| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,§,§,§
,§
*
Division of Dermatology,
Division of Pulmonary Medicine,
Department of Microbiology and Immunology, and
§
Molecular Biology Institute, University of California School of Medicine, Los Angeles, CA 90095;
¶
Howard Hughes Medical Institute, Albert Einstein College of Medicine, Bronx, NY 10461;
||
Office of the Dean, Harvard School of Public Health, Boston, MA 02115;
#
Division of Immunology and Transplantation Biology, Department of Pediatrics, Stanford University, Stanford, CA, 94305; and
**
Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
helical bundle motif,
with the helices enriched for positively charged amino acids,
including arginine and lysine residues. Denaturation of the polypeptide
reduced the helical content from 49 to 18% resulted in complete
inhibition of antimicrobial activity. Chemical modification of the
arginine, but not the lysine, residues also blocked the antimicrobial
activity and interfered with the ability of granulysin to adhere to
Escherichia coli and Mycobacterium
tuberculosis. Granulysin increased the permeability of
bacterial membranes, as judged by its ability to allow access of
cytosolic ß-galactosidase to its impermeant substrate. By electron
microscopy, granulysin triggered fluid accumulation in the periplasm of
M. tuberculosis, consistent with osmotic perturbation.
These data suggest that the ability of granulysin to kill microbial
pathogens is dependent on direct interaction with the microbial cell
wall and/or membrane, leading to increased permeability and
lysis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antimicrobial proteins and peptides are diverse in their
structure and mechanism by which they kill infectious agents. Many
antimicrobial peptides, however, have a secondary structure based on
either helices or ß sheets. The antimicrobial activity of these
peptides is dependent on their ability to form multimers that
facilitate pore formation leading to cell death (6). In
addition, most antimicrobial peptides are cationic, although amino acid
usage varies, including arginine, histidine, and lysine (7, 8). The putative structure of granulysin is a four helical
bundle similar to the amoebapore family members (3).
Because granulysin is the only antimicrobial peptide to have been
identified in T cells, we thought it important to clarify its mechanism
of action.
Experiments were performed to correlate the structure and
function of granulysin using biophysical approaches. Synthetic peptides
of granulysin conforming to a putative helix-loop-helix motif (aa
1–35, 36–70, and 31–50) retained 50–80% of anti-bacterial
activity, whereas those peptides without this predicted structure (aa
1–20, 16–35, 46–65, 61–80) had <20% activity. The structural
model also predicts that the helices are amphipathic, including 15
positively charged amino acids: 12 arginine (16%) and three lysine
residues. Chemical modification of the arginine residues caused
complete inhibition of the antimicrobial effects of granulysin;
however, modification of the lysine residues did not inhibit the
antimicrobial activity. Granulysin altered bacterial membranes by
increasing their permeability, inducing lesions on the
surface of bacteria and separation of the cell wall and membranes from
the cytoplasm. These data suggest that the ability of granulysin to
kill microbial pathogens is dependent on interactions with the
microbial cell wall or membrane leading to increased permeability and
osmotic lysis. Differences in structure and function of granulysin
compared with amoebapores suggest that granulysin has evolved as a
potent weapon of T cells to combat microbial pathogens.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Granulysin was produced in Escherichia coli BL21 (DE3) transformed with the kanamycin-selective vector, pET28, containing a hexahistidine fusion tag (Novagen, Madison, WI) as previously described (2). Briefly, the transformed E. coli were grown in 2x YT and induced with 1 mM isopropyl-ß-D-thiogalactoside (Fisher Scientific, Pittsburgh, PA). The bacteria were harvested and denatured in 6 M guanidine HCl/0.05 mM Tris-HCl (pH 7.4). Granulysin was purified via nickel affinity chromatography according to the manufacturer’s recommendation (Qiagen, Valencia, CA) and eluted with 0.2 M imidazole, then reduced with 10 mM DTT. The denatured granulysin was renatured in 0.75 M arginine, 0.05 M Tris-HCl (pH 8), 0.05 M KCl, 0.1 mM EDTA, and 10 mM oxidized DTT at a 1/5 dilution with constant stirring for 48 h at 4°C. The renaturing buffer was exchanged by dialysis with 2 mM sodium phosphate and 13 mM NaCl (pH 7.2), then lyophilized. The granulysin pellet was rehydrated and treated with thrombin for 16 h to cleave the hexahistidine tag. Following thrombin cleavage, the protein was loaded onto a Rainin C18 reverse phase chromatography column (Braintree, MA) and eluted by a linear gradient of 10–60% aqueous acetonitrile in 0.1% trifluoroacetic acid. The fractions containing granulysin, as determined by Coomassie staining of a 15% SDS-PAGE gel, were lyophilized and hydrated in 10 mM sodium phosphate (pH 7.2) unless otherwise noted. The final protein concentration was determined using the bicinchoninic acid protein assay (Pierce, Rockford, IL) with BSA as a standard. Protein purity was assessed by Coomassie staining of 15% SDS-PAGE gels and was >95%. Additionally, purified granulysin was analyzed by matrix-assisted laser desorption ionization mass spectrometry and shown to contain one species at 9081 Da that corresponded to the calculated m.w., which is 9070.4 assuming that four of the five cysteines are involved in disulfide bridges.
