, PCR amplification
with primer nos. 3720 and 5310 to create MPOpro/1-m. The
resulting PCR product was digested by HindIII and
BamHI, followed by isolation on an agarose gel and cloning
into the desired expression vector.
cDNA for sTNFR1 was created with PCR technique as described above for
the cDNA of 1-m. Thus, pADTNF-R [18
] was
used as a template with the upstream primer
5'-TTCAGGTACCGCCGCCACCATGGGCCTCTCCACCGTGCCT
(no. 2200) and the downstream primer
5'-CTTCAGGGATCCTCAATTCTCAATCTGGGGTAGGCA (no.
3586). The construct was digested by KpnI and
BamHI and cloned into pCDNA3.
cDNA for MPOpro/sTNFR1 was constructed in a two-step spliced overhang
extension PCR by analogy with the creation of cDNA of
MPOpro/1-m. cDNA of full-length MPO [17
]
and pADTNF-R [18
], respectively, were used as templates
in the two primary PCR reactions with the following PCR primers:
upstream
5'-GACTTCGGTACCGCCACCATGGGGGTTCCCTTCTTCTCT (no.
3587) plus downstream 5'-CCCCACGTCCTGGTAGGCGCA (no. 3588) and upstream
5'-TGCGCCTACCAGGACGTGGGGGATAGTGTGTGTCCCCAAGGA (no. 3591) plus
downstream
5'-CTTCAGGGATCCTCAATTCTCAATCTGGGGTAGGCA (no.
3586), respectively. In the second PCR amplification splicing with
primer nos. 3587 and 3586, the cDNA of fusion protein MPOpro/sTNFR1 was
created with the restriction enzyme sites for KpnI and
BamHI. Subsequently, the cDNA was cloned into the desired
expression vector. All PCRs were performed in a Perkin-Elmer 480
thermal cycler using Pfu polymerase (Stratagene, San Diego,
CA), according to the manufacturer’s instructions.
Green fluorescent protein (GFP) and MPOpro/GFP constructs
Constructs for expression in promyelocytic cell line
cells used a pEF vector [12
], kindly provided by Dr.
Mary Dinauer (Indiana University, Indianapolis) and used previously for
studies of gp91phox biosynthesis [19
,
20
]. The GFP construct was derived from pADRSVhGFPpA,
provided by Dr. Bev Davidson (Vector Core, University of Iowa, Iowa
City).
Stably transfected cell lines
32D cl3 cells [21
, 22
] were
transfected with pcDNA3 constructs using the Bio-Rad Gene PulserTM
(Bio-Rad, Hercules, CA) with electrical settings of 960 µF and
260–300 V, as described previously [23
,
24
]. After electrophoresis, 1 mg/ml geneticin was added
to select for recombinant clones containing the geneticin-resistance of
pcDNA3. Individual antibiotic-resistant cell clones were expanded in
suspension culture and screened for the expression of the transfected
protein by biosynthetic radiolabeling. The clones with appropriate
expression were chosen for the experiments described. Stably
transfected K562 cells were generated using the expression vector pREP
as previously described [11
, 25
]. For
stable transfection of PLB-985 cells, the vector pEF was used
[12
]. Cell cultures were incubated in 5%
CO2 at 37°C in a fully humidified atmosphere.
Exponentially growing cells were used in all experiments.
Biosynthetic radiolabeling
Biosynthetic radiolabeling of the newly synthesized proteins by
stably transfected cell lines was carried out as described previously
[11
, 24
25
26
27
]. Cells were starved for 30 min
followed by incubation with
[35S]methionine/[35S]cysteine for
radiolabeling of newly synthesized proteins (pulse-labeling). For chase
of the radiolabeled protein variants, the cells were resuspended in
complete nonradioactive medium after radiolabeling. At various time
intervals, cells were withdrawn and subjected to analysis.
Peroxidase assay
The peroxidase activity of recombinant proteins was assessed in
activity gels, as previously described [28
]. Transfected
cells were lysed in a nondenaturing sample buffer and separated in a
native polyacrylamide gel. Samples were electrophoresed toward the
cathode until the dye front was 
0.5 cm from the bottom of the gel.
