- Email: [email protected]
Generation of a Rat Bronchiolar Epithelial Cell cDNA Library: Isolation of a Proline Rich Protein Highly Enriched in Bronchiolar Epithelial Cells
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
225, 877–882 (1996)
1266
Generation of a Rat Bronchiolar Epithelial Cell cDNA Library: Isolation of a Proline Rich Protein Highly Enriched in Bronchiolar Epithelial Cells Colin D. Bingle1 Department of Toxicology, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, London, EC1A 7ED, United Kingdom Received July 17, 1996 As a tool for the identification of novel gene products expressed in rat bronchiolar epithelial (BE) cells we generated a cDNA library prepared from BE cells, enriched to contain ú40% Clara cells. Using a simple differential screening strategy we isolated and partially characterized 47 clones, 29 of which corresponded to gene products known to be expressed within the bronchiolar epithelium. A further 15 clones contained sequences which were not present in the EMBL database and therefore represented potential novel BE cell enriched clones. One of these clones, CC-4, was characterized further and shown by northern blotting to be highly enriched in the BE cell population. The deduced protein product of CC-4 is a proline rich protein of 29.5 kd. These studies suggest that this cDNA library may be a useful tool for the identification of novel lung restricted gene products. q 1996 Academic Press, Inc.
Clara (Non-ciliated bronchiolar epithelial) cells are non-mucous secretory cells located in the surface epithelium of the pulmonary airways. These cells contain large amounts of endoplasmic reticulum and secretory granules. The secretory granules have been shown to contain, surfactant apo-proteins A, B and D (SP-A, SP-B and SP-D) (1,2), proteases (3), antiproteases (4) and Clara cell secretory protein (CCSP) (5,6). In addition to the large amount of secretory products Clara cells also contain high concentrations of xenobiotic metabolizing enzymes (7) and are therefore a site of xenobiotic mediated damage (8). Along with secretory and metabolic roles, studies of epithelial cell regeneration after experimental injury suggest that Clara cells may function as progenitor cells within the bronchiolar epithelium (9). Studies of Clara cells have been hampered by the difficulty in obtaining and maintaining enriched populations of cells in vitro (10). In an effort to identify novel Clara cell enriched gene products we have isolated rat BE cells highly enriched for Clara cells and generated a cDNA library enriched in Clara cell gene products. Our studies using this library and a simple differential screening approach suggest that the library will be a valuable tool for the identification of other novel protein products enriched in BE cells. MATERIALS AND METHODS Isolation and characterization of rat BE cells. BE cells were isolated from Sprague-Dawley rats (300g) by a modification of the method of Jones et al (11), from an initial single cell suspension prepared as described (12). The final fraction was pelleted at 800g for 10 min and the cells were resuspended in DMEM and plated onto IgG coated bacteriological plastic for 1 hr to remove macrophages and lymphocytes. Cells were harvested by panning and resuspended in DMEM containing 10% fetal calf serum (FCS). Aliquots of the final suspension were cytospun onto Superfrost plus slides (Fisher) and the number of Clara cells were determined by nitroblue tetrazolium (NBT) staining. The enrichment of expression of rat CCSP was followed by Northern blotting of total RNA hybridized with a probe to rat CCSP (7). Production of BE cell cDNA library. BE cell poly A/ RNA was purified directly from fresh cells using the MicroFast Track kit (Invitrogen). A size selected directionally cloned library was produced from 3.6mg of poly A/ RNA
1
Fax: 0171-6066300. E-mail: [email protected]. 877 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5238
/
6907$$$441
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. RNA blot analysis of Clara cell secretory protein expression (CCSP) in a population of BE cells. Total RNA was isolated from the indicated sources and 15mg (lane 1) and 1mg (lanes 2-4) was electrophoresed on a 1% agarose/formaldehyde gel, transferred to nylon and hybridized with a 32P-labelled antisense cRNA probe to rat CCSP. The blot was washed and exposed to film for 10 minutes.