Peptide synthesis
Peptides were synthesized using F-moc chemistry on an Applied Biosystems (Foster City, CA) automatic peptide synthesizer and were purified to >95% homogeneity by reverse phase HPLC, and peptide composition was confirmed by mass spectrometry and amino acid analysis. Stock peptide solutions were prepared at 20 mg/ml in DMSO. The sequences for the peptides used are as follows: 1–35, GRDYRTSLTIVQKLKKMVDKPTQRSVSNAATRVSR; 36–70, TGRSRWRDVSRNFMRRYQSRVIQGLVAGETAQQIS; 1–20, GRDYRTSLTIVQKLKKMVDK; 31–50, TRVSRTGRSRWRDVSRNFMR; 16–35, KMVDKPTQRSVSNAATRVSR; 46–65, RNFMRRYQSRVIQGLVAGET; and 61–80, VAGETAQQISEDLR.
Instead of constructing the 62–74 peptide, additional sequence from the 15-kDa form of granulysin was used to construct the 61–80 peptide and maintain this peptide at a similar length as the others. Amoebapore A was purified from trophozoites of E. histolytica using reverse phase HPLC as the final purification step (9). Melittin was obtained from Sigma (St. Louis, MO).
CFU assay
The CFU assay was performed as described previously (5, 10). Briefly, E. coli (strain ML-35p) was grown in trypticase soy broth (TSB) collected in mid-log phase and washed three times with 10 mM sodium phosphate, pH 7.4, supplemented with 0.03% TSB. Various concentrations of granulysin and granulysin peptides with 60 µl of 2.5 x 105 bacteria/ml were incubated at 37°C for 3 h. After incubation, the bacteria were placed on ice, and 10-fold dilutions were prepared and plated on TSB agar plates. The plates were incubated for 2 h at 37°C, then overnight at room temperature. For kinetic studies, bacteria were incubated with 0.5 and 2.5 µM granulysin and then diluted and plated at the indicated times. To evaluate the effects of pH, granulysin was hydrated in 10 mM sodium phosphate supplemented with 0.03% TSB at various pH units. The samples were incubated at 37°C for 3 h and processed as described above. To evaluate the effects of ionic concentration on granulysin, E. coli were incubated with various concentrations of granulysin in 10 mM sodium phosphate, pH 7.4, supplemented with 0.03% TSB and 40, 90, 140, and 290 mM NaCl. After the incubation period, the samples were diluted and plated as described above. M. tuberculosis was recovered from frozen aliquots and washed twice in Middlebrook 7H9 medium. To prevent clumping, mild sonication was performed before and after washing. Bacteria (1 x 104) were added in 90 µl/well of a 96-well flat-bottom tissue culture plate. Recombinant granulysin and synthetic peptides were dissolved and diluted in 10 mM sodium phosphate buffer, pH 7.2, supplemented with 0.03% TSB, and incubated with bacteria at 37°C for 72 h. After incubation, serial dilutions were plated on Middlebrook 7H11 agar plates, and the colonies were counted after 2 wk.
Radial diffusion assay
The radial diffusion assay was used to determine the relative bactericidal activity of chemically modified granulysin and granulysin peptides. A modified assay (11) was performed. Briefly, ML-35p E. coli was collected at mid-log phase, and 4 x 106 bacteria were pour-plated in 10 ml of 1% agarose dissolved in 10 mM sodium phosphate, pH 7.4, supplemented with 0.3% TSB. After the agarose solidified, 3-mm holes were punched and removed from the agarose. Then, 5 µl of the samples were added to the wells and incubated at 37°C for 3 h. After the incubation, 10 ml of 1% agarose and 6% TSB were poured as an overlay. The plates were incubated for an additional 16 h at 37°C. The antimicrobial effects were evaluated by measuring the amount of clearing in millimeters and multiplying by a factor of 10 to yield radial diffusion units.
