of January 22, 2019.This information is current as

Cytotoxic Lymphocyte FunctionsGranzymes C and/or F Are Important for Granzyme B and the Downstream

Timothy J. LeyandXuefang Cao, Rajesh Behl, Jane A. Ratner, Zhi Hong Lu

Paula A. Revell, William J. Grossman, Dori A. Thomas,

http://www.jimmunol.org/content/174/4/2124doi: 10.4049/jimmunol.174.4.2124

2005; 174:2124-2131; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/174/4/2124.full#ref-list-1

, 19 of which you can access for free at: cites 35 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2005 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on January 22, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

Granzyme B and the Downstream Granzymes C and/or F AreImportant for Cytotoxic Lymphocyte Functions1

Paula A. Revell,2* William J. Grossman,2,3† Dori A. Thomas,2,4* Xuefang Cao,2*Rajesh Behl,2,5* Jane A. Ratner,* Zhi Hong Lu,* and Timothy J. Ley6*

Although the functions of granzyme A (GzmA) and GzmB are well-defined, a number of orphan granzymes of unknown functionare also expressed in cytotoxic lymphocytes. Previously, we showed that a targeted loss-of-function mutation for GzmB wasassociated with reduced expression of several downstream orphan granzyme genes in the lymphokine-activated killer cell com-partment. To determine whether this was caused by the retained phosphoglycerate kinase I gene promoter (PGK-neo) cassette inthe GzmB gene, we retargeted the GzmB gene with a LoxP-flanked PGK-neo cassette, then removed the cassette in embryonic stemcells by transiently expressing Cre recombinase. Mice homozygous for the GzmB null mutation containing the PGK-neo cassette(GzmB�/�/�PGK-neo) displayed reduced expression of the closely linked GzmC and F genes in their MLR-derived CTLs andlymphokine-activated killer cells; removal of the PGK-neo cassette (GzmB�/�/�PGK-neo) restored the expression of both genes.Cytotoxic lymphocytes derived from mice with the retained PGK-neo cassette (GzmB�/�/�PGK-neo) had a more severe cytotoxicdefect than those deficient for GzmB only (GzmB�/�/�PGK-neo). Similarly, GzmB�/�/�PGK-neo mice displayed a defect in theallogeneic clearance of P815 tumor cells, whereas GzmB�/�/�PGK-neo mice did not. These results suggest that the retainedPGK-neo cassette in the GzmB gene causes a knockdown of GzmC and F expression, and also suggest that these granzymes arerelevant for the function of cytotoxic lymphocytes in vitro and in vivo. The Journal of Immunology, 2005, 174: 2124–2131.

C ytotoxic lymphocytes can use multiple pathways to in-duce target cell death, including secretory mechanisms(e.g., TNF-� and IFN-�) and contact-dependent mecha-

nisms, including the granule exocytosis pathway and the Fas path-way (1–3). In several studies the killing of tumor cells has beenshown to be dependent on the granule exocytosis pathway, butvariable results have been obtained with different tumor targets indifferent laboratories, suggesting that more than one mechanism isrelevant, and that different mechanisms may be important for dif-ferent tumors (1, 4–11).

The secretory granules of cytotoxic lymphocytes (CTL, NK, andLAK cells) migrate along microtubules to the site of contact be-tween the effector and target cells, where a tight immunologicsynapse is formed (1–3). There, the contents of the cytotoxic gran-ules are secreted into the small intercellular space, where perforincan polymerize in the presence of calcium to form membrane

perturbations that may be relevant for the entry of other granuleproteins, including granzymes. Perforin also appears to play a criticalrole in the trafficking of granzymes once they enter the target cell.Granzymes, a class of neutral serine proteases, are thought to inducetarget cell death by cleaving critical intracellular substrates (1, 2).Granzyme A (GzmA)7 and B have very different substrate specifici-ties, and appear to cleave nonoverlapping sets of target cell proteins(1, 2, 12, 13). Although GzmB induces classical apoptosis associatedwith mitochondrial depolarization, nuclear collapse and fragmenta-tion, and activation of caspase-activated DNase, GzmA induces celldeath that has many of the features of apoptosis, but is clearly differentfrom death induced by GzmB (1, 2, 14–16). The apparent nonredun-dancy of these two granzymes might be explained in part by the factthat viruses and tumor cells sometimes express inhibitors of gran-zymes, presumably to thwart the ability of cytotoxic lymphocytes tokill these cells (1). Under these circumstances, alternative granzymeswith different specificities would be required to induce cell death.

In both humans and mice, a variety of additional granzymes ofunknown function (orphans) are expressed in cytotoxic lympho-cytes (17). In the mouse a large number of functional granzymegenes are found downstream from GzmB in a locus commonlyknown as the GzmB gene cluster. GzmC lies �24 kb downstreamfrom GzmB and is most closely related to human GzmH, which is theonly orphan granzyme gene found downstream from human GzmB.In the mouse, however, several additional granzyme genes lie down-stream from GzmC, including (5�-�3�) GzmF, G, L, N, D, and E,followed by cathepsin G (a promyelocyte-specific serine proteinase ofsimilar structure) and several mast cell chymases (17).

In our initial GzmB knockout mouse, we removed the codingsequences from exon 1 and part of intron 1 and replaced these

*Division of Oncology, Departments of Internal Medicine and Genetics, SitemanCancer Center, Washington University School of Medicine, and †Division of Hema-tology/Oncology, Department of Pediatrics, St. Louis Children’s Hospital, St. Louis,MO 63110

Received for publication September 3, 2004. Accepted for publication October24, 2004.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by grants from the National Institutes of Health(5T32AI07163 (to P.A.R.) and DK49786 (to T.J.L.)) and the Hope Street Kids Foun-dation (to W.J.G.).2 P.A.R., W.J.G., D.A.T., X.C., and R.B. contributed equally to this work.3 Current address: Division of Hematology/Oncology, Department of Pediatrics,Medical College of Wisconsin, Milwaukee, WI 53226.4 Current address: Department of Cancer Biology and Genetics, Memorial Sloan Ket-tering Cancer Center, New York, NY 10021.5 Current address: Center for Human Genetics, Bangalore 560066, India.6 Address correspondence and reprint requests to Dr. Timothy J. Ley, WashingtonUniversity School of Medicine, 660 South Euclid Avenue, Campus Box 8007, St.Louis, MO 63110. E-mail address: tley@im.wustl.edu

7 Abbreviations used in this paper: Gzm, granzyme; 7-AAD, 7-aminoactinomycin;ES, embryonic stem cell; FloKA, flow-based killing assay; LAK, lymphokine-acti-vated killer; PGK-neo, phosphoglycerate kinase I gene promoter; qRT-PCR, quanti-tative RT-PCR; rh, recombinant human; [125I]UdR, [125I]5-iodo-2[prime]-deoxyuri-dine; WT, wild type; LCR, locus control region.