The gel was rinsed briefly in distilled water, and peroxidase activity
was detected by staining with trimethylbenzidine HCl as described
previously [28
].
Subcellular fractionation
Subcellular fractionation was performed on continuous gradients
of Percoll, as described previously [24
], as well as in
10–60% sucrose [25
]. Where indicated, the cell
homogenate was centrifuged in a Percoll density gradient, after which
nine fractions were collected with all the cytosol in fraction 9. The
distribution of lysosomes and Golgi elements in the gradient was
determined by assaying ß-hexosaminidase and galactosyl transferase
[29
, 30
]. The peak activities of
ß-hexosaminidase and galactosyl transferase in Percoll fractions of
32D cells were localized in fractions 2 and 6, respectively. The 59-kDa
heavy subunit of MPO expressed in 32D cells is localized predominantly
to fractions 1–2, whereas the precursors of MPO are found in fractions
5–8 [17
]. Subcellular fractionation in a sucrose
gradient was performed as described [25
]. Assays for
cytochrome c reductase and ß-glucuronidase, markers for the ER and
lysosome, respectively, were performed. The 59-kDa heavy subunit of
normal MPO expressed in stably transfected K562 cells sediments
predominantly in fractions 10–16 in this gradient, coincident with the
lysosomal marker ß-glucuronidase, whereas the 90-kD precursors of MPO
sediment in fractions 5–12, correlating well with the distribution of
the ER markers CRT and cytochrome c reductase [25
].
Immunoprecipitation
Immunoprecipitations from 32D cells were performed as described
previously [5
]. Immunoprecipitations from transfected
K562 cells and PLB-985 cells used previously described buffers for
solubilization [11
].
Immunoelectron microscopy
32D cells stably transfected with 1-m or with
MPOpro/1-m were fixed for 24 h in 4%
paraformaldehyde in 0.1 M PHEM buffer (pH 6.9) and then
processed for ultrathin cryosectioning as described previously
[31
]. Cryosections (45 nm) were cut at -125°C using
diamond knives (Drukker Cuijk, The Netherlands) in an
ultracryomicrotome (Leica Aktiengesellschaft, Vienna, Austria) and were
transferred with a mixture of sucrose and cellulose onto formvar-coated
copper grids [32
]. The grids were placed on 35-mm Petri
dishes containing 2% gelatin. Double-immunolabeling was performed
using a previously described procedure [33
] with 10- and
15-nm protein A-conjugated colloidal gold probes (EM Lab, Utrecht
University, The Netherlands). Staining used rat monoclonal ID4B against
mouse lysosome-associated membrane protein 1 (LAMP-1; CD107a;
Pharmingen Leiden, The Netherlands) and rabbit anti-1-m.
After immunolabeling, the cryosections were embedded in a mixture of
methylcellulose and uranyl acetate and examined with a Philips CM 10
electron microscope (Eindhoven, The Netherlands). For the controls, the
primary antibody was replaced by a nonrelevant rabbit antiserum.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-m is sorted for storage in myeloid cells1-m was investigated in stably
transfected 32D and K562 cells. The 1-m of plasma is
mainly manufactured normally by liver cells and secreted after cleavage
from a precursor protein [34
, 35
]. Two
forms of newly synthesized 1-m with different molecular
sizes were observed during radiolabeling of transfected 32D cells. A
significant fraction of the synthesized high molecular-size species was
secreted rapidly into the culture medium (Fig. 1 A
). The remaining intracellular forms of 1-m were
degraded slowly, as judged by chase of the radiolabeled material (Fig. 1A)
. To further investigate the fate of the intracellularly retained
1-m, subcellular fractionation experiments were
performed. After 30 min of pulse-radiolabeling, most of the
radiolabeled 1-m was present in light fractions (nos.
6–8), corresponding to ER and Golgi (Fig. 1B)
. With time, however,
labeled 1-m appeared in dense fractions (nos. 1–2),
indicating transfer to storage granules. These data show unexpectedly
that a nonmyeloid, normally secreted protein can be sorted for storage
when expressed in myeloid cells.