Chemical modification for radial diffusion assay
To chemically modify granulysin, granulysin peptides and amoebapore A, we incubated 100 µM protein, 40 mM butanedione (BAD) or 20 mM citraconic anhydride (CAH; Sigma), and 10 mM sodium borate, pH 8.0, for 2 h. The samples were placed in the wells of the radial diffusion plate and incubated as described above.
Chemical modification for Western blot analysis
To 1 µg of granulysin, varying doses of BAD and CAH were used for chemical modification. The modified protein was incubated with 1.5 x 104 E. coli or 2.5 x 105 M. tuberculosis in 10 mM sodium phosphate, pH 7.4, supplemented with 0.03% TSB for 1 h, then washed three times with 10 mM sodium phosphate, pH 7.4, to remove unbound granulysin. The bacteria were suspended in 20% Tween 20 and subjected to two rounds of freezing and boiling. The level of granulysin retained with the E. coli was determined by Western blot analysis. Samples were run on a 15% SDS-PAGE and blotted to nitrocellulose. Granulysin was detected using the anti-granulysin mAb DH4 (12).
Membrane permeability of granulysin and granulysin peptides
The membrane permeability effects of granulysin and amoebapore A were assayed by adding protein to stationary phase E. coli ML-35p. Bacteria were grown for 16 h at 37°C in TSB broth, then washed three times in 10 mM sodium phosphate, pH 7.4. O-nitrophenyl-ß-D-galactoside (ONPG; Sigma; 2.5 mM), 1 x 107 bacteria, and 2.5 µM of protein were added to a well of a microtiter plate (Costar, Cambridge, MA). As a control, 1 µM melittin was incubated with E. coli as described above. The plate was read in a microtiter plate reader at 410 nm. To assess the effect of chemically modified granulysin on membrane permeability, 2.5 µM granulysin was preincubated with 40 mM BAD and then added to the assay system described above. BAD (40 mM) had no effect on the assay system (data not shown).
Circular dichroic analysis of granulysin
Circular dichroic measurements were performed on an Aviv 62A DS
circular dichroic spectrapolarimeter (Aviv, Lakewood, NJ) using a
quartz cuvette with a pathlength of 10 mm at 25°C as described
previously (13). The data are reported as the average of
10 scans at 50 nm/min with a 0.1-nm step resolution. Spectra were
recorded in the far UV at a protein concentration of 1 mg/ml in 10 mM
sodium phosphate buffer, pH 7.2. For the evaluation at various pH
units, granulysin was diluted in 10 mM sodium phosphate at the
appropriate pH. Data are reported as the mean residue ellipticity,
[
], in units of degrees cm2/decimole.
A model of granulysin was constructed based on the nuclear magnetic
resonance (NMR) structure of NK-lysin (PDB accession code 1NKL)
(14). Briefly, using the fold recognition server
(http://fold.doe-mbi.ucla.edu/), the sequence of granulysin was
threaded onto the NMR backbone of NK-lysin (15). Rasmol
freeware (http://www.ncbi.nlm.nih.gov/Structure/CN3D/rasmol.html) was
used to visual the three-dimensional structure of granulysin.
Estimation of the helical concentration of granulysin was performed
using Prosec secondary structural analysis software (Aviv).
Transmission electron microscopy
For transmission electron microscopy 5 x 106 tubercle bacilli (Erdman strain) were incubated with either 30 µM purified granulysin or control protein for 80 h. Bacilli were then fixed in 2.5% glutaraldehyde in 0.1 M sodium cacodylate, pH 7.4, buffered solution for 6 h. Bacteria were postfixed in 1% osmium tetroxide, en bloc stained in uranyl acetate, dehydrated in a graded ethanol series, then embedded in LX112 (Ladd Research, Burlington, VT) under vacuum in beem capsules and polymerized at 60°C. Thin sections were stained with uranyl acid and lead citrate before examination on a JEOL 1200EX or JEOL 100CX transmission electron microscope (Peabody, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ability of peptide constituents of the full-length 9-kDa human
granulysin were tested for antimicrobial activity against E.