primed with oligo dT primers, using the Uni-Zap XR cDNA vector kit (Stratagene). The library was packaged and amplified once using XL1-Blue MRF1 cells. Differential screening of the BE cell library. 1mg of poly A/ RNA from isolated BE cells and rat liver was used as a template for random primed first strand cDNA probe production using AMV reverse transcriptase and [32P] dCTP. The probes were used to screen duplicate filters each of which had õ 2,000 plaque forming units (pfu) per plate. 20,000pfu were screened in total. The filters were hybridized at 427C overnight using standard buffers and washed at 507C in 0.21SSC/0.1% SDS. Positive BE cell plaques were picked for further analysis. Inserts were recovered into pBS by excision rescue. Clones were arrayed on filters and were hybridized with cDNA probes to CCSP, SP-A, -B, -C and -D. Clones which failed hybridize were sequenced, using T3 and Sequenase 2 (Amersham). One of the novel sequences (CC-4) was chosen for further study. To obtain full length clones, 300,000 pfus of the BE cell library were screened using antisense cRNA probes using standard conditions. These clones were fully sequenced in both directions by the dideoxychain termination method. Nucleotide and protein sequence analysis was performed using the PC/GENE software system. RNA isolation and analysis. Total RNA from primary cells and rat liver was isolated by Caesium chloride centrifugation following dissolution in guanidinium isothiocyanate. RNA samples were electrophoresed, transferred to nylon membranes and analyzed by hybridization with 32P-labelled cRNA antisense probes as described (6).
RESULTS AND DISCUSSION
Isolation and characterization of rat BE cells. Using the method outlined above we were able to isolate ú11107 BE cells per rat. Differential staining with nitro blue tetrazolium revealed that 38{2 % (nÅ22) of the cells were classified as Clara cells (results not shown). To confirm that we were isolating a population of cells highly enriched for Clara cells, total RNA prepared from whole lung, the initial cell digest and the isolated cell population was hybridized with a rat CCSP cRNA probe. The enrichment of CCSP expressing cells compared to total lung is clearly shown in both the primary cell digest and the final cell isolate (Fig. 1). By re-hybridizing the blot with a probe to rat SP-C (a typeII cell marker), the final cell isolate was also shown to be relatively denuded in type II cells compared to the primary cell digest (results not shown). Construction and differential screening of the BE cell cDNA library. The initial titer of the library was 4.9 1 106 pfu and random amplification of individual plaques suggested that the average insert size was ú1 Kb (results not shown). Using a simple differential screening protocol, we picked 47 plaques which gave a strong signal with the BE cell cDNA probe but which gave little detectable signal with the liver cDNA probe. Although the protocol was intended to isolate novel BE cell enriched gene products it also allowed us to confirm that the library contained transcripts whose expression is enriched in BE cells. To identify such clones we initially screened them with probes for CCSP and the surfactant apoproteins A, B, C and D. 23 of the clones hybridized to these probes representing 18 which hybridized with CCSP, 878
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. RNA blot analysis of CC-4 expression in BE cells and liver. 5mg of total RNA from Clara cells (lane 1) and rat liver (lane 2) was electrophoresed on a 1% agarose/formaldehyde gel transferred to nylon and hybridized with a 32P-labelled antisense cDNA probe to CC-4. The blot was washed and exposed to film at 0707C for 2 hours.
3 with SP-A, 1 with SP-B and 1 with SP-C. We then subjected 24 of the remaining hybridization negative clones to a single run of sequencing using the T3 primer. The resulting DNA sequences were searched against available databases and nine were found to correspond to known sequences. There were an additional 4 CCSP clones, 2 were Lysozyme, 1 was ubiquitin, 1 was ribosomal protein L-34 and 1 was ribosomal protein S-3. Characterization of CC4 cDNA. We chose to further characterise one of the 15 unknown clones, CC-4, based on the observation that its mRNA was highly enriched in the BE cell population compared to the liver (Fig. 2). The northern blotting data also suggested that the CC-4 cDNA was not full length and so the library was rescreened to isolate additional clones. From 300,000 pfus screened ú30 positive clones were isolated. Sequencing the largest insert resulted in the identification of a 1612bp cDNA (Fig. 3) (Acession number U61729) which contained a single open reading frame coding for a protein of 271 amino acids with an estimated molecular weight of 29,502. The ATG was not found to be in a particularly favourable context for initiation. A G was present in position /4 which is commonly found but no purine was present at position 03, which