The Journal of Immunology

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

sequences with a functional phosphoglycerate kinase I gene pro-moter (PGK)-neo cassette (18). We later learned that mixed strainmice containing this mutation display significantly reduced the ex-pression of several downstream granzyme genes (C, F, and D) inthe lymphokine-activated killer (LAK) cell compartment (19). Todetermine whether this was a neighborhood knockdown effectcaused by the retained PGK-neo cassette, we retargeted the GzmBgene with a LoxP-flanked PGK-neo cassette, then removed it fromthe targeted ES cells with Cre recombinase. Removal of the PGK-neo cassette restored the expression of the downstream granzymes,which allowed us to compare the functions of cytotoxic lympho-cytes deficient for GzmB only (GzmB�/�/�PGK-neo) or GzmBplus reduced levels of GzmC and F (GzmB�/�/�PGK-neo). Toour surprise, cytotoxic lymphocytes deficient for GzmB, C, and Fdisplayed a more striking defect in killing than cytotoxic lympho-cytes deficient for GzmB only; this was true not only for LAKcells, but also for CTL derived from MLR. To validate these re-sults in vivo, we assessed the ability of these mice to clear anallogeneic tumor (P815, H-2Kd) and again found that mice defi-cient for GzmB, C, and F have a more significant clearance defectthan those deficient for GzmB only. In sum, these results haveshown that the retained PGK-neo cassette in the GzmB gene doescause a neighborhood effect, and the reduced expression of GzmCand F is relevant for the functions of CTLs and LAK cells in vitroand in vivo.

Materials and MethodsConstruction of the GzmB targeting vector

A 4.16-kb EcoRI fragment containing the entire GzmB locus was sub-cloned into pUC19 to generate the backbone of the GzmB targeting vector,similar to that previously described (18). A 350-bp AvrII fragment con-taining the initiation and start sites and the entire first exon (and part of thefirst intron) of the GzmB gene was replaced with a 1.6-kb NotI fragmentcontaining a LoxP-flanked neomycin phosphotransferase gene driven byPGK-neo (20). The resulting targeting vector contained 1.03 kb of genomicDNA from the GzmB 5� flank upstream from the PGK-neo insertion and2.75 kb of genomic DNA containing exons 2–5 downstream from PGK-neo. The targeting vector was linearized with SalI and gel-purified beforeelectroporation into Rw4 embryonic stem cells (ES; 129/SvJ).

Generation of 129/SvJ mice containing the GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo mutations

GzmB�/�/�PGK-neo ES were generated by electroporation of the GzmBtargeting vector in Rw4 ES cells, neomycin-resistant clones were ex-panded, and homologous recombination events were identified by Southernblot analysis using an external probe (data not shown) (18). Correct tar-geting was confirmed by Southern blot analysis using an internal probeafter EcoRI digestion (Fig. 1). ES clones with specific homologous recom-bination were injected into C57BL/6 blastocysts that were implanted intopseudopregnant Swiss-Webster mice. Male chimeras were bred with 129/SvJ female mice, and germline transmission of the mutant alleles wasconfirmed by Southern blot analysis. Heterozygous animals were inter-crossed to generate homozygous GzmB�/�/�PGK-neo deficient mice.GzmB�/�/�PGK-neo ES clones were generated by transiently transfectingGzmB�/�/�PGK-neo ES clones with a Turbo-Cre-expressing plasmid aspreviously described (20, 21), resulting in the recombination of LoxP sitesand the deletion of the internal PGK-neo cassette. Resultant GzmB�/�/�PGK-neo clones were screened for neomycin sensitivity, and deletion ofthe PGK-neo cassette was confirmed by Southern blot analysis (Fig. 1).Mutant mice were generated exactly as described above. Heterozygousanimals were intercrossed to generate homozygous GzmB�/�/�PGK-neodeficient mice. Both mutant strains were maintained as homozygous breed-ing colonies in the 129/SvJ background.

MLR and LAK cell preparations for in vitro cytotoxicity andflow cytometry assays

MLR and LAK cell preparations were performed as previously described(18, 22). For MLR preparations, stimulator (BALB/c wild-type (WT);H-2Kd) and effector (129/SvJ WT, GzmB�/�/�PGK-neo, and GzmB�/�/�PGK-neo; all H-2Kb) spleens were harvested into PBS, and single cell

suspensions were made by gently crushing spleens between frosted micro-scope slides. Red cells were lysed in ACK buffer (154 mM NH4Cl, 10 mMKHCO3, and 0.1 mM EDTA; sterile-filtered) and washed twice in PBS.The cell suspension from each BALB/c spleen was resuspended in 5 ml ofcomplete K10 medium (RPMI 1640, 10% heat-inactivated FCS, 10 mMHEPES, 1% nonessential amino acids, 1% sodium pyruvate, 1% L-glu-tamine, 1� penicillin/streptomycin, 0.57 �M �-ME) and 50 U/ml recom-binant human IL-2 (rhIL-2) and irradiated (2000 cGy). The cell suspen-sions from WT, GzmB�/�/�PGK-neo, and GzmB�/�/�PGK-neo spleenswere each resuspended in 40 ml of complete K10 medium and 50 U/mlrhIL-2. Irradiated BALB/c stimulators and 129/SvJ, GzmB�/�/�PGK-neo,or GzmB�/�/�PGK-neo responders were combined into 45 ml of com-plete K10 medium and 50 U/ml rhIL-2. Samples were mixed, plated inthree 100-mm non-tissue culture-treated plates, and incubated for 4 days.Before FloKA, the mononuclear cells were purified over Ficoll (Sigma-Aldrich), washed with PBS, and resuspended in complete K10 medium and50 U/ml rhIL-2.

Quantitative real-time RT-PCR (qRT-PCR)