coli using a CFU assay. Three peptides (peptides 1–35, 36–70,
and 31–50) had significant antimicrobial activity, reducing by >4
logs the number of viable E. coli (Fig. 1
A). Four peptides had no or
only weak antimicrobial activity (peptides 1–20, 16–35, 46–65, and
61–80), reducing by <2 logs the number of viable bacteria (Fig. 1B). These results were confirmed using a radial diffusion
assay (data not shown). The predicted structure of the peptides with
antimicrobial activity corresponded to protein segments of granulysin
with a helix-loop-helix motif according to a model of granulysin based
on the NMR structure of NK-lysin (Fig. 1D). In addition,
these peptides were rich in positively charged amino acids, arginine,
and lysine residues. In contrast, the peptides that had weaker
antimicrobial activity either corresponded to the protein segment of
granulysin with a single helix or helical-loop structure or had
fewer positively charged residues. These data correlate a
helix-loop-helix motif containing positively charged residues with the
antimicrobial activity of granulysin.
|
Because we have previously shown that granulysin killed M.
tuberculosis, it was reasonable to determine the antimicrobial
effect of the granulysin peptides on this pathogen. Similar to the
results observed when granulysin was incubated with E. coli,
full-length granulysin had the strongest antimicrobial effect against
M. tuberculosis, with 65% killing at 25 µM. When M.
tuberculosis was treated with peptides 1–35, 36–70, and 31–50,
killing was observed in a dose-dependent fashion (Fig. 1
C).
However, compared with granulysin only 25–40% of the bacteria were
killed when treated with 25 µM peptide, with as much as 45–60%
killing at a peptide concentration of 100 µM. Similar to the
experiments with E. coli, peptide 1–20 had no effect on
M. tuberculosis. However, peptide 31–50, which showed at
least 2-fold less activity than peptides 1–35 and 36–70 against
E. coli, showed equal activity to these two peptides against
M. tuberculosis. These differences may be due to differences
in the cell wall and membrane structure between the two organisms and
their ability to interact with the granulysin peptides.
Secondary structural analysis of recombinant granulysin
The putative three-dimensional structure of granulysin modeled on
the backbone of the NK lysin NMR data indicated that granulysin
conforms to a four helical bundle, with a predicted helical
content of 
74%. The helical content of recombinant granulysin
was measured by circular dichroism according to its molecular
ellipticity. We observed spectral minimum at both 222 and 208 nm,
indicating a highly helical structure, and based on secondary
structural analysis, this was determined to be 49%. The difference
between the measured and calculated helical content may reflect
inaccuracies in the estimation or indicate that part of the granulysin
did not refold during purification. We found no difference in the
spectra of granulysin from pH 8.2 to pH 3.0, indicating that granulysin
is an acid-stable protein (Fig. 2A). To denature granulysin,
granulysin was boiled, but this did not alter its circular dichroic
spectra. However, boiling in the presence of 2-ME, to destroy the
disulfide linkages, altered the circular dichroic spectra with a
calculated reduction in helical content to 18% (Fig. 2B). The loss of granulysin helices correlated directly
with decreased antimicrobial activity, as measured by the radial
diffusion assay (Fig. 2C). These data further implicate the
three-dimensional structure of granulysin in its antimicrobial
activity.
|
To further characterize the antimicrobial activity of granulysin
according to pH and ionic strength, a series of experiments was
performed. The kinetic activity of granulysin was analyzed to determine
the rate at which granulysin kills E. coli. By incubating
E. coli with either 0.5 or 2.5 µM granulysin at pH 7.2 and
measuring CFU vs time, we observed a 5-fold decrease in CFU after
2 h (Fig. 3A). The effect
of pH on granulysin antimicrobial activity was examined from a range of
pH 8 to pH 4.5 (Fig. 3B). Granulysin at pH 8 had the highest
specific activity against E. coli, with nearly a 7-log
reduction in CFU with 0.1 µM. Granulysin at pH 7.2 showed a 3-log
reduction in CFU at 0.1 µM. There was very little difference in the
antimicrobial effects of granulysin from pH 6 to pH 4.5. The ability of
granulysin to lyse liposomes was similarly pH dependent
(12).
|
C). At 10 mM salt, granulysin exhibited the highest level
of killing, with a 6-log decrease in CFU with 2.5 µM granulysin. At
50 mM salt the level of killing decreased by a factor of 10. By 100 mM
salt there was only a 1.5 log decrease at the highest concentration
tested, 25 µM granulysin. In summary, granulysin effectively killed
E. coli at 120 min and optimally did so at neutral pH and
low ionic strength. These data are reminiscent of the effects of pH and
ionic strength on defensins and suggest that electrostatic interactions
are the predominant determinants of granulysin activity. Evaluation of chemically modified granulysin
We therefore reasoned that granulysin was acting at least in part through electrostatic interaction of its cationic amino acid side chains with the negatively charged bacterial cell membrane. Because 17.5% of the protein is composed of arginine residues (13 of 74 residues), and 5.5% are lysine residues (4 of 74 residues), granulysin is a highly basic protein with a calculated pI of 10.8. To establish the role the basic amino acids in the antimicrobial activity, we altered both arginine and lysine residues with specific chemical modifiers. To verify the specificity of the modification, amoebapore A was also studied as a control. Amoebapore A has been shown previously to have antimicrobial activity against a wide range of microbes, and modification of its lysines inhibited binding to synthetic phospholipid vesicles. Unlike granulysin, the amoebapore protein does not contain any arginines.