occurs in 97% of vertebrate mRNAs (13). mRNAs with highly unfavourable initiation sites have been found to encode regulatory proteins such as growth factors and cytokines and it has been suggested that a weak context may be a deliberated ploy to regulate expression of proteins whose over expression may be harmful to the cell (13). The 5* untranslated region of the cDNA is 176 nucleotides long which is longer than the majority of mRNAs, which are between 20-100 nucleotides (14). A major exception are proto-oncogenes, many of which have been shown to have longer leader sequences than normal. There is no evidence that longer than normal leader sequences have any effect on translational efficiency (13). The protein as a whole contains 14.4% proline with the majority being located in the N-terminal half of the protein. Proline rich proteins (PRP) occur widely in both prokaryotes and eukaryotes. Many are found to contain repetitive proline rich regions (XP)n or (XPX)n which form extended structures or longer proline rich sequences of 5-8 residues in length which are tandemly repeated many times (15). In other PRP, the proline rich regions are arranged in a non-repetative manner, as is the case with proteins which bind to SH3 domains (16). The exact function of proline rich regions in proteins remains unclear, however in many proteins it appears that they mediate functionally important binding reactions (15). The predictated amino acid sequence of CC4 contains no signal peptide suggesting that the protein product is not secreted, however, the region of highly basic amino acids from residues 60-66 (KKRRKKK, underlined in Figure 4) has the characteristics of a potential nuclear localization signal (17). The presence of the helix-breaking amino acid proline to the N-terminal side of the motif at position 59 allows this region to be putatively identified as a class A NLS (18). 879
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. The complete nucleotide and deduced amino acid sequence of CC-4. Numbers refer to the nucleotides of the full length cDNA. The initiation codon is shown in bold. The putative nuclear localization signal is underlined and the shaded areas represent sequence not present in the human homologue B4-2.
Absolute identification of this motif as a functional NLS requires further studies with chimeric cytosolic proteins. Searching of both cDNA and protein databases revealed no significant homology with any previously published rodent sequences. Searching of the human cDNA data base revealed that CC-4 is highly homologous to a recently published cDNA sequence, B4-2 (accession number U03105) isolated by differential screening of a human Natural Killer cell line library (19). Close alignment of the translated open reading frames reveal (Fig.4) that 880
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. A comparison of the deduced amino acid sequence of CC-4 with its human homologue. Identical amino acids are shaded and the dashes indicate regions of no homology.
the two proteins are highly homologous over large regions but they also contain two highly divergent areas. Between amino acids 35-56 and 109-116 of the rat sequence there is no homology (indicated by the shaded areas in Fig. 4). Additionally in the same regions of the human protein there are inserted areas of sequence which are not present in the rat protein (Fig. 4). Such regions of lack of homology in an otherwise highly homologous protein suggest that the rat protein may be alternatively spliced. However sequencing of multiple clones isolated in the second screen suggested that they all contained the same sequence. In addition RT-PCR performed using a series of oligonucleotides surrounding the divergent regions produced only bands of the expected size (results not shown). The observation that CC-4 is highly similar to a human cDNA recently isolated by differential screening of a NK cell line library, suggests that CC-4 isoforms may be expressed in multiple cell types in a cell specific altenatively-spliced fashion. Consistant with this notion the human homologue is expressed in multiple cell types by RNA blot analysis (19). The exact cellular localization of CC4 within the bronchiolar epithelium and identification of its function awaits further experiments and the nature of the specific isoforms will require isolation of the corresponding gene. In conclusion, we have constructed a cDNA library from an enriched population of rat BE cells. We have shown that this library contains abundant clones known to be expressed in BE epithelial cells and shown that it may also be a valuable tool for the identification of novel BE cell enriched gene products. ACKNOWLEDGMENTS The work described in this paper was undertaken whilst C.D.B. was a Post-doctoral fellow in the laboratory of Dr. Jonathan D. Gitlin at Washington University School of Medicine, St. Louis, Missouri. The BE cell populations were provided by Dr. Mike Moxley at St. Louis University.