Total RNA was isolated from 10 � 106 LAK cells or Ficoll-purified cellsfrom MLRs using the RNeasy Mini Kit (Qiagen). One microgram of totalRNA from each sample was reverse-transcribed into cDNA using the Taq-Man reverse transcriptase (ABI; Applied Biosystems) and analyzed byPCR. The relative mRNA abundance of each granzyme mRNA was mea-sured by the 7300 real-time PCR system from Applied Biosystems, usingprimer sets that were proven to amplify only the self granzyme cDNAusing the conditions described below. The primer sets used in these studieswere as follows: GzmA-specific primer set: forward, 5�-AGA CCG TATATG GCT CTA CT-3�; and reverse, 5�-CCC TCA CGT GTA TAT TCATC-3�; GzmB-specific primer set: forward, 5�-ATC AAG GAT CAG CAGCCT GA-3�; and reverse, 5�-TGA TGT CAT TGG AGA ATG TCT-3�;GzmC-specific primer set: forward, 5�-ATG AGT TTC TGA AAG TTGGTG-3�; and reverse, 5�-TCA TTA GAA CGG TCA TCA GG-3�; GzmF-specific primer set: forward, 5�-TGA GGT TTG TGA AAG ATA ATG-3�;and reverse, 5�-TCA CTG GTG TTG TCC TTA TC-3�; GzmG-specificprimer set: forward, 5�-GTT CAT TAA GTC TGT GGA TAT CG-3�; andreverse, 5�-GTC TTG GAA TAG GTG TAC CAG-3�; GzmD-specificprimer set: forward, 5�-AAA CAG CTC TGT CCA AAG CTC-3�; andreverse, 5�-CAA ATC TCT GTG GTC TCA GTG-3�; and GzmE-specificprimer set: forward, 5�-CTG CTC ACT GCA GGA ACA GGA-3�; andreverse, 5�-CAA ATC TCT GTG GTC TCA GTG-3�. The qRT-PCRs wereconducted in the presence of the DNA intercalating dye SYBR Green. PCRconditions used for all primer sets were as follows: 95°C hot start for 10min, followed by 40 amplification cycles of 95°C for 30 s (denaturing),59°C for 40 s (annealing), and 72°C for 40 s (extension). A PCR ampli-fication profile was derived by recording the SYBR Green fluorescenceintensity, which was in linear relation to the amount of formed PCR prod-uct (as a function of PCR cycle number). Standard curves were generatedby plotting the PCR cycle number at which a reaction entered exponentialamplification vs the amount of input DNA of plasmids containing thecDNA of each granzyme gene. The standard curves were then used for adetermination of sample template concentration.

Western blot analysis

Ficoll-purified MLR and LAK cell lysates were prepared and analyzedusing standard Western blotting techniques as previously described (16,22). Primary Abs included rabbit anti-mouse GzmA antiserum (MA2A)(16), rabbit anti-mouse GzmB antiserum (526B) (22), rabbit anti-mouseGzmC antiserum (428A) (23), and goat anti-mouse �-actin-HRP antiserum(Santa Cruz Biotechnology). Primary Abs were detected using secondarygoat anti-rabbit HRP or horse anti-goat HRP with standard chemilumines-cence procedures (Amersham Biosciences). Recombinant murine GzmA,B, and C were purified as previously described (16, 23, 24). One hundrednanograms of each recombinant granzyme was used to demonstrate anti-serum specificity. Twenty-five to 50 �g of MLR or LAK cell protein wasused in each lane.

Animals

WT C57BL/6J, BALB/c, and 129/SvJ mice were obtained from The Jack-son Laboratory. GzmB�/�/�PGK-neo- and GzmB�/�/�PGK-neo mice inthe 129/SvJ background were generated as described above. The gld/gldand perforin�/� mice in C57BL/6 backgrounds were previously described(25). GzmB�/�/�PGK-neo mice in the C57BL/6J background were gen-erated by backcrossing our previously described mutant mice (18) toC57BL/6J mice for 10 generations. All mice were bred and kept in patho-gen-free housing in accordance with Washington University School of

2125The Journal of Immunology

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

Medicine animal care guidelines, using protocols approved by the animalstudies committee.

Intracellular granzyme staining and flow cytometry

Unstimulated splenocytes or MLR cells were stained for surface markersand intracellular GzmB expression using a GzmB-specific Ab as previ-ously described (26). Briefly, 1 � 106 cells were first labeled with fluo-rochrome-conjugated Abs against cell surface markers (anti-mouse CD4,CD8, and DX5; BD Pharmingen). Samples were fixed and permeabilized(Cytofix/Cytoperm; BD Pharmingen), then stained with primary conju-gated anti-GzmB Ab (GB12; Caltag Laboratories) diluted at 1/400 in stain-ing buffer. During all steps of staining, permeabilization, and washing,anti-murine Fc�RII/III-blocking Ab (0.1 �g/�l, final concentration; BDPharmingen) and human albumin (1%; Aventis Behring) were used toblock nonspecific FcR Ab binding. Samples were analyzed on a FACScan(BD Biosciences). All FACScans depicted are representative of four ormore independent experiments.

Flow-based killing assay (FloKA) and [125I]5-iodo-2[prime]-deoxyuridine ([125I]UdR) release assay

CTL and LAK cells were assessed by FloKA as previously described (26).Briefly, allogeneic (P815 and TA3-H-2Kd; American Type Culture Col-lection) or MHC class I-deficient (RMAS and YAC-1; American TypeCulture Collection) murine target cells were washed with PBS and labeledwith 125 nM (final concentration) CFSE (Molecular Probes). Labeling re-actions were stopped with complete K10 medium. Labeled cells wereadded to 96-well, V-bottom, tissue culture-treated plates (Corning Glass)along with the indicated effector cells (CTLs from MLRs, or LAK cells;see above) in complete medium containing 50 U/ml rhIL-2. E:T cell ratiosvaried from 5:1 to 20:1 in different experiments. Immediately before anal-ysis, 1 �g/ml (final concentration) of 7-aminoactinomycin D (7-AAD; Cal-biochem) was added to each sample. Control samples (targets only) hadeffectors added immediately before analysis at the indicated E:T cell ratio.[125I]UdR release assays were performed as previously described (18, 27).All experiments examining WT vs Gzm-deficient mice were performed inparallel. All samples were analyzed in duplicate. Data presented in graphicform are the combination of three or more individual experiments. The datashown in individual FACS scans are representative of three or more indi-vidual experiments. Statistical analyses were performed by one-way-ANOVAwith Bonferroni post-test analysis or Student’s t test, using PRISM version3.0a for Macintosh (GraphPad).

Allogeneic tumor cell clearance in vivo

All mice were sublethally irradiated (400 cGy) on day �1 and subse-quently injected i.v. in the lateral tail vein with 1 � 107 P815 mastocytomacells on day 0. WT, GzmB�/�/�PGK-neo, perforin�/�, and gld/gld mice(all on the C57BL/6J background) were monitored over a 60-day period.WT, GzmB�/�/�PGK-neo, and GzmB�/�/�PGK-neo mice (on the 129/SvJ background) were similarly injected and monitored.