Modification of the arginine residues with BAD significantly decreased
the antimicrobial activity of granulysin and the three active peptides
against E. coli, but had no effect on the activity of
amoebapore A (Fig. 4A). In
contrast, modification of the lysine residues with CAH had no effect on
the antimicrobial activity of granulysin, but significantly inhibited
the antimicrobial activity of amoebapore A (Fig. 4B). To
verify that granulysin was modified, MALDI-TOF mass spectroscopy was
performed and showed an increase in the mass of the modified granulysin
with BAD, but not CAH, indicating that the lysines were not available
for modification (data not shown) and therefore may not have a role in
the antimicrobial activity. Together, these data point to a
differential role of arginine and lysine residues in the antimicrobial
activity of granulysin and amoebapore A.
|
C). Treatment of
granulysin with increasing levels of BAD completely inhibited the
ability of granulysin to bind to E. coli (Fig. 4C). This decrease in Ab recognition as observed by Western
blot analysis was not due to modification of the Ab epitope, because
modified granulysin could be detected independently of the
concentration of the BAD used (data not shown). BAD did not alter the
secondary structure of granulysin as judged by circular dichroism
studies (data not shown). Additionally, modification of the granulysin
with CAH did not inhibit the interaction of granulysin with E.
coli (Fig. 4D). Thus, the ability of granulysin to
directly interact with bacteria is dependent on its positively charged
arginine residues. We also tested whether BAD-treated granulysin would
bind to M. tuberculosis (Fig. 4E). We found that
BAD pretreatment of granulysin inhibited granulysin binding to M.
tuberculosis in a dose-dependent manner. These data provide
evidence the positively charged arginine residues of granulysin are
required for binding to bacterial membranes. Granulysin increases membrane permeability of E. coli
We examined whether the binding of granulysin to the surface of
bacteria leads to an increase in membrane permeability. Membrane
permeability was measured in a permease-deficient strain of E.
coli by the accessibility of cytosolic ß-galactosidase to its
membrane-impermeant substrate, ONPG. Granulysin increased membrane
permeability in a dose- and time-dependent manner (Fig. 5A). In contrast, amoebapore A
had no effect on membrane permeability. The known membranolytic
peptide, melittin, is shown as a positive control.
|
B). The addition of modifiers
alone to this assay system did not elicit a response (data not shown).
We also tested the control protein amoebapore A, which killed E.
coli in a manner dependent on its lysine residues. However,
amoebapore A did not permeabilize the bacterial membrane (Fig. 6A), indicating that while
these two proteins both interact by electrostatic mechanisms with
bacteria, they eventually cause cell death through different
mechanisms.
|
To directly visualize the effect of granulysin on M.
tuberculosis, granulysin was incubated with the bacteria for
varying incubation times and viewed by transmission electron
microscopy. Fig. 6 shows an untreated bacterium as well as that treated
with granulysin for 24 h (middle panel) and 48 h
(bottom panel). Compared with the untreated control, the
granulysin-treated mycobacterium exhibited cell membrane and cell wall
separation similar to that observed in osmotic lysis. We speculate that
these data are consistent with the possibility that granulysin is
creating pores in the bacterial membrane or alternatively perturbing
the membrane releasing the cytoplasm. A rapid release of the cytoplasm
may lead to the separation of the membranes and eventually cell
death.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Amino acid sequence comparison indicates that granulysin is a member of the SAPLIP family of lipid binding proteins. Within this family, granulysin is most similar to NK lysin (43% identify and 67% similarity), a porcine granule protein with antibacterial activity (16). Other SAPLIP family members include the amoebapores, antibacterial peptides that amoebas use to kill bacterial prey (9). To determine the minimum active portion of the granulysin molecule retaining antimicrobial activity, peptides were synthesized to encompass the domains of structural homology with other members of this antimicrobial protein family. Although we cannot be certain that these peptides assume the same structure and function as their corresponding segments do in the intact protein, three of the peptides tested retained significant antimicrobial activity and were found to have a highly cationic charge, with contributions from both arginine and lysine residues. Similar to other antimicrobial peptides, granulysin and these constituent peptides had optimal activity at neutral to slightly basic pH and low ionic strength buffers, suggesting an essential role for an ionic interaction between the positively charged arginine residues of granulysin and the negatively charged bacterial membrane (17, 18). Chemical modification of the arginine residues with BAD not only inhibited the antimicrobial activity of granulysin, but also interfered with the protein interaction with the bacterial cell wall or membrane. Previous studies indicate the role of the negatively charged bacterial cell membrane with cationic antimicrobial peptides (19). Taken together, these data indicate that ionic interactions between granulysin and bacterial membranes are required for optimal antimicrobial activity.
