SHORT COMMUNICATION Human Perforin (PRFI) Maps to lOq22, a Region That Is Syntenic with Mouse Chromosome 10 THOMAS M. FINK,* MICHAEL ZIMMER,t SANDRA WEITZ,* JURG TSCHOPP,$ DIETER E. JENNE,$ AND PETER LIGHTER* *Deutsches Krebsforschungszentrum, Schwerpunkt Angewandte Tumorvirologie, fijr Experimentelle Medizin, Abt. Molekulare Germany; t Max-Planck-lnstitut Germany; and +/nstitute of Biochemistry, University of Lausanne, Received
January
10, 1992;
Perforin (PRFl) is a cytolytic, channel-forming protein of cytolytic T cells, natural killer cells, and granulated metrial gland cells and plays a crucial role in the killer cell-mediated elimination of virally infected host cells, tumor cells, and allotransplants. Two-thirds of the perforin sequence is homologous to the lytic, channel-forming complement proteins C6, C7, CSa, C&3, and C9. Using cosmid DNA containing the PRFl gene as a probe for fluorescence in situ hybridization, we have reevaluated its chromosomal location. Previously assigned to chromosome 17ql l-q21, it has now been mapped to lOq22. The human PRFl locus lies within a conserved synteny segment present on mouse chromosome 10, consistent with the previous chromosomal assignment of mouse perforin. The perforin locus is not linked to any of the genes of the terminal complement system. 8 1992 Academic Press. Inc.
Perforin (PRFl; synonyms: PFPl, pore-forming protein 1, cytolysin, CS-related protein), like other granule constituents-e.g., proteoglycans and granzymes-of cytolytic effector cells, is released at the contact area between target and effector cells and induces target cell death (11, 15, 16). Human perforin and mouse perforin are transcribed from single-copy genes (5, 14). The human perforin gene has previously been mapped to chromosome 17qll-q21 by a combination of somatic cell hybrid analysis and radioactive in situ hybridization (la), whereas the mouse perforin has been assigned to mouse chromosome 10 (14). Two circumstances prompted us to reexamine the published human mapping data. First, perforin clones were undetectable in a human chromosome 17 reference library (10). In the same library nine cosmid clones for human vitronectin, which maps to the centromeric region of 17qll (4), were readily identified. Second, the lack of any known region of homology between human chromosome 17 and mouse chromosome 10 (2) further nourished our suspicion about an incorrect assignment of the human perforin gene. To obtain a perforin probe suitable for mapping by nonradioactive in situ hybridization, we screened a huGENOMICS 13, 1300-1302 (19%) 0888-7543/9‘2 $5.00 Copyright El 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
1300
revised
April
Im Neuenheimer Fe/d 280, D-6900 Heidelberg, Neuroendokrinologie, D-3400 Gbttingen, CH-1066 Epalinges, Switzerland
13, 1992
man cosmid library constructed from total human DNA of peripheral blood leukocytes in the vector pcos2 EMBL using a labeled human perforin cDNA fragment (the NcoI-Hind111 fragment at the 3’ end) kindly provided by E. Podack (5). Two cosmid clones, PERF-1 and PERF-5, that overlap to about 75% were identified. The identity of these clones was confirmed by Southern blot analysis using as probes the perforin cDNA and oligonucleotides JT29 (5’-ATGGCAGCCCGTCTGCTC-3’) and JT30 (5’-ATTAACGACCTGCTGTTC-3’), which are specific for exons 2 and 3, respectively. As predicted from the complete genomic perforin sequence (5), JT29 hybridized to the 5-kb BamHI fragment containing exons 1 and 2, and JT30 to the 1-kb BumHI fragment containing the 5’portion of exon 3. The 2.3-kb PstI fragment covering exon 2 and the 5’ portion of exon 3 was detected with both oligonucleotide probes. For in situ hybridization experiments total cosmid DNA of the clone PERF-5 was biotin-labeled and used for chromosomal in situ suppression hybridization to elongated metaphase chromosomes as described (6, 7). Specific fluorescent signals of the perforin probe were found on both sister chromatids of both chromosome 10 homologues in 90% of the evaluated metaphases (see Figs. 1A and 1B; 71 metaphases were analyzed from five independent experiments). Additional fluorescence signals on other chromosomes were not observed. Chromosomal assignment was confirmed by cohybridization with the centromere-specific alphoid probe palORP8 (DlOZl (3)) (see Fig. 1B) and by copainting of chromosome 10 with a DNA library enriched in chromosome 10 sequences (1) kindly provided by J. Gray (see Fig. 1C). The localization to band lOq22 was established by chromosome banding with 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) (see Fig. 1A). Thus, the previously reported map location 17qll-q21 for human perforin (12) cannot be reconciled with our data. This discrepancy might be explained by unknown chromosomal material present in the hybrid lines of the cell panel used for the previous chromosome assignment. Given the subchromosomal location of the human perforin gene as lOq22 and the presence of mouse per-