ResultsCreation of mice with targeted mutations in the GzmB gene

We generated GzmB�/�/�PGK-neo mice using the targeting vec-tor shown in Fig. 1A (top panel). The targeted GzmB allele con-tained a 350-bp deletion that starts in the 5� untranslated region of thegene, and removes all the coding sequences of exon 1, and part ofintron 1 (18). Targeted ES cells containing the mutant GzmB�/�/�PGK-neo allele contained the PGK-neo selection cassette flankedby two LoxP sites. We generated GzmB�/�/�PGK-neo ES cellsby transiently transfecting the GzmB�/�/�PGK-neo ES cloneswith pTurbo-Cre, which resulted in recombination of the two LoxPsites and deletion of the internal PGK-neo selection cassette (Fig.1A, top panel). Using Southern blot analysis with internal andexternal probes along with new restriction sites introduced by thetargeting vector, we demonstrated correct targeting of the vector intwo independent Rw4 (129/SvJ) ES clones; precise deletion of thePGK-neo cassette in the targeted clones was likewise demon-strated with Southern blot analysis. The correctly targeted ESclones were injected into C57BL/6 blastocysts, which were thenimplanted into pseudopregnant female recipients. All ES clonesgave rise to chimeric male mice that were then directly mated to

129/SvJ mice, and the mutations were transmitted through the germ-line. The expected EcoRI fragments were detected in the tail DNA ofmice heterozygous for the mutations (4.4 kb GzmB�/�/�PGK-neo,2.8 kb GzmB�/�/�PGK-neo, and 4.2 kb WT) with Southern blotanalysis using an internal probe (Fig. 1, A and B) and also an externalprobe (data not shown). Heterozygous matings yielded offspring withthe expected Mendelian frequencies. Homozygous GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo mice were found to have nor-mal development and fertility.

Neighborhood effect on downstream granzyme gene expressioncaused by the retained PGK-neo cassette

We measured the expression of multiple granzyme genes in cellsderived from MLRs and LAK preparations from both GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo mice using qRT-PCR withprimer pairs specific for each granzyme gene (Fig. 2). All primerpairs were shown to amplify only the cloned self target cDNA inqRT-PCRs (data not shown). However, annealing efficiencies var-ied among the primer pairs, so comparisons of the absolute abun-dances of different granzyme mRNAs are not valid. We thereforenormalized the abundance of each granzyme mRNA to the leveldetected in WT LAK cells. In all experiments activations wereperformed concurrently with all three genotypes. Neither GzmBmutant strain had detectable GzmB mRNA expression in LAKcells or CTLs (Fig. 2). There was no significant difference in theexpression levels of GzmA in any of the mutant mice, because

FIGURE 1. Targeting strategy and screening. A, Targeting vector used forhomologous recombination containing the PGK-neo selection cassette flankedby two LoxP recombination sites. Recombination of the targeting vector withthe endogenous allele results in formation of GzmB�/�/�PGK-neo ES cells.Transfection of a Cre-expressing plasmid in GzmB�/�/�PGK-neo ES cellsresults in excision of the PGK-Neo selection cassette through recombination ofthe flanking LoxP sites and formation of GzmB�/�/�PGK-neo ES cells. E,EcoRI. Probe A was used for the Southern blot shown in B. B, Southern blotanalysis of EcoRI-digested tail DNA derived from WT mice or heterozygousmice containing the GzmB�/�/�PGK-neo or GzmB�/�/�PGK-neo muta-tions. The expected fragment sizes are shown.

2126 ORPHAN GRANZYME AND CTL

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

GzmA is located on a different chromosome than GzmB (17). TheqRT-PCR analysis of GzmC and F mRNAs in LAK cells derivedfrom GzmB�/�/�PGK-neo mice demonstrated that expressionwas reduced 5- to 6-fold compared with that in WT LAK cells, butwas normal in LAK cells derived from GzmB�/�/�PGK-neo mice(Fig. 2). GzmC and F expression in MLR-derived cells fromGzmB�/�/�PGK-neo mice was reduced compared with that inWT cells, but expression levels were very low, and the differencewas not statistically significant. GzmC and F expression inGzmB�/�/�PGK-neo CTLs was reproducibly increased comparedwith that in WT animals (Fig. 2). mRNAs for GzmD, E, and Gwere not detectably expressed in MLR-derived CTLs, but all weredetected in LAK cells and were not altered in GzmB�/�/�PGK-neo derived LAKs.

Examination of granzyme protein levels from MLR and LAKcell lysates yielded a similar expression profile (Fig. 3). MLR andLAK cell lysates from both GzmB�/�/�PGK-neo and GzmB�/

�/�PGK-neo mice lacked a detectable GzmB signal using West-ern blot analysis (Fig. 3, left and middle panels). LAK cells fromGzmB�/�/�PGK-neo mice had an �80% reduction in GzmC pro-tein expression compared with WT LAKs (Fig. 3, middle panels).In contrast, LAK cell lysates from GzmB�/�/�PGK-neo mice hadrestoration of GzmC protein expression to WT levels. Analysis ofMLR-derived protein lysates from GzmB�/�/�PGK-neo micedemonstrated GzmC protein expression levels that were elevated overthe GzmC protein levels found in wild-type CTLs (Fig. 3, left panels),

similar to those observed with mRNA analysis. GzmA protein ex-pression was equivalent in LAK cell lysates from WT and the twoGzmB knockout strains (Fig. 3, middle panels); GzmA protein couldnot be detected in day 4 MLR-derived lysates, probably because thisgene is activated later than GzmB and C (28). Recombinant murinegranzymes were used for each Western blot as a control for antiserumspecificity (Fig. 3, right panels), and each blot was analyzed for �-ac-tin protein levels as an internal control for protein loading. SpecificAbs for GzmD, E, F, and G are not yet available.

Normal lymphocyte subsets in GzmB-deficient mice

To ensure that the observed differences in granzyme mRNA andprotein levels between GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo mice were not due to skewed granzyme-expressingpopulations, we examined these mice for the main lymphocytepopulations known to express granzymes, namely CD4� andCD8� T cells and NK cells (DX5-positive cells). As shown inFig. 4A, no differences were found in the percentages of theselymphocyte subpopulations between WT mice and the twoGzmB-deficient mouse strains. Resting WT splenocytes do notcontain detectable amounts of GzmB (Fig. 4B), in contrast toresting human PBMCs, where 15–25% of resting CD8� T cellscontain detectable amounts of this enzyme (26).

We next examined the intracellular GzmB expression profilefrom WT MLR-derived cells using flow cytometry and an intra-cellular staining assay for GzmB (26). These studies demonstratedthat the majority of GzmB-expressing cells in MLRs are CD8� Tcells (Fig. 4C); however, a small percentage of CD4� T cells alsoexpress GzmB, a finding consistent with that of human PBMC(26). As expected, no GzmB intracellular expression was detectedin either CD4� or CD8� T cells in MLRs from GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo mice (Fig. 4C). The polyclonal Abagainst GzmC used in the Western blots in Fig. 3 was not usefulin flow-based studies; mAbs specific for murine GzmA or C arenot yet available. A small amount of nonspecific staining (presum-ably representing activated macrophages) was observed in MLR-derived cells from all mouse strains (Fig. 4C), which was not pre-vented by Fc�RII-III-blocking Abs.