When its primary structure is modeled on the related protein porcine
NK-lysin (20), granulysin is predicted to form a four
helical bundle. Based on computer modeling and secondary structural
analysis, the putative helical conformation of granulysin displays
hydrophobic residues on one surface and hydrophilic residues on the
opposite surface (data not shown). Such a structure is common among
proteins and peptides that interact with the phospholipid bilayers of
cell membranes. Circular dichroic analysis from three separate
preparations of recombinant granulysin indicates a high helical
content of 49%. Additionally, we have shown that the specific
three-dimensional structure of granulysin is required for its
antimicrobial activity, because denaturation completely abolished its
ability to kill E. coli. Studies of synthetic peptides
corresponding to selected regions of granulysin suggest that one or
more helical segments may be required for activity. The three
peptides of granulysin that retained antimicrobial activity were
predicted to conform to the helix-loop-helix motif. Similarly, a
previous study found that synthetic peptides of granulysin conforming
to a helix-loop-helix region spanning the second and third helices were
anti-mycobacterial (21). The helix-loop-helix
structure of the antimicrobial peptides of granulysin is similar to the
structure of cecropins, antimicrobial peptides present in insects
(22, 23).
Our data indicate that granulysin kills bacteria by perturbing the cell membrane. When granulysin was incubated in the presence of E. coli and a substrate for the cytoplasmic ß-galactosidase, ONPG, granulysin permeabilized the membrane and allowed the cytoplasmic enzyme to hydrolyze the normally excluded substrate. Transmission electron microscopy of granulysin-treated M. tuberculosis demonstrated the separation of the outer cell wall and membrane from the cytoplasm and the accumulation of electron-lucent material, presumably water, in the cytoplasm. Previous analysis of granulysin-treated bacteria by scanning electron microscopy indicated distortions in the outer cell wall and membrane of the pathogen (5). Together, these data are consistent with incorporation of granulysin into the membrane, perhaps forming pores in the bacterial membrane and disrupting the osmotic gradient between the cytoplasm and the external environment. Alternatively, the incorporation of granulysin into the bacterial membrane could lead to outer membrane expansion and disruption of the barrier to the extracellular environment.
Two types of membrane-active proteins have been shown to kill or lyse a
broad range of targets. These two species of proteins include 1) small
peptides (typically 16–40 aa), e.g., magainins, cecropins, defensins,
protegrins, and tachyplesins (8, 17, 24, 25); and 2)
larger proteins with multiple helical domains, e.g., amoebapores,
porcine NK-lysin, and human granulysin (16, 26). All these
proteins are cationic amphipathic species whose basic residues are
thought to interact with the negatively charged membrane. Yet the
mechanisms by which two of these proteins, granulysin and amoebapores,
kill pathogens appear different. Although both are positively charged
molecules, the antimicrobial activity of granulysin was dependent on
arginine residues, while that of amoebapores was dependent on lysine
residues. The permeabilization experiments indicate that granulysin can
perturb both the inner and outer bacterial membranes, but amoebapore A
had no measurable effect on the E. coli inner membrane.
Over a billion years of evolution, the conformation of SAPLIP family members has been preserved, from the amoebapores of amoebas to granulysin of humans (2, 3). This evolutionary change coincides with the development of the immune system; amoebapores represent a basic form of innate immunity that amoebas use to prevent growth of bacterial prey, whereas granulysin is secreted by human T cells, thus integrating this form of antimicrobial defense into the adaptive immune response. Amoebapores kill bacteria and are cytotoxic to human cells, whereas granulysin kills tumor cells, but is relatively inactive against normal human cells, requiring perforin as a cofactor to enter cells and to kill intracellular pathogens. Because of its low toxicity to normal host cells, granulysin is an attractive template for the development of novel antimicrobial therapeutics to combat the emergence of drug-resistant pathogens.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert L. Modlin, Division of Dermatology, University of California, 52-121 CHS, 10833 Le Conte Avenue, Los Angeles, CA 90095.