 View larger version (40K): [in a new window] |
Figure 1. Targeting 1-m to dense cytoplasmic organelles. 32D
cells expressing 1-m were radiolabeled biosynthetically
for 30 min, followed by chase of the label for the indicated time
periods. At depicted time points (A), 20 x 106 cells
were removed and after lysis, subjected to immunoprecipitation with
anti-1-m. In addition, 1-m was
precipitated from the incubation medium at each chase time point. At
indicated time periods (B), 100 x 106 cells were
removed and homogenized, after which subcellular fractionation of the
postnuclear supernatant was performed by centrifugation in Percoll,
with subsequent collection of nine subcellular fractions, fraction 9
containing all cytosol. The fractions were lysed and subjected to
immunoprecipitation with anti-1-m. The
immunoprecipitates were analyzed by SDS-PAGE followed by fluorography.
The fluorograms were exposed for 11 and 10 days, respectively. The
positions of 1-m are indicated to the right with arrows.
Peak activities of ß-hexosaminidase, fraction 2, and galactosyl
transferase, fraction 6, indicate the position of lysosomes and Golgi
elements, respectively.
|
1-m on its fate in
stably transfected 32D and K562 cells. Two forms of
MPOpro/1-m were synthesized in 32D cells: The 43-kDa
species was predominant intracellularly, and the 47-kDa protein was
predominant extracellularly (Fig. 2 A
). Both forms diminished during chase of the radiolabel, and the
coincident appearance of a 32-kDa protein was not recognized by the
anti-MPOpro antibody (unpublished results) and therefore presumably
represented free 1-m. Chloroquine treatment to block
lysosomal hydrolysis did not inhibit the release of 1-m
from the chimera (unpublished results), suggesting that the cleavage
was not associated with the lysosome. However, processing and secretion
of MPOpro/1-m were inhibited by BFA (Fig. 2B)
, which
blocks egress of protein from ER, indicating that these two events
occurred in a Golgi or post-Golgi compartment, access to which should
be prevented by BFA [36
, 37
]. Upon
subcellular fractionation, most of the MPOpro/1-m
synthesized early was in the light fractions corresponding to the ER
and the Golgi, but in contrast to 1-m (Fig. 1B)
, only a
small fraction of intact MPOpro/1-m was ever visible in
dense fractions (Fig. 2C)
. Instead, a lower molecular-weight form,
corresponding to cleaved 1-m (compare with Fig. 2A
),
appeared in dense fractions (Fig. 2C)
, indicating processing of
MPOpro/1-m to 1-m during transfer to
dense organelles. Thus, the addition of the MPOpro peptide sequence at
the amino terminus of 1-m delayed but did not inhibit
its delivery to the dense organelles.
 View larger version (45K): [in a new window] |
Figure 2. Processing MPOpro/1-m with cleavage of
1-m and targeting to dense cytoplasmic organelles. 32D
cells expressing MPOpro/1-m were radiolabled
biosynthetically for 30 min followed by chase of the label for the
indicated time periods. At depicted time points (A), 20 x
106 cells were removed and after lysis, subjected to
immunoprecipitation with anti-1-m. In addition,
MPOpro/1-m was precipitated from the incubation medium
at each chase time point. The same experiment was carried out in the
presence of 5 µg/mL BFA (B), present during starvation,
radiolabeling, and chase of the label. At indicated time points (C),
100 x 106 cells were removed and homogenized, after
which subcellular fractionation of the postnuclear supernatant was
performed by centrifugation in Percoll, as described in the legend to
Figure 1
. The immunoprecipitates were analyzed by SDS-PAGE and
fluorography. The fluorograms were exposed for 11, 7, and 11 days,
respectively. The positions of MPOpro/1-m and
1-m are indicated with arrows to the right. Peak
activities of ß-hexosaminidase, fraction 2, and galactosyl
transferase, fraction 6, indicate the position of lysosomes and Golgi
elements, respectively.