Defects in cytotoxicity in vitro

To determine whether the two GzmB knockout strains have de-fects in cytotoxicity, we first examined CTLs generated from MLRfor their ability to kill allogeneic P815 or TA3 tumor cells (bothfrom an H-2Kd background) using a flow-based killing assay(FloKA). We found that GzmB�/�/�PGK-neo CTLs were signif-icantly less efficient at inducing target cell death than WT CTLs atall time points tested (representative data are shown in Fig. 5A, R1and R2 gates). The same cytotoxic defects were detected with both

FIGURE 2. Quantitative RT-PCR analysis of gran-zyme mRNA abundance in MLR and LAK cells. Quan-titative RT-PCR was performed on three independentMLR and three independent LAK cell preparations frommice of the described genotypes. All data for each specificgranzyme mRNA was normalized to the level of expres-sion detected in WT LAK cells. Because primer pairs an-nealed to their targets with different efficiencies, compar-ison of absolute mRNA abundance among differentgranzymes is not valid. Values shown are the mean � SD.�, p 0.05; ��, p 0.01; ���, p 0.001.

FIGURE 3. Western blot analysis of cell lysates made from LAK cellsand MLR-derived CTLs. Purified murine recombinant granzymes (rGzms)were used for specificity controls of Abs in each Western blot (top, middle,and bottom panels). Data shown are representative of three or more inde-pendent experiments.

2127The Journal of Immunology

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

P815 and TA3 targets, and the findings were consistent and highlyreproducible. The most striking defect was the inability of theseeffectors to generate the small, 7-AADlow�high cells that are lo-cated in the R1 gate, which probably represent cells in the latestages of apoptosis. CTLs derived from both Gzm-deficient strainsinduced an increased number of target cells in the R2 gate (largecells that are 7-AADhigh, probably representing cells with earlyapoptotic events), suggesting that the induction of target celldeath is delayed with GzmB-deficient effectors, a finding that cor-roborates earlier studies of CTL function with these mice (18).GzmB�/�/�PGK-neo generated CTLs demonstrated a less severe

cytotoxic defect than GzmB�/�/�PGK-neo CTLs (Figs. 5A and6A, R1 gates). A similar defect in the production of late stageapoptotic events was seen with LAK cells generated from bothGzm-deficient strains against two different NK-sensitive target celllines (RMAS and YAC-1 cells; Fig. 6B). The most severe cyto-toxic defect was observed with GzmB�/�/�PGK-neo LAK cells,with GzmB�/�/�PGK-neo LAKs again demonstrating an inter-mediate defect (Fig. 6B). Although the severity of the cytotoxicphenotype was partly dependent upon the target cells used, thetrends were the same for all target cells tested.

When we examined GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo LAKs for their ability to release dsDNA fragmentsusing classic [125I]UdR release assays, we detected no differencebetween the two Gzm-deficient strains regardless of time or E:Tcell ratio (Fig. 5B). CTLs from both Gzm-deficient strains dis-played a severe defect in [125I]UdR clearance at 2 h, but this defectwas not apparent after 8 h (Fig. 5B), similar to results previouslydescribed (18). No abnormalities were detected in 51Cr releasewith GzmB-deficient LAK cells, as expected (data not shown).

Defects in tumor clearance in vivo

To determine whether the cytotoxic defects detected in vitro wererelevant in vivo, we examined the clearance of P815 cells in sub-lethally irradiated (400 cGy) C57BL/6 mice deficient for perforin,GzmB�/�/�PGK-neo, or FasL (gld). The GzmB�/�/�PGK-neoand perforin-deficient mice were equivalently impaired in theirability to survive a P815 challenge, compared with WT mice (Fig.7A). In contrast, gld/gld mice were able to clear P815 cells asefficiently as WT mice (Fig. 7A).

To examine the role of orphan granzymes in allogeneic tumorclearance, we compared the clearance of P815 cells in sublethallyirradiated (400 cGy) 129/SvJ mice (WT vs GzmB�/�/�PGK-neovs GzmB�/�/�PGK-neo). GzmB�/�/�PGK-neo mice were sig-nificantly impaired in their ability to clear P815 cells (Fig. 7B). Incontrast, GzmB�/�/�PGK-neo mice cleared P815 cells as effi-ciently as WT 129/SvJ mice.

DiscussionIn this report we describe two strains of mice that are deficient forGzmB. In one, a retained PGK-neo cassette in the GzmB genecauses knockdown of the most closely linked genes, GzmC and F.Removal of the PGK-neo cassette from the GzmB gene by Cre-mediated recombination restored the expression of GzmC and F tonormal levels in LAK cells; however, in MLR-derived CTLs,GzmC and F were expressed at higher than normal levels. Thecytotoxic potential of MLR-generated CTLs from both GzmB-de-ficient strains was reduced, but the defect was more severe in micewith the retained PGK-neo cassette, implying that GzmC and/or Fmay be relevant for in vitro cytotoxicity. Finally, these results werecorroborated in vivo, where the allogeneic clearance of P815 tu-mor cells was significantly impaired in GzmB�/�/�PGK-neomice, but was normal in mice deficient for GzmB only. These datasuggest that the orphan GzmC and/or F are relevant for cytotoxiclymphocyte functions in vitro and in vivo.

Retained PGK-neo cassettes in a targeted locus can cause theexpression of tightly linked genes to be reduced (19, 20, 29–31).Although many examples of this neighborhood effect have beenreported, the mechanism and rules that govern the effect are not yetclear. The interposition of an active selectable marker cassette be-tween a locus control region (LCR) and its functional targets cansignificantly reduce the expression of those linked genes over tensof kilobases. This effect can often be eliminated by removing theselectable marker cassette, using LoxP/Cre (or Frt/Flp) recombi-nation technology (20, 29–31). Indeed, removal of the PGK-neo

FIGURE 4. Flow cytometric analysis of lymphocyte subpopulationsand GzmB expression. A, CD4�, CD8�, and NK (DX5) cell subpopulationanalysis in unstimulated splenocytes from WT, GzmB�/�/�PGK-neo, andGzmB�/�/�PGK-neo mice, showing similar cellular compositions. B,Lack of GzmB expression in unstimulated CD4�, CD8�, and NK (DX5)cell populations in WT spleens. C, Intracellular GzmB detection in WTCD4� and CD8� cells from 4-day MLRs. The lack of detectable GzmBexpression in GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neo MLR cellsis demonstrated, with the exception of nonspecific background macrophagestaining, which is demonstrated in the top panels.

2128 ORPHAN GRANZYME AND CTL

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

cassette from the GzmB gene restored expression of GzmC and F,which are located 24.7 and 44.8 kb downstream from GzmB, re-spectively. The expression of the granzyme genes located furtherdownstream (G, D, and E) was not affected by the retained PGK-neo cassette, which suggests that the effect is limited to the 5� endof this multigene cluster. We previously examined the effect of aretained PGK-neo cassette in the murine cathepsin G gene, whichlies just downstream from the GzmE gene (32); this retained cas-sette knocked down the expression of the mast cell chymase genejust downstream from it, but did not affect the expression of thegranzymes found upstream, suggesting a unidirectional componentof the neighborhood effect.