3 Abbreviations used in this paper: SAPLIP, saposin-like protein; TSB, trypticase soy broth; BAD, butanedione; CAH, citraconic anhydride; ONPG, o-nitrophenyl-ß-D-galactoside; NMR, nuclear magnetic resonance.
Received for publication May 1, 2000. Accepted for publication September 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-helix bundle motif of cytolytic peptides from distantly related organisms. Fold. Des. 2:47.[Medline]
This article has been cited by other articles:
|
|
 |
|
  B. Rivas-Santiago, C. J. Serrano, and J. A. Enciso-Moreno Susceptibility to Infectious Diseases Based on Antimicrobial Peptide Production Infect. Immun., November 1, 2009; 77(11): 4690 - 4695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Hogg, G. C. Bowick, N. K. Herzog, M. W. Cloyd, and J. J. Endsley Induction of granulysin in CD8+ T cells by IL-21 and IL-15 is suppressed by human immunodeficiency virus-1 J. Leukoc. Biol., November 1, 2009; 86(5): 1191 - 1203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Palm-Apergi, A. Lorents, K. Padari, M. Pooga, and M. Hallbrink The membrane repair response masks membrane disturbances caused by cell-penetrating peptide uptake FASEB J, January 1, 2009; 23(1): 214 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. G. da Silva, D. Unks, S.-c. Lyu, J. Ma, R. Zbozien-Pacamaj, X. Chen, A. M. Krensky, and C. Clayberger In vitro and in vivo antimicrobial activity of granulysin-derived peptides against Vibrio cholerae J. Antimicrob. Chemother., May 1, 2008; 61(5): 1103 - 1109. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Andersson, A. Samarina, J. Fink, S. Rahman, and S. Grundstrom Impaired Expression of Perforin and Granulysin in CD8+ T Cells at the Site of Infection in Human Chronic Pulmonary Tuberculosis Infect. Immun., November 1, 2007; 75(11): 5210 - 5222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. F. Zheng, L. L. Ma, G. J. Jones, M. J. Gill, A. M. Krensky, P. Kubes, and C. H. Mody Cytotoxic CD4+ T cells use granulysin to kill Cryptococcus neoformans, and activation of this pathway is defective in HIV patients Blood, March 1, 2007; 109(5): 2049 - 2057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. P. Huang, S.-C. Lyu, C. Clayberger, and A. M. Krensky Granulysin-Mediated Tumor Rejection in Transgenic Mice J. Immunol., January 1, 2007; 178(1): 77 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Andra, T. Gutsmann, P. Garidel, and K. Brandenburg Invited review: Mechanisms of endotoxin neutralization by synthetic cationic compounds Innate Immunity, October 1, 2006; 12(5): 261 - 277. [Abstract] [PDF]
|
|
|
|
|
|
M. A. Ryan, H. T. Akinbi, A. G. Serrano, J. Perez-Gil, H. Wu, F. X. McCormack, and T. E. Weaver Antimicrobial Activity of Native and Synthetic Surfactant Protein B Peptides J. Immunol., January 1, 2006; 176(1): 416 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Stegelmann, M. Bastian, K. Swoboda, R. Bhat, V. Kiessler, A. M. Krensky, M. Roellinghoff, R. L. Modlin, and S. Stenger Coordinate Expression of CC Chemokine Ligand 5, Granulysin, and Perforin in CD8+ T Cells Provides a Host Defense Mechanism against Mycobacterium tuberculosis J. Immunol., December 1, 2005; 175(11): 7474 - 7483. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Feng, N. Huang, Q. Wu, and B. Wang HMGN2: a novel antimicrobial effector molecule of human mononuclear leukocytes? J. Leukoc. Biol., November 1, 2005; 78(5): 1136 - 1141. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. A. Linde, S. Grundstrom, E. Nordling, E. Refai, P. J. Brennan, and M. Andersson Conserved Structure and Function in the Granulysin and NK-Lysin Peptide Family Infect. Immun., October 1, 2005; 73(10): 6332 - 6339. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Oliaro, S. Dudal, J. Liautard, J.-B. Andrault, J.-P. Liautard, and V. Lafont V{gamma}9V{delta}2 T cells use a combination of mechanisms to limit the spread of the pathogenic bacteria Brucella J. Leukoc. Biol., May 1, 2005; 77(5): 652 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Deng, S. Chen, Q. Li, S.