|
1-m in
32D cells reflected a limitation of the expression system rather than
an intrinsic property of the construct, we compared the fate of
wild-type and chimeric 1-m in K562 cells. Transfected
K562 cells secreted the 28-kDa wild-type 1-m, whereas
the majority of the 43-kDa chimeric species remained intracellular and
was not secreted at the same rate as the parent protein (unpublished
results). In analysis of subcellular fractions separated on 10–60%
gradients of sucrose, the MPOpro/1-m did not cosediment
with dense organelles containing mature MPO when expressed but rather
were retained in the ER as judged by cosedimentation of biochemical
markers for ER (unpublished results). MPOpro/1-m was not
retained in the ER by association with CRT or CLN (unpublished
results).
Results from immunoelectron microscopy using double immunogold-labeling
revealed that 1-m colocalized with the lysosomal marker
LAMP-1 [38
, 39
] in 32D cells transfected
with 1-m (Fig. 3 A
). Specific labeling was associated with multivesicular bodies
that correspond to the abnormal granules of 32D cells. These results
are consistent with those from subcellular fractionation experiments
demonstrating targeting of 1-m to dense organelles (Fig. 1B)
. As seen in cells expressing 1-m alone, the
1-m of cells transfected with chimeric
MPOpro/1-m colocalized with LAMP-1 in multivesicular
bodies (Fig. 3B)
. This product likely represents 1-m
released from the chimeric protein and accumulated in granules,
supported by the finding of almost exclusively free 1-m
in the dense organelles, as judged by results from radiolabeling
experiments (Fig. 2C)
. Taken together, these data demonstrated the
destination of the 1-m released from the chimera to be
dense, multivesicular bodies, but the specific localization of the
intact chimeric MPOpro/1-m was not identified.
 View larger version (110K): [in a new window] |
Figure 3. Colocalization of 1-m and LAMP-1. Ultrathin cryosections
from 32D cells transfected with 1-m (A) or with
MPOpro/1-m (B) were double-labeled with rat anti-LAMP-1,
followed by rabbit anti-rat immunoglobulin G (IgG) and protein A gold
(10 nm), and with rabbit anti-1-m, followed by protein A
gold (15 nm). The pattern of labeling in both transfectants is very
similar; 1-m and LAMP-1 colocate in small and large
multivesicular bodies. Original bar = 200 nm.
|
1-m was sensitive to digestion with Endo-H, indicating
presence of high mannose glycoforms produced in the ER (Fig. 4 A
). Conversely, the larger species of 1-m, present
intra- and extracellularly, was resistant to Endo-H, indicating the
presence of complex glycoforms normally produced in the Golgi complex
by trimming mannose and adding new terminal sugars. The results
indicate that the smaller size 1-m, after ER export, is
converted to the larger size 1-m as a result of complex
oligosaccharide formation. Digestion with N-glycosidase F to
remove all N-linked carbohydrates revealed two intracellular species of
1-m with distinct protein core sizes, of which only the
larger form was detected extracellularly (Fig. 4A)
, indicating that
proteolytic trimming of 1-m occurred in storage
organelles but not during constitutive secretion.
MPOpro/1-m showed a much slower acquisition of complex
carbohydrates occurring in the Golgi complex as compared with
1-m (compare Fig. 4
A and B). Thus, a partially
preserved susceptibility to Endo-H, indicating lack of complex
glycoforms, was observed after 3 h of chase of radiolabeled
MPOpro/1-m, a time when 1-m was fully
Endo-H-resistant. These data are consistent with a retarded rate of
carbohydrate processing in MPOpro/1-m as compared with
1-m, most likely because of a prolonged ER retention of
the chimera.