In our previously described mouse model of GzmB deficiency(18), the GzmB gene was targeted with a standard PGK-neo cas-sette, and expression studies were performed in mice on a mixedstrain background (129/SvOla�C57BL/6). In that mouse model,the expressions of GzmC, F, and D were all significantly reducedin LAK cells (19). In the current study both targeted mutationswere made in a 129/SvJ ES line (Rw4). Chimeric males made fromthese ES cells were bred to 129/SvJ females, so that the mutationswere immediately anchored in a pure 129/SvJ background. In thecurrent model we detected a striking reduction in GzmC and F ex-pression in LAK cells, but detected no reduction in GzmD expression.This may reflect strain-specific differences in the mouse models, or itcould explain differences in the measurement technology. In the pre-vious study experiments were performed using S1 nuclease protectionassays (19, 32); in the current study, quantitative RT-PCR studies

were performed, using primers absolutely specific for each granzymemRNA. Regardless, it is clear that the retained PGK-neo cassettecaused a knockdown of GzmC and F expression in both mutantmouse strains, and that this expression was restored upon removal ofthe cassette.

We do not yet understand why expression of the GzmC and Fgenes is higher than normal in MLR-derived CTLs from GzmB�/�/�PGK-neo mice. The targeted mutation of the GzmB gene createsa deletion between two naturally occurring AvrII sites. The 5� endof the deletion is in the very short 5�-untranslated region of GzmB.The deletion extends 350 bp downstream and includes all codingsequences within the first exon and part of the first intron of thegene. The transcription start site of the gene is therefore removed.It is possible that removal of the transcriptional start site disruptsa functional interaction between a putative LCR and the 5� end ofthe GzmB gene, allowing the LCR to scan downstream for the nextavailable promoter (i.e., GzmC). Alternatively, it is possible thathigh level transcription of the GzmB gene normally causes inter-ference with the transcription of GzmC and F; this transcriptionalinterference would be relieved by the start site deletion. However,the expression levels of GzmC and F are normal in LAK cells fromGzmB�/�/�PGK-neo mice, making it less likely that these mech-anisms are involved. Alternatively, it is possible that a specificsubpopulation of cells within an MLR preferentially expressesGzmC and F in this setting; we cannot address this issue experi-mentally as yet, because mAbs that permit flow-based detection ofthese granzymes are not yet available.

FIGURE 5. Functional assays of CTLs. A, Represen-tative FloKA data of allogeneic P815 and TA3 targetcells coincubated with MLR-derived CTL at an E:T cellratio of 20:1 for the indicated times. CFSE-positive tar-get cells were gated and analyzed for forward scatterand 7-AAD incorporation. The R1 gate contains small7-AADlow�high cells that are in the late stages of apo-ptotic cell death. The R2 gate contains large 7-AADhigh

cells that are in the early stages of cell death. Both Gzm-deficient strains have a defect in the killing of P815 andTA3 allogeneic tumor cells, specifically in the percent-ages of late apoptotic cells (R1 gate), with GzmB�/�/�PGK-neo CTLs demonstrating the most severe defect.Similar trends were obtained using E:T cell ratios of 5:1and 10:1. B, [125I]UdR release assay using LAK celleffectors coincubated with [125I]UdR-labeled YAC-1target cells for 2 h (left panel) or 8 h (right panel). LAKcells from both GzmB-deficient strains demonstrated asimilar severe defect in the release of [125I]UdR at 2 h atall E:T cell ratios tested. This defect had resolved at 8 h,as previously described (18).

2129The Journal of Immunology

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

The generation of these two GzmB-deficient mouse strains hasallowed us to assess the importance of granzyme genes down-stream from GzmB in cytotoxic killing assays in vitro and in vivo.Traditional [125I]UdR and 51Cr release assays did not reveal sig-nificant differences between the cytotoxic profiles of LAK cellswith and without the PGK-neo cassette. However, FloKA revealedthat the CTLs derived from GzmB�/�/�PGK-neo mice had amore significant killing defect than those from GzmB�/�/�PGK-neo mice. Previous work by Lecoeur et al. (33, 34) revealed thattarget cells with the FSClow/7-AADhigh phenotype represent lateapoptotic target cells (as defined by increased activation ofcaspases-3 and -8, as well as increased mitochondrial depolariza-tion, DNA fragmentation, and annexin V staining). In contrast,FSChigh/7-AADhigh target cells (R2 gate) were shown to representcells in the early stages of apoptosis (defined by lower levels ofmitochondrial depolarization, DNA fragmentation, and annexin Vstaining) (33, 34). In this study we noted a significant decrease inthe percentage of late apoptotic target cells (e.g., FSClow/7-AADlow�high; R1 gate) induced by MLR-generated CTLs andLAK cells from both GzmB-deficient strains, but the defect wasmore severe with CTLs from the GzmB�/�/�PGK-neo mice.CTLs from both strains caused excess target cells to appear in theearly apoptotic gate (R2), suggesting that the full induction of tar-get cell death requires additional time if GzmB is missing, con-

sistent with previous results (18). Alternatively, the differences inthe target cell FLoKA patterns between the two strains may rep-resent specific, biochemically unique apoptotic events induced byGzmC and/or F. Indeed, data from our laboratory have demon-strated that GzmC can directly induce mitochondrial depolariza-tion and nuclear collapse in a noncaspase-dependent manner (23).Regardless, these findings suggest that the FloKA assay can detectsubtle changes in target cells undergoing cell death that are notdetected by traditional cytotoxicity assays (that rely on the releaseof radiolabeled proteins and/or DNA from target cells). The strongcorrelation between the cytotoxic defects detected by this methodand the tumor clearance defect detected in vivo suggests that thesechanges may indeed be physiologically relevant.

We detected a strong defect in the clearance of P815 tumorcells in perforin-deficient mice and in mice with our originalGzmB�/�/�PGK-neo mutation on the C57BL/6 background.These data strongly suggest that GzmB and/or the orphan gran-zymes affected by this mutation (i.e., GzmC, F, and D) are relevantfor the perforin-dependent clearance of this tumor cell line in vivo.These data are similar to those reported by Pardo et al. (4), whoshowed that GzmA and B cluster enzymes are important for per-forin-dependent clearance of RMAS cells in vivo. However, ourdata are different from those reported by Smyth, Trapani, and col-leagues (5, 6), who detected perforin-dependent, but granzyme-independent, clearance of several tumor cell lines in vivo. Thereasons for these discrepancies are not yet clear. Differences inlaboratory stocks of tumor cell lines could be relevant, and subtledifferences in the strain backgrounds of the mutant mice could alsobe important, because GzmB�/�/�PGK-neo mice on theC57BL/6J background are clearly more susceptible to death thanthe same mice on the 129/SvJ background (see Fig. 7). Clearly,additional studies will be required to understand the causes ofthese laboratory-to-laboratory differences.