SHORT
1301
COMMUNICATION
FIG. 1. Mapping of the human perforin gene to lOq22. (A) Sixty nanograms of biotin-labeled cosmid DNA containing the perforin gene (PERF-6) was combined with 30 pg of human Cot1 DNA (BRL) in a 30-~1 hybridization cocktail and hybridized to human metaphase chromosomes. The bound probe was visualized via FITC conjugated to avidin (see arrows). Chromosomal counterstaining with DAPI resulted in a Q-banding-like pattern. Note that the highly specific FITC signals (yellow) are located within lOq22. Digitized images of FITC and DAPI fluorescence were generated separately with the use of a CCD camera (Photometrics), processed, and overlaid electronically. (A’) DAPI image of A to visualize lOq22 (see arrowheads). (B) Colocalization of the PERF-5 probe (arrow) and the centromere-specific probe palORP8 (red signals, see arrowheads). Five nanograms of digoxigenin-labeled palORP8 DNA in 5 ~1 of hybridization cocktail was denatured separately and was applied to the slides together with the hybridization cocktail containing biotin-labeled PERF-5 probe. (C) Simultaneous detection of PERF-5 (arrow) and “red-painted” chromosome 10. Six hundred nanograms of digoxigenin-labeled DNA from the chromosome lo-derived library pBSl0 was added to the hybridization cocktail containing biotin-labeled PERF-5 probe. In B and C, the digoxigenin-modified chromosome lo-specific DNA probes were detected via rhodamine conjugated to anti-digoxigenin antibodies. Digitized images of emitted FITC and rhodamine fluorescence as well as differential interference contrast (DIC) images to visualize the chromosomal outline were generated using a confocal scanning microscope (LSMlO, Zeiss Oberkochen) as described (4). Separately recorded images of FITC, rhodamine, and DIC were processed and overlayed electronically. To optimize the alignment of images, the overlay in all three panels was adjusted according to the background fluorescence of the chromosomal bodies, which was electronically enhanced (not shown). Photographs were taken directly from the video screen.
forin on chromosome 10, we predict that the mouse perforin locus falls into the previously identified region of synteny between human and mouse chromosome 10 (2, 9,14). This syntenic region is defined by the cell division cycle control protein 2 gene, the hexokinase-1 gene at lOq22, the inorganic pyrophosphatase gene (PP in the human, Pyp in the mouse), the early growth response gene 2 (2), and the gene for core peptide of proteoglycans (8) that are found in protein granules of neutrophils, mast cells, natural killer cells, and cytolytic T cells (13). ACKNOWLEDGMENTS We thank Sibylle Schmid and Chantal Mattmann for technical assistance, J. Spiess, Giittingen, for continuing interest and support, and M. Schwab for sharing equipment for digital imaging microscopy.
This work was supported by the Verein zur Fiirderung chung in Deutschland (to P.L.) and by a grant from tional Science Foundation (to D.E.J.).
der Krebsforsthe Swiss Na-
REFERENCES Collins, C. C., Kuo, W. L., Segraves, R., Pinkel, D., Fuscoe, J. C., and Gray, J. W. (1991). Construction and characterization of plasmid libraries enriched in sequences from single human chromosomes. Genomics 11: 997-1006. Davisson, M. T., Lalley, P. A., Peters, J., Doolittle, D. P., Hillyard, A. L., and Searle, A. G. (1991). Report of the comparative subcommittee for human and mouse homologies. Cytogenet. Cell Genet. 58: 1152-1189. Devilee, P., Kievits, T., Waye, J. S., Pearson, P. L., and Huntington, F. W. (1988). Chromosome-specific alpha satellite DNA: Isolation and mapping of a polymorphic alphoid repeat from human chromosome 10. Genomics 3: l-7.
1302
SHORT
COMMUNICATION
4.
Fink, T. M., Jenne, D. E., and Lichter, P. (1992). The vitronectin (complement S-protein) gene maps to the merit region of 17q. Hum. Genet. 88: 569-572.
5.
Lichtenheld, M. G., and Podack, E. R. (1989). Structure of the human perforin gene: A simple gene organization with interesting potential regulatory sequences. J. Immunol. 143: 42674274.
6.