 View larger version (29K): [in a new window] |
Figure 4. Digestion of 1-m and MPOpro/1-m with
glycosidases. 32D cells expressing 1-m (A) or
MPOpro/1-m (B) were radiolabeled biosynthetically for 30
min followed by chase of the label for 3 h. At these time points,
100 x 106 cells were removed and after lysis,
subjected to immunoprecipitation with anti-1-m. In
addition, the incubation medium was analyzed at the end of the chase
period. Aliquoted immunoprecipitates were incubated with 0.2 units/mL
endoglycosidase H (+ Endo-H) or 7 units/mL N-glyc (+
N-glycosidase F) or were served as controls (-). The samples were
analyzed as described in the legend to Figure 1
. The fluorograms were
exposed for 21 days.
|
 View larger version (37K): [in a new window] |
Figure 5. Biosynthesis of sTNFR1 and MPOpro/sTNFR1 in 32D cells. 32D cells
expressing sTNFR (top panel) or MPOpro/sTNFR (bottom panel) were
radiolabeled biosynthetically for 30 min followed by chase of the label
for the indicated time periods. At depicted time points, 20 x
106 cells were removed and after lysis, subjected to
immunoprecipitation with anti-sTNFR1. The incubation medium was also
assayed. Immunoprecipitates were analyzed by SDS-PAGE and fluorography.
The positions of sTNFR1 or MPOpro/sTNFR1, respectively, are indicated
with arrows. The fluorograms were exposed for 5 weeks.
|
 View larger version (24K): [in a new window] |
Figure 6. Subcellular localization of MPO and MPOpro/sTNFR1 in K562 cells. Stable
transfectants expressing normal MPO or MPOpro/sTNFR were cavitated and
separated on the same 10–60% sucrose gradient. Fractions from the
gradient were separated by SDS-PAGE, and duplicate immunoblots were
probed with antibody against MPO (mature MPO) or with antibody against
sTNFR1.
|
We first confirmed that the MPOpro/GFP construct was synthesized in the ER of the PLB-985 cells by assessing its glycosylation status. Whereas GFP lacks carbohydrate, MPOpro/GFP was glycosylated, because the addition of asparagine 139 in the propeptide resulted in tunicamycin-inhibitable N-linked glycosylation of the recombinant protein in the ER (unpublished results). Despite synthesis of MPOpro/GFP in the ER, it did not associate with the molecular chaperones CRT or CLN during their biosynthesis. Lysates of pulse-chased MPOpro/GFP transfectants were immunoprecipitated under nondenaturing conditions with antibodies to MPO (examining endogenous MPO production), CRT (recovering CRT-associated MPO precursors), and GFP (precipitating MPOpro/GFP) and were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by autoradiography (Fig. 7 ). Under these conditions, multicomponent protein complexes remain intact [26 , 27 ], but in contrast to MPO, MPOpro/GFP was not precipitated in a complex with CRT (Fig. 7) or CLN (unpublished results). Whereas GFP was expressed stably and remained intracellular in PLB-985 cells (unpublished results), MPOpro/GFP was degraded relatively rapidly in the ER by a process dependent on the proteasome, as judged from inhibition of degradation by N-acetyl-L-leucinyl-L-leucinyl-L-norleucinyl (ALLnL; unpublished results). Thus, in the PLB-985 cell line—cells competent to synthesize and target endogenous MPO—the addition of the propeptide to GFP was not sufficient for directing GFP to granule storage.
 View larger version (39K): [in a new window] |
Figure 7. Coprecipitation of CRT and MPO or MPOpro/GFP. Wild-type PLB-985 cells
(PLB-985) or transfected PLB-985 cells expressing MPOpro/GFP (4G4) were
radiolabled biosynthetically and chased for 0 or 4 h. Cell lysates
were immunoprecipitated with antibodies to MPO (M), CRT (C), or GFP (G)
to recover MPO, CRT-associated MPO or proMPO, or MPOpro/GFP.
|
 View larger version (40K): [in a new window] |
Figure 8. Biosynthesis of prodeletion mutant in K562 cells. (A) Transfectants
stably expressing wild-type MPO (top panel) or the (-pro) mutant
(bottom panel) were radiolabled biosynthetically for 1 h and
chased for 1–4 h. Cell lysates were immunoprecipitated with anti-MPO
to recover all MPO-related protein (MPO), sequentially with anti-CLN
and anti-MPO (CLN) to recover CLN-associated (-pro), or with
anti-CRT and anti-MPO (CRT) to recover CRT-associated (-pro). CLN
and the MPO precursor migrate nearly identically and could not be
distinguished in one-dimensional gels (top panel). However, (-pro) and
CLN were separated and identified after SDS-PAGE (bottom panel). (B)
Lysates from the parental K562 cells (K562), K562 transfectants stably
expressing wild-type MPO (MPO), K562 transfectants stably expressing
(-pro) [-(pro)], and PLB-985 cells (PLB) were separated by PAGE in a
native system and stained for peroxidase activity.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-m was expressed in
hematopoietic cell lines and sorted for storage in dense cytoplasmic
organelles. We demonstrated as well that fusion of the propeptide of
MPO to create chimeras altered the distribution of the unmodified
protein and resulted in ER retention.