-c. Lyu, C. Clayberger, and A. M. Krensky Granulysin, a Cytolytic Molecule, Is Also a Chemoattractant and Proinflammatory Activator J. Immunol., May 1, 2005; 174(9): 5243 - 5248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Walch, E. Eppler, C. Dumrese, H. Barman, P. Groscurth, and U. Ziegler Uptake of Granulysin via Lipid Rafts Leads to Lysis of Intracellular Listeria innocua J. Immunol., April 1, 2005; 174(7): 4220 - 4227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Li, C. Dong, A. Deng, M. Katsumata, A. Nakadai, T. Kawada, S. Okada, C. Clayberger, and A. M. Krensky Hemolysis of Erythrocytes by Granulysin-Derived Peptides but Not by Granulysin Antimicrob. Agents Chemother., January 1, 2005; 49(1): 388 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Levy Antimicrobial proteins and peptides: anti-infective molecules of mammalian leukocytes J. Leukoc. Biol., November 1, 2004; 76(5): 909 - 925. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. L. Ma, C. L. C. Wang, G. G. Neely, S. Epelman, A. M. Krensky, and C. H. Mody NK Cells Use Perforin Rather than Granulysin for Anticryptococcal Activity J. Immunol., September 1, 2004; 173(5): 3357 - 3365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Endsley, J. L. Furrer, M. A. Endsley, M. A. McIntosh, A. C. Maue, W. R. Waters, D. R. Lee, and D. M. Estes Characterization of Bovine Homologues of Granulysin and NK-lysin J. Immunol., August 15, 2004; 173(4): 2607 - 2614. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Lauterbach, K. M. Kerksiek, D. H. Busch, E. Berto, A. Bozac, P. Mavromara, R. Manservigi, A. L. Epstein, P. Marconi, and T. Brocker Protection from Bacterial Infection by a Single Vaccination with Replication-Deficient Mutant Herpes Simplex Virus Type 1 J. Virol., April 15, 2004; 78(8): 4020 - 4028. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Lewinsohn, A. S. Heinzel, J. M. Gardner, L. Zhu, M. R. Alderson, and D. M. Lewinsohn Mycobacterium tuberculosis-specific CD8+ T Cells Preferentially Recognize Heavily Infected Cells Am. J. Respir. Crit. Care Med., December 1, 2003; 168(11): 1346 - 1352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Jacobs, H. Bruhn, I. Gaworski, B. Fleischer, and M. Leippe NK-Lysin and Its Shortened Analog NK-2 Exhibit Potent Activities against Trypanosoma cruzi Antimicrob. Agents Chemother., February 1, 2003; 47(2): 607 - 613. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Borelli, F. Vita, S. Shankar, M. R. Soranzo, E. Banfi, G. Scialino, C. Brochetta, and G. Zabucchi Human Eosinophil Peroxidase Induces Surface Alteration, Killing, and Lysis of Mycobacterium tuberculosis Infect. Immun., February 1, 2003; 71(2): 605 - 613. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. L. Ma, J. C. L. Spurrell, J. F. Wang, G. G. Neely, S. Epelman, A. M. Krensky, and C. H. Mody CD8 T Cell-Mediated Killing of Cryptococcus neoformans Requires Granulysin and Is Dependent on CD4 T Cells and IL-15 J. Immunol., November 15, 2002; 169(10): 5787 - 5795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. D. Chan, J. Chan, and N. W. Schluger What is the Role of Nitric Oxide in Murine and Human Host Defense against Tuberculosis? . Current Knowledge Am. J. Respir. Cell Mol. Biol., November 1, 2001; 25(5): 606 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Merlo, D. Saverino, C. Tenca, C. E. Grossi, S. Bruno, and E. Ciccone CD85/LIR-1/ILT2 and CD152 (Cytotoxic T Lymphocyte Antigen 4) Inhibitory Molecules Down-Regulate the Cytolytic Activity of Human CD4+ T-Cell Clones Specific for Mycobacterium tuberculosis Infect. Immun., October 1, 2001; 69(10): 6022 - 6029. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ogawa, K. Tanaka, A. Ishii, Y. Nakamura, S. Kondo, K. Sugamura, S. Takano, M. Nakamura, and K. Nagata A Novel Serum Protein That Is Selectively Produced by Cytotoxic Lymphocytes J. Immunol., May 15, 2001; 166(10): 6404 - 6412. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