Lichter, P., Cremer, T., Borden, J., Manuelidis, L., and Ward, D. C. (1966). Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum. Genet. 80: 224234.
7.
human centro-
Lichter, P., Tang, C. C., Call, K., Hermanson, G., Evans, G. A., Housman, D., and Ward, D. C. (1990). High resolution mapping of human chromosome 11 by in situ hybridization with cosmid clones. Science 247: 64-69.
8.
Mattei, M. G., Perin, J.-P., Alliel, P. M., Bonnet, F., Maillet, P., Passage, E., Mattei, J.-F., and Jolles, P. (1989). Localization of human platelet proteoglycan gene to chromosome 10, band q22.1, by in situ hybridization. Hum. Genet. 82: 87-86.
9.
Nadeau, J. H. (1989). Maps of linkage synteny tween mouse and man. Trends Genet. 5: 82-86.
10.
Nizetic, Lehrach,
D., Zehetner, H. (1991).
G., Monaco, Construction,
homologies
screening 21: Their Sci. USA
12.
Shinkai, Y., Yoshida, M. C., Maeda, K., Kobata, T., Muruyama, K., Yodoi, J., Yagita, H., and Okumura, K. (1989). Molecular cloning and chromosomal assignment of a human perforin (PFP) gene. Immunogenetics 30: 452-457.
13.
Stevens, R. L., Kamada, M. M., and Serafin, W. E. (1989). Structure and function of the family of proteoglycans that reside in the secretory granules of natural killer cells and other effector cells of the immune response. Curr. Top. Microbial. Immunol. 140: 93-108. Trapani, J. A., Kwon, B. S., Kozag, C. A., Chintamaneni, C., Young, J. D., and DuPont, B. (1990). Genomic organization of the mouse pore-forming protein (perforin) gene and localization to chromosome 10: Similarities to and differences from C9. J. Exp. Med. 1’71: 545-557.
15.
Tschopp, cell lysis 279-302.
16.
Young, (1990).
H., and Lichtenheld, in cytolysis? Annu.
X and Acad.
Podack, central 129-157.
14.
E. R., Hengartner, role of perforin
libraries of human chromosomes as reference libraries. hoc. Natl.
11.
be-
A. P. L. G., Young, B. D., and arraying, and high-density
of large insert potential use 88: 3233-3237.
J., and Nabholz, M. (1990). by cytolytic T lymphocytes.
M. G. (1991). Rev. Immurwl.
A 9:
Perforin-mediated target Annu. Rev. Immunol. 8:
L. H., Liu, C. C., Joag, S., Rafii, How lymphocytes kill. Annu. Rev.
S., and Young, Med. 41: 45-54.
J. D.
January
10, 1992;
Perforin (PRFl) is a cytolytic, channel-forming protein of cytolytic T cells, natural killer cells, and granulated metrial gland cells and plays a crucial role in the killer cell-mediated elimination of virally infected host cells, tumor cells, and allotransplants. Two-thirds of the perforin sequence is homologous to the lytic, channel-forming complement proteins C6, C7, CSa, C&3, and C9. Using cosmid DNA containing the PRFl gene as a probe for fluorescence in situ hybridization, we have reevaluated its chromosomal location. Previously assigned to chromosome 17ql l-q21, it has now been mapped to lOq22. The human PRFl locus lies within a conserved synteny segment present on mouse chromosome 10, consistent with the previous chromosomal assignment of mouse perforin. The perforin locus is not linked to any of the genes of the terminal complement system. 8 1992 Academic Press. Inc.