Hematopoietic cells contain an array of granule compartments, whose
composition reflects the timing and coordination of gene expression
[2
]. Therefore, specific sorting mechanisms for
targeting to different granule subtypes may not be required. However, a
mechanism is needed for retrieval of proteins from the secretory
pathway to allow sorting for storage, because constitutive secretion
also occurs in myeloid cells, as demonstrated for proforms of newly
synthesized granule proteins [3
]. We have demonstrated
previously that LBP, a nonmyeloid protein that is normally secreted
after biosynthesis in liver cells, can be retained and targeted for
dense organelles when expressed in myeloid cells [5
]. We
now demonstrate that this is also the case also for 1-m,
which was chosen for this study, because it is secreted constitutively
by normal liver cells. Therefore, its sorting for storage in myeloid
cells was unexpected, but this and our previous results suggest that
retention for storage in myeloid cells may reflect cell-specific rather
than protein-specific determinants. Thus, even nonmyeloid proteins can
be targeted for storage when the conformation is acceptable to the
sorting machinery. A precursor protein, 1-m-bikunin is
synthesized in hepatocytes and cleaved in the Golgi followed by
secretion of 1-m and bikunin [35
].
Glycosylation may be important for secretion inasmuch as the secretion
of carbohydrate-free 1-m from recombinant insect cells
was reduced significantly [45
]. Our results demonstrate
that the 1-m synthesized in myeloid cells accomplished
complex glycoforms in the Golgi, followed by retention for storage or
partial secretion. We therefore have no reason to anticipate that
abnormal glycosylation would be responsible for the intracellular
retention for storage of 1-m in myeloid cells.
Furthermore, it is noteworthy that 1-m belongs to the
lipocalin protein family, and human neutrophils normally synthesize and
store NGAL, another lipocalin, in specific granules [1
].
Therefore, the targeting of 1-m in myeloid cells may
reflect similarities in targeting sequences between these two
structurally related proteins.
One essential observation is that myeloid cells can discriminate and
selectively target some proteins for storage. The routing of myeloid
and nonmyeloid proteins likely depends on their structural determinants
inasmuch as not all soluble proteins delivered to the trans-Golgi
network are sorted for storage in myeloid cells. For instance,
the behavior of the soluble TNFR1 that was exported from ER and
secreted contrasts with that of LBP [5
] and
1-m, which were intracellularly retained. Similarly, we
have observed that propeptide-deleted pro-MPO is secreted
constitutively in myeloid cell lines without accumulation of the final
product in granules [17
]. This behavior is not a
property peculiar to transfected, cultured cell lines; results from
normal bone marrow cells have indicated that differences exist in
sorting granule proteins between constitutive secretion and retrieval
for storage in granules [46
], e.g., a significant
fraction of lysozyme was routed for constitutive secretion. The
specific physicochemical properties that dictate targeting to specific
storage compartments in myeloid cells remain to be elucidated.
Unlike endocrine, neuroendocrine, and exocrine cells, which manufacture
granules for regulated secretion of their stored products, myeloid
cells often use the endosomal pathway for manufacturing
lysosome-related organelles. This involves transfer of coated vesicles
from the TGN to late endosomes, often with the appearance of
multivesicular bodies that form mature, lysosome-related storage
organelles [47
]. The endosomal pathway was involved in
targeting 1-m for dense organelles as judged by the
results from immunogold-labeling in 32D cells. Thus, 1-m
colocalized with the lysosomal marker LAMP-1 [38
,
39
], and much of the labeling was associated with
organelles having the features of multivesicular bodies corresponding
to the vacuole-like abnormal granules of 32D cells [48
].