FIGURE 6. Summary of CTL and LAK cytotoxicity studies. A, FloKAanalysis of CTL cytotoxicity against allogeneic targets (P815 and TA3tumor cells) from WT, GzmB�/�/�PGK-neo, and GzmB�/�/�PGK-neomice at a 20:1 E:T cell ratio after 3 h of incubation. GzmB�/�/�PGK-neoand GzmB�/�/�PGK-neo CTLs demonstrate significantly decreased abil-ity to induce late stage apoptotic cells (R1 gated cells) vs WT in P815 andTA3 target cells. Both GzmB�/�/�PGK-neo and GzmB�/�/�PGK-neoCTLs had a significantly higher fraction of large 7-AADhigh cells (P815and TA3) in the early stages of cell death (R2 gated cells) compared withWT CTLs. B, FloKA analysis of LAK cytotoxicity against NK-sensitivetargets (RMAS and YAC-1 tumor cells) from WT, GzmB�/�/�PGK-neo,and GzmB�/�/�PGK-neo mice at a 10:1 E:T cell ratio after 4 h of coin-cubation. All effectors demonstrated significant target cell killing comparedwith targets only (p 0.0001). Data shown are the combination of fourindependent experiments, each performed in duplicate. Values shown arethe mean � SD. �, p 0.05; ��, p 0.01; ���, p 0.001.

FIGURE 7. In vivo allogeneic tumor cell clearance. A, Sublethally ir-radiated mice (H-2Kb, C57BL/6J background) were injected with 1 � 107

allogeneic P815 tumor cells and monitored for survival. GzmB�/�/�PGK-neo and perforin�/� mice had significantly diminished survival comparedwith WT and gld/gld mice. B, Sublethally irradiated mice (H-2Kb,129/SvJbackground) were injected with 1 � 107 allogeneic P815 tumor cells andmonitored for survival. GzmB�/�/�PGK-neo deficient mice had a signif-icantly diminished survival compared with WT and GzmB�/�/�PGK-neomice.

2130 ORPHAN GRANZYME AND CTL

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

In addition to the defect in allogeneic tumor cell clearance inGzmB�/�/�PGK-neo mice, we have detected a defect in the con-trol of gammaherpesvirus (�HV68) latency and reactivation inthese mice (35); this defect is more severe than that in mice defi-cient for GzmB only. These data support the observations reportedin this study and implicate GzmC and/or F in the control of gam-maherpesviruses in vivo.

The neighborhood effect in LAK cells derived from GzmB�/�/�PGK-neo mice predominantly affects GzmC and F. Recent stud-ies in our laboratory have demonstrated that GzmC can inducetarget cell death that has many of the features of apoptosis, includ-ing nuclear collapse, annexin V positivity, and mitochondrial de-polarization. However, GzmC does not cause nuclear fragmenta-tion or double-stranded intranucleosomal DNA cleavage, and itdoes not cleave any of the recognized substrates of GzmB (23).The ability of GzmC to cause cell death is related to its proteaseactivity, because a mutation in its active site serine significantlyreduces its ability to kill target cells in vitro; however, the speci-ficity of this protease and its relevant substrates are currently un-known. Because GzmC can induce target cell death, its reducedexpression in GzmB�/�/�PGK-neo-derived CTLs may well ex-plain the reduced cytotoxicity of these cells, at least in part. Theability of GzmF to induce cell death has not yet been examined;likewise, its substrates are not yet known. Regardless, it is impor-tant to note that the overexpression of GzmC and F in MLR-de-rived CTLs does not rescue the cytotoxicity defect caused by theloss of GzmB, suggesting that the enzymes are not functionallyredundant. Additional studies and knockout mice will be requiredto dissect the relative contributions of GzmC and F to the pheno-types observed in this report.

In summary, these studies have shown that a retained PGK-neocassette in the GzmB gene causes a knockdown of the expressionlevels of GzmC and F just downstream from the GzmB gene.CTLs derived from these mice have a more severe cytotoxic defectthan those from mice deficient for GzmB only. These data clearlyimplicate GzmC and F in CTL functions and suggest the need foradditional studies of their substrate specificities and killing activities.Furthermore, because GzmC is the probable orthologue of humanGzmH (17), it will be important to learn whether human GzmH isimportant for the ability of human cytotoxic lymphocytes to kill theirtargets.

AcknowledgmentsWe thank the Siteman Cancer Center High Speed Cell Sorter Core and theEmbryonic Stem Cell Core for technical support, and Kelly Schrimpf andMieke Hoock for excellent mouse husbandry. Nancy Reidelberger pro-vided expert editorial assistance.

References1. Russell, J. H., and T. J. Ley. 2002. Lymphocyte-mediated cytotoxicity. Annu.

Rev. Immunol. 20:323.2. Lieberman, J. 2003. The ABCs of granule-mediated cytotoxicity: new weapons in

the arsenal. Nat. Rev. Immunol. 3:361.3. Barry, M., and R. C. Bleackley. 2002. Cytotoxic T lymphocytes: all roads lead to

death. Nat. Rev. Immunol. 2:401.4. Pardo, J., S. Balkow, A. Anel, and M. Simon. 2003. Granzymes are essential for

natural killer cell-mediated and perf-facilitated tumor control. Eur. J. Immunol.32:2881.

5. Davis, J. E., M. J. Smyth, and J. A. Trapani. 2001. Granzyme A and B-deficientkiller lymphocytes are defective in eliciting DNA fragmentation but retain potentin vivo anti-tumor capacity. Eur. J. Immunol. 31:39.

6. Smyth, M. J., S. E. Street, and J. A. Trapani. 2003. Cutting edge: granzymes A andB are not essential for perforin-mediated tumor rejection. J. Immunol. 171:515.

7. Yasukawa, M., H. Ohminami, J. Arai, Y. Kasahara, Y. Ishida, and S. Fujita. 2000.Granule exocytosis, and not the fas/fas ligand system, is the main pathway ofcytotoxicity mediated by alloantigen-specific CD4� as well as CD8� cytotoxic Tlymphocytes in humans. Blood 95:2352.

8. Smith, M. J., K. Y T. Thia, S. E. A. Street, D. MacGregor, D. I. Godfrey, andJ. A. Trapani. 2000. Perforin-mediated cytotoxicity is critical for surveillance ofspontaneous lymphoma. J. Exp. Med. 192:755.

9. Tsukada, N., T. Kobata, Y. Aizawa, H. Yagita, and K. Okumura. 1999. Graft-versus-leukemia effect and graft-versus-host disease can be differentiated by cy-totoxic mechanisms in a murine model of allogeneic bone marrow transplanta-tion. Blood 93:2738.

10. Pan, L., T. Teshima, G. R. Hill, D. Bungard, Y. S. Brinson, V. S. Reddy,K. R. Cooke, and J. L. Ferrara. 1999. Granulocyte colony-stimulating factor-mobilized allogeneic stem cell transplantation maintains graft-versus-leukemiaeffects through a perforin-dependent pathway while preventing graft-versus-hostdisease. Blood 93:4071.