Perforin (PRFl; synonyms: PFPl, pore-forming protein 1, cytolysin, CS-related protein), like other granule constituents-e.g., proteoglycans and granzymes-of cytolytic effector cells, is released at the contact area between target and effector cells and induces target cell death (11, 15, 16). Human perforin and mouse perforin are transcribed from single-copy genes (5, 14). The human perforin gene has previously been mapped to chromosome 17qll-q21 by a combination of somatic cell hybrid analysis and radioactive in situ hybridization (la), whereas the mouse perforin has been assigned to mouse chromosome 10 (14). Two circumstances prompted us to reexamine the published human mapping data. First, perforin clones were undetectable in a human chromosome 17 reference library (10). In the same library nine cosmid clones for human vitronectin, which maps to the centromeric region of 17qll (4), were readily identified. Second, the lack of any known region of homology between human chromosome 17 and mouse chromosome 10 (2) further nourished our suspicion about an incorrect assignment of the human perforin gene. To obtain a perforin probe suitable for mapping by nonradioactive in situ hybridization, we screened a huGENOMICS 13, 1300-1302 (19%) 0888-7543/9‘2 $5.00 Copyright El 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
1300
revised
April
Im Neuenheimer Fe/d 280, D-6900 Heidelberg, Neuroendokrinologie, D-3400 Gbttingen, CH-1066 Epalinges, Switzerland
13, 1992
man cosmid library constructed from total human DNA of peripheral blood leukocytes in the vector pcos2 EMBL using a labeled human perforin cDNA fragment (the NcoI-Hind111 fragment at the 3’ end) kindly provided by E. Podack (5). Two cosmid clones, PERF-1 and PERF-5, that overlap to about 75% were identified. The identity of these clones was confirmed by Southern blot analysis using as probes the perforin cDNA and oligonucleotides JT29 (5’-ATGGCAGCCCGTCTGCTC-3’) and JT30 (5’-ATTAACGACCTGCTGTTC-3’), which are specific for exons 2 and 3, respectively. As predicted from the complete genomic perforin sequence (5), JT29 hybridized to the 5-kb BamHI fragment containing exons 1 and 2, and JT30 to the 1-kb BumHI fragment containing the 5’portion of exon 3. The 2.3-kb PstI fragment covering exon 2 and the 5’ portion of exon 3 was detected with both oligonucleotide probes. For in situ hybridization experiments total cosmid DNA of the clone PERF-5 was biotin-labeled and used for chromosomal in situ suppression hybridization to elongated metaphase chromosomes as described (6, 7). Specific fluorescent signals of the perforin probe were found on both sister chromatids of both chromosome 10 homologues in 90% of the evaluated metaphases (see Figs. 1A and 1B; 71 metaphases were analyzed from five independent experiments). Additional fluorescence signals on other chromosomes were not observed. Chromosomal assignment was confirmed by cohybridization with the centromere-specific alphoid probe palORP8 (DlOZl (3)) (see Fig. 1B) and by copainting of chromosome 10 with a DNA library enriched in chromosome 10 sequences (1) kindly provided by J. Gray (see Fig. 1C). The localization to band lOq22 was established by chromosome banding with 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) (see Fig. 1A). Thus, the previously reported map location 17qll-q21 for human perforin (12) cannot be reconciled with our data. This discrepancy might be explained by unknown chromosomal material present in the hybrid lines of the cell panel used for the previous chromosome assignment. Given the subchromosomal location of the human perforin gene as lOq22 and the presence of mouse per-
SHORT
1301
COMMUNICATION
FIG. 1. Mapping of the human perforin gene to lOq22. (A) Sixty nanograms of biotin-labeled cosmid DNA containing the perforin gene (PERF-6) was combined with 30 pg of human Cot1 DNA (BRL) in a 30-~1 hybridization cocktail and hybridized to human metaphase chromosomes. The bound probe was visualized via FITC conjugated to avidin (see arrows). Chromosomal counterstaining with DAPI resulted in a Q-banding-like pattern. Note that the highly specific FITC signals (yellow) are located within lOq22. Digitized images of FITC and DAPI fluorescence were generated separately with the use of a CCD camera (Photometrics), processed, and overlaid electronically. (A’) DAPI image of A to visualize lOq22 (see arrowheads). (B) Colocalization of the PERF-5 probe (arrow) and the centromere-specific probe palORP8 (red signals, see arrowheads). Five nanograms of digoxigenin-labeled palORP8 DNA in 5 ~1 of hybridization cocktail was denatured separately and was applied to the slides together with the hybridization cocktail containing biotin-labeled PERF-5 probe. (C) Simultaneous detection of PERF-5 (arrow) and “red-painted” chromosome 10. Six hundred nanograms of digoxigenin-labeled DNA from the chromosome lo-derived library pBSl0 was added to the hybridization cocktail containing biotin-labeled PERF-5 probe. In B and C, the digoxigenin-modified chromosome lo-specific DNA probes were detected via rhodamine conjugated to anti-digoxigenin antibodies. Digitized images of emitted FITC and rhodamine fluorescence as well as differential interference contrast (DIC) images to visualize the chromosomal outline were generated using a confocal scanning microscope (LSMlO, Zeiss Oberkochen) as described (4). Separately recorded images of FITC, rhodamine, and DIC were processed and overlayed electronically. To optimize the alignment of images, the overlay in all three panels was adjusted according to the background fluorescence of the chromosomal bodies, which was electronically enhanced (not shown). Photographs were taken directly from the video screen.
forin on chromosome 10, we predict that the mouse perforin locus falls into the previously identified region of synteny between human and mouse chromosome 10 (2, 9,14). This syntenic region is defined by the cell division cycle control protein 2 gene, the hexokinase-1 gene at lOq22, the inorganic pyrophosphatase gene (PP in the human, Pyp in the mouse), the early growth response gene 2 (2), and the gene for core peptide of proteoglycans (8) that are found in protein granules of neutrophils, mast cells, natural killer cells, and cytolytic T cells (13). ACKNOWLEDGMENTS We thank Sibylle Schmid and Chantal Mattmann for technical assistance, J. Spiess, Giittingen, for continuing interest and support, and M. Schwab for sharing equipment for digital imaging microscopy.