Azurophil granules of neutrophils are lysosome-like and may belong to
the lysosome-related organelles [47
], although they lack
LAMP-1 and LAMP-2 [49
]. Thus, all lysosome-related
organelles, including azurophil granules, are defective in the
Chediak-Higashi syndrome [47
]. Endogenous granule
proteins of 32D cells are more stable than exogenously introduced
1-m, which is eventually degraded in the dense
organelles. This may result from abnormal packing of
1-m, which is normally not present in this environment,
thus resulting in a lack of protection against proteases.
These studies were also designed to assess the potential role of the
propeptide of MPO for the intracellular targeting of the enzyme. A
deletion mutant in which the propeptide was eliminated from the MPO
precursor aborted proper delivery of mature MPO to the dense organelles
of the 32D cell line, despite the fact that constitutive secretion was
not globally affected [17
]. Moreover, the propeptide of
MPO targets a fusion protein of MPOpro and lysozyme to the lysosomes of
Chinese hamster ovary cells, followed by cleavage of the MPO propeptide
from the lysozyme moiety [50
]. Similarly, we tracked the
subcellular fate of transfected chimeric constructs to determine if the
addition of the propeptide of MPO would redirect proteins from the
constitutive secretory pathway to sorting for storage in granules and
thereby support a functional role for that domain in proper cellular
targeting. Thus, we created chimeric proteins that would exhibit a
"gain of function" if the propeptide in fact represented a
targeting domain. A chimera of MPOpro and 1-m was
targeted to dense organelles of 32D cells after cleavage of the
1-m moiety, but the potential information gained by
adding the MPOpro to 1-m was undermined by our
unexpected finding of targeting for storage of the normally secreted
1-m alone. It should be noted that a substantial portion
of newly synthesized 1-m was released extracellularly
(Fig. 1A)
. Therefore, a decreased secretion to the medium is consistent
with increased intracellular retention and sorting. However, the
relatively slow appearance of complex carbohydrates on
MPOpro/1-m, as compared with 1-m alone,
suggested that the half-life of MPOpro/1-m in the ER was
prolonged. Although this observation supported a role for the
propeptide in a transient ER retention, this ER retention may not
specifically reflect a targeting function for the propeptide. Clearly,
the relative fates of 1-m and MPOpro/1-m
differed when expressed in K562 cells. In this human hematopoietic cell
line, wild-type 1-m was secreted into the culture
medium, whereas the fusion protein was trapped intracellularly,
primarily within the ER, without detectable targeting for storage in
dense organelles. Therefore, a role for the propeptide in sorting for
storage in granules could be neither confirmed nor excluded from these
experiments.
32D cells and K562 cells transfected with wild-type sTNFR1 or the MPOpro/sTNFR1 demonstrated the same differential phenotypes, namely secretion of the wild-type proteins, but intracellular retention of the fusion proteins without secretion to the same extent as that seen for the parental protein. Results from subcellular fractionation studies in K562 cells demonstrated that the chimeric species containing the MPO propeptide was retained in fractions that cosedimented with markers for the ER. The chimeric protein was not retained in the ER by association with molecular chaperones CRT and CLN (unpublished results), indicating that the presence of the propeptide was not sufficient to mediate interactions with those ER lectins. Last, data from studies comparing the subcellular localization of GFP and MPOpro/GFP in PLB-985 cells, human promyelocytic cells constitutively synthesizing, and packaging normal MPO [26 , 40 ] demonstrated that the addition of the propeptide to GFP caused ER retention followed by proteasome-dependent degradation, thus making it difficult to judge whether the propeptide could affect sorting at more distal locations of the secretory pathway.