11. Schmaltz, C., O. Alpdogan, K. J. Horndasch, S. J. Muriglan, B. J. Kappel,T. Teshima, J. L. Ferrara, S. J. Burakoff, and M. R. van den Brink. 2001. Dif-ferential use of Fas ligand and perforin cytotoxic pathways by donor T cells ingraft-versus-host disease and graft-versus-leukemia effect. Blood 97:2886.

12. Kam, C. M., D. Hudig, and J. C. Powers. 2000. Granzymes (lymphocyte serineproteases): characterization with natural and synthetic substrates and inhibitors.Biochim. Biophys. Acta. 1477:307.

13. Bredemeyer, A. J., R. M. Lewis, J. P. Malone, A. E. Davis, A. E. Gross,R. R. Townsend, and T. J. Ley. 2004. A proteomic approach for the discovery ofprotease substrates. Proc. Natl. Acad. Sci. USA 101:11785.

14. Ebnet, K., M. Hausmann, F. Lehmann-Grube, A. Mullbacher, M. Kopf,M. Lamers, and M. M. Simon. 1995. Granzyme A-deficient mice retain potentcell-mediated cytotoxicity. EMBO J. 14:4230.

15. Beresford, P. J., Z. Xia, A. H. Greenberg, and J. Lieberman. 1999. Granzyme Aloading induces rapid cytolysis and a novel form of DNA damage independentlyof caspase activation. Immunity 10:585.

16. Shresta, S., T. A. Graubert, D. A. Thomas, S. Z. Raptis, and T. J. Ley. 1999.Granzyme A initiates an alternative pathway for granule-mediated apoptosis. Im-munity 10:595.

17. Grossman, W. J., P. A. Revell, Z. H. Lu, H. J. Johnson, A. J. Bredemeyer, and T. J. Ley.2003. The orphan granzymes of humans and mice. Curr. Opin. Immunol. 15:544.

18. Heusel, J. W., R. L. Wesselschmidt, S. Shresta, J. H. Russell, and T. J. Ley. 1994.Cytotoxic lymphocytes require granzyme B for the rapid induction of DNA frag-mentation and apoptosis in allogeneic target cells. Cell 76:977.

19. Pham, C. T., D. M. MacIvor, B. A. Hug, J. W. Heusel, and T. J. Ley. 1996.Long-range disruption of gene expression by a selectable marker cassette. Proc.Natl. Acad. Sci. USA 93:13090.

20. Hug, B., R. Wesselschmidt, S. Fiering, M. Bender, E. Epner, M. Groudine, andT. J. Ley. 1996. Analysis of mice containing a targeted deletion of �-globin locuscontrol region 5� hypersensitive site 3. Mol. Cell. Biol. 16:2906.

21. Westervelt, P., A. A. Lane, J. L. Pollock, K. Oldfather, M. S. Holt,D. B. Zimonjic, N. C. Popescu, J. F. DiPersio, and T. J. Ley. 2003. High pen-etrance mouse model of acute promyelocytic leukemia with very low levels ofPML/RAR� expression. Blood 102:1857.

22. Shresta, S., D. M. MacIvor, J. W. Heusel, J. H. Russell, and T. J. Ley. 1995.Natural killer and lymphokine-activated killer cells require granzyme B for therapid induction of apoptosis in susceptible target cells. Proc. Natl. Acad. Sci. USA92:5679.

23. Johnson, H., L. Scorrano, S. J. Korsmeyer, and T. J. Ley. 2003. Cell death in-duced by granzyme C. Blood 101:3093.

24. Pham, C. T., D. A. Thomas, J. D. Mercer, and T. J. Ley. 1998. Production of fullyactive recombinant murine granzyme B in yeast. J. Biol. Chem. 273:1629.

25. Graubert, T. A., J. F. DiPersio, J. H. Russell, and T. J. Ley. 1997. Perforin/granzyme-dependent and independent mechanisms are both important for thedevelopment of graft-versus-host disease after murine bone marrow transplanta-tion. J. Clin. Invest. 100:904.

26. Grossman, W. J., J. W. Verbsky, B. L. Tollefson, C. Kemper, J. P. Atkinson, andT. J. Ley. 2004. Differential expression of granzymes A and B in human cytotoxiclymphocyte subsets and T regulatory cells. Blood 10:1182.

27. Russell, J. H., V. Masakowski, T. Rucinsky, and G. Phillips. 1982. Mechanismsof immune lysis. III. Characterization of the nature and kinetics of the cytotoxicT lymphocyte-induced nuclear lesion in the target. J. Immunol. 128:2087.

28. Kelso, A., E. O. Costelloe, B. J. Johnson, P. Groves, K. Buttigieg, andD. R. Fitzpatrick. 2002. The genes for perforin, granzymes A-C and IFN-� aredifferentially expressed in single CD8� T cells during primary activation. Int.Immunol. 14:605.

29. Kim, C. G., E. M. Epner, W. C. Forrester, and M. Groudine. 1992. Inactivationof the human �-globin gene by targeted insertion into the �a-globin locus controlregion. Genes Dev. 6:928.

30. Fiering, S., E. Epner, K. Robinson, Y. Zhuang, A. Telling, M. Hu, D. I. Martin,T. Enver, T. J. Ley, and M. Groudine. 1995. Targeted deletion of 5�HS2 of themurine �-globin LCR reveals that it is not essential for proper regulation of the�-globin locus. Genes Dev 9:2203.

31. Fiering, S., C. G. Kim, E. M. Epner, and M. Groudine. 1993. An “in-out” strategyusing gene targeting and FLP recombinase for the functional dissection of com-plex DNA regulatory elements: analysis of the �-globin locus control region.Proc. Natl. Acad. Sci. USA 90:8469.

32. MacIvor, D. M., S. D. Shapiro, C. T. Pham, A. Belaaouaj, S. N. Abraham, and T. J. Ley.1999. Normal neutrophil function in cathepsin G-deficient mice. Blood 94:4282.

33. Lecoeur, H., Fevrier, M., Garcia, S., Riviere, Y., and Gougeon, M. L. 2001. Anovel flow cytometric assay for quantitation and multiparametric characterizationof cell-mediated cytotoxicity. J. Immunol. Methods 253:177.

34. Lecoeur, H., L. M. de Oliveria-Pinto, and M. L. Gougeon. 2002. Multiparametric flowcytometric analysis of biochemical and functional events associated with apoptosis andoncosis using the 7-aminoactinomycin D assay. J. Immunol. Methods 265:81.

35. Loh, J., D. A. Thomas, P. A. Revell, T. J. Ley, and H. W. Virgin. 2004. Gran-zymes and Caspase 3 play important roles in the control of gammaherpesviruslatency. J. Virol. 78:12519.

2131The Journal of Immunology

by guest on January 22, 2019http://w

ww

.jimm

unol.org/D

ownloaded from