This work was supported by the Verein zur Fiirderung chung in Deutschland (to P.L.) and by a grant from tional Science Foundation (to D.E.J.).
der Krebsforsthe Swiss Na-
REFERENCES Collins, C. C., Kuo, W. L., Segraves, R., Pinkel, D., Fuscoe, J. C., and Gray, J. W. (1991). Construction and characterization of plasmid libraries enriched in sequences from single human chromosomes. Genomics 11: 997-1006. Davisson, M. T., Lalley, P. A., Peters, J., Doolittle, D. P., Hillyard, A. L., and Searle, A. G. (1991). Report of the comparative subcommittee for human and mouse homologies. Cytogenet. Cell Genet. 58: 1152-1189. Devilee, P., Kievits, T., Waye, J. S., Pearson, P. L., and Huntington, F. W. (1988). Chromosome-specific alpha satellite DNA: Isolation and mapping of a polymorphic alphoid repeat from human chromosome 10. Genomics 3: l-7.
1302
SHORT
COMMUNICATION
4.
Fink, T. M., Jenne, D. E., and Lichter, P. (1992). The vitronectin (complement S-protein) gene maps to the merit region of 17q. Hum. Genet. 88: 569-572.
5.
Lichtenheld, M. G., and Podack, E. R. (1989). Structure of the human perforin gene: A simple gene organization with interesting potential regulatory sequences. J. Immunol. 143: 42674274.
6.
Lichter, P., Cremer, T., Borden, J., Manuelidis, L., and Ward, D. C. (1966). Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum. Genet. 80: 224234.
7.
human centro-
Lichter, P., Tang, C. C., Call, K., Hermanson, G., Evans, G. A., Housman, D., and Ward, D. C. (1990). High resolution mapping of human chromosome 11 by in situ hybridization with cosmid clones. Science 247: 64-69.
8.
Mattei, M. G., Perin, J.-P., Alliel, P. M., Bonnet, F., Maillet, P., Passage, E., Mattei, J.-F., and Jolles, P. (1989). Localization of human platelet proteoglycan gene to chromosome 10, band q22.1, by in situ hybridization. Hum. Genet. 82: 87-86.
9.
Nadeau, J. H. (1989). Maps of linkage synteny tween mouse and man. Trends Genet. 5: 82-86.
10.
Nizetic, Lehrach,
D., Zehetner, H. (1991).
G., Monaco, Construction,
homologies
screening 21: Their Sci. USA
12.
Shinkai, Y., Yoshida, M. C., Maeda, K., Kobata, T., Muruyama, K., Yodoi, J., Yagita, H., and Okumura, K. (1989). Molecular cloning and chromosomal assignment of a human perforin (PFP) gene. Immunogenetics 30: 452-457.
13.
Stevens, R. L., Kamada, M. M., and Serafin, W. E. (1989). Structure and function of the family of proteoglycans that reside in the secretory granules of natural killer cells and other effector cells of the immune response. Curr. Top. Microbial. Immunol. 140: 93-108. Trapani, J. A., Kwon, B. S., Kozag, C. A., Chintamaneni, C., Young, J. D., and DuPont, B. (1990). Genomic organization of the mouse pore-forming protein (perforin) gene and localization to chromosome 10: Similarities to and differences from C9. J. Exp. Med. 1’71: 545-557.
15.
Tschopp, cell lysis 279-302.
16.
Young, (1990).
H., and Lichtenheld, in cytolysis? Annu.
X and Acad.
Podack, central 129-157.
14.
E. R., Hengartner, role of perforin
libraries of human chromosomes as reference libraries. hoc. Natl.
11.
be-
A. P. L. G., Young, B. D., and arraying, and high-density
of large insert potential use 88: 3233-3237.
J., and Nabholz, M. (1990). by cytolytic T lymphocytes.
M. G. (1991). Rev. Immurwl.
A 9:
Perforin-mediated target Annu. Rev. Immunol. 8:
L. H., Liu, C. C., Joag, S., Rafii, How lymphocytes kill. Annu. Rev.
S., and Young, Med. 41: 45-54.
J. D.