We also examined the fate of the propeptide deletion mutant expressed in K562 cells. The (-pro)MPO construct failed to egress from the ER and decayed rapidly in a proteasome-dependent fashion. Coexpression of the propeptide with the (-pro)MPO construct failed to reconstitute proper targeting or function of the deletion mutant (unpublished results). These data suggest that deletion of the propeptide irreversibly compromised proper functioning of the MPO precursor, resulting in a species rapidly eliminated by the ER-associated degradation system and the proteasome and that the propeptide did not function as an intramolecular chaperone. Conversely, previous results [17 ] showed that (-pro)MPO expressed in 32D cells was secreted, albeit less than normal, without being targeted for storage, indicating a role of the propeptide for targeting for storage in granules in 32D cells [17 ]. The relative fates of (-pro)MPO in these two cell lines may parallel the extent of heme incorporation into the precursor; whereas there is limited incorporation of heme into (-pro)MPO expressed in 32D cells [17 ], the same construct expressed in K-562 cells lacked heme. If heme is necessary to allow ER export of (-pro)MPO, as is the case in the biosynthesis of wild-type MPO [41 ], one would anticipate that any compromise in heme insertion per se would create a phenotype similar to that of maturation arrest of the precursor.
Collectively, these data demonstrate that fusion of the propeptide of MPO altered subcellular targeting relative to that of the unmodified protein, but that the propeptide alone was not sufficient to target the reporter protein to an intracellular site where further proteolytic processing and maturation of MPO precursors occur. However, as the chimeric proteins were retained in the ER, we cannot exclude a role for the propeptide in directing MPO to dense organelles, as indicated by previous studies demonstrating that processing of MPO precursors and efficient sorting for storage require an intact propeptide [17 ]. A BLAST search of the Swissprot protein databank for sequences related to the MPO propeptide recovered only precursor forms of other members of the family of animal peroxidases, including eosinophil peroxidase, thyroid peroxidase, and lactoperoxidase, perhaps suggesting a function shared by these structurally related proteins. In that light, the propeptide may, in addition to being necessary for granule targeting, provide a mechanism for retention of the apopro-form of animal peroxidases in the ER, independent of association with the chaperones CRT and CLN, to facilitate heme insertion and proper folding and thus generate a precursor competent to exit the ER. Alternatively, ER retention may reflect abnormal folding followed by degradation, as seen for some of the chimeras investigated.
We conclude that nonmyeloid proteins that are normally secreted can be targeted for storage when expressed in myeloid cells, indicating the existence of tissue-specific retention mechanisms for intracellular storage, and that myeloid cells selectively route some proteins but not others for storage. In addition, the propeptide of MPO may play a role in retention mechanisms necessary for proper maturation and storage of MPO.
Received September 11, 2001; revised October 18, 2001; accepted October 22, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-microglobulin are important for secretion and tissue localization but not for immunological properties Glycobiology 10,891-900This article has been cited by other articles:
 |
|
 |
|
  M. Goedken, S. McCormick, K. G. Leidal, K. Suzuki, Y. Kameoka, J. M. Astern, M. Huang, A. Cherkasov, and W. M. Nauseef Impact of Two Novel Mutations on the Structure and Function of Human Myeloperoxidase J. Biol. Chem., September 21, 2007; 282(38): 27994 - 28003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Gao, M. Hansson, J. Calafat, H. Tapper, and I. Olsson Sorting soluble tumor necrosis factor (TNF) receptor for storage and regulated secretion in hematopoietic cells J. Leukoc. Biol., October 1, 2004; 76(4): 876 - 885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Rosen, Y. Gao, E. Johnsson, and I. Olsson Artificially controlled aggregation of proteins and targeting in hematopoietic cells J. Leukoc. Biol., November 1, 2003; 74(5): 800 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lemansky, M. Gerecitano-Schmidek, R. C. Das, B. Schmidt, and A. Hasilik Targeting myeloperoxidase to azurophilic granules in HL-60 cells J. Leukoc. Biol., October 1, 2003; 74(4): 542 - 550. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Gao, H. Rosen, E. Johnsson, J. Calafat, H. Tapper, and I. Olsson Sorting of soluble TNF-receptor for granule storage in hematopoietic cells as a principle for targeting of selected proteins to inflamed sites Blood, July 15, 2003; 102(2): 682 - 688. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