REFERENCES 1. Phelps, D. S., and Floros, J. (1991). Exp. Lung Res. 17, 985–995. 2. Crouch, E., Purghi, D., Kuan, S. F., and Persson, A. (1992). Am. J. Physiol. 263, L60–L66. 881
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
3. Kido, H., Yokogoshi, Y., Saki, K., Tashiro, M., Fukutomi, A., and Katunuma, N. (1992). J. Biol. Chem. 267, 13573. 4. Mooren, H. W. D., Kramps, J. A., Franken, C., Meijer, C. J. L. M., and Dijkman. (1983). Thorax 38, 180–183. 5. Bedetti, C. D., Singh, J., Sing, G., Katyal, S. L. O., and Wong-chong, M-L. (1987) J. Histochem. Cytochem. 35, 789–794. 6. Hackett, B. P., Shimizu, N., and Gitlin, J. D. (1992). Am. J. Physiol. 262, L399–L406. 7. Chichester, C. H., Philpot, R. M., Weir, A. J., Buckpitt, A. R., and Plopper, C. G. (1991). Am. J. Respir. Cell. Mol. Biol. 4, 179–186. 8. Massaro, G. D., Sing, G., Mason, R., Plopper, C. G., Malkinson, A. M., and Gail, D. B. (1994). Am. J. Physiol. 266, L101–L106. 9. Stripp, B. R., Maxson, K., Mera, R., and Singh, G. (1995). Am. J. Physiol. 267, L791–L799. 10. Oreffo, V. I., Morgan, A., and Richards, R. J. (1990). Environ. Health. Persp. 85, 51–64. 11. Jones, K. G., Holland, J. F., and Fouts, J. R. (1982). Cancer Res. 42, 4658–4663. 12. Dobbs, L. G., Gonzalez, R., and Williams, M. C. (1986). Am. Rev. Respir. Dis. 134, 141–145. 13. Kozak, M. (1991). J. Cell. Biol. 115, 887–903. 14. Kozak, M. (1987). Nuc. Acid. Res. 15, 8125–8148. 15. Williamson, M. P. (1994). Biochem. J. 297, 249–260. 16. Ren, R., Mayer, B. J., Cicchetti, P., and Baltimore, D. (1993). Science 259, 1157–1161. 17. Boulikas, T. (1993). CRC. Crit. Rev. Euk. Gene. Exp. 3, 193–227. 18. Dang, C. V., and Lee, M. W. (1989). J. Biol. Chem. 264, 18019– 19. Chen, J., Lui, L., and Pohajdak, B. (1995). Biochem. Biophys. Acta. 1264, 19–22.
882
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
225, 877–882 (1996)
1266
Generation of a Rat Bronchiolar Epithelial Cell cDNA Library: Isolation of a Proline Rich Protein Highly Enriched in Bronchiolar Epithelial Cells Colin D. Bingle1 Department of Toxicology, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, London, EC1A 7ED, United Kingdom Received July 17, 1996 As a tool for the identification of novel gene products expressed in rat bronchiolar epithelial (BE) cells we generated a cDNA library prepared from BE cells, enriched to contain ú40% Clara cells. Using a simple differential screening strategy we isolated and partially characterized 47 clones, 29 of which corresponded to gene products known to be expressed within the bronchiolar epithelium. A further 15 clones contained sequences which were not present in the EMBL database and therefore represented potential novel BE cell enriched clones. One of these clones, CC-4, was characterized further and shown by northern blotting to be highly enriched in the BE cell population. The deduced protein product of CC-4 is a proline rich protein of 29.5 kd. These studies suggest that this cDNA library may be a useful tool for the identification of novel lung restricted gene products. q 1996 Academic Press, Inc.
Clara (Non-ciliated bronchiolar epithelial) cells are non-mucous secretory cells located in the surface epithelium of the pulmonary airways. These cells contain large amounts of endoplasmic reticulum and secretory granules. The secretory granules have been shown to contain, surfactant apo-proteins A, B and D (SP-A, SP-B and SP-D) (1,2), proteases (3), antiproteases (4) and Clara cell secretory protein (CCSP) (5,6). In addition to the large amount of secretory products Clara cells also contain high concentrations of xenobiotic metabolizing enzymes (7) and are therefore a site of xenobiotic mediated damage (8). Along with secretory and metabolic roles, studies of epithelial cell regeneration after experimental injury suggest that Clara cells may function as progenitor cells within the bronchiolar epithelium (9). Studies of Clara cells have been hampered by the difficulty in obtaining and maintaining enriched populations of cells in vitro (10). In an effort to identify novel Clara cell enriched gene products we have isolated rat BE cells highly enriched for Clara cells and generated a cDNA library enriched in Clara cell gene products. Our studies using this library and a simple differential screening approach suggest that the library will be a valuable tool for the identification of other novel protein products enriched in BE cells. MATERIALS AND METHODS Isolation and characterization of rat BE cells. BE cells were isolated from Sprague-Dawley rats (300g) by a modification of the method of Jones et al (11), from an initial single cell suspension prepared as described (12). The final fraction was pelleted at 800g for 10 min and the cells were resuspended in DMEM and plated onto IgG coated bacteriological plastic for 1 hr to remove macrophages and lymphocytes. Cells were harvested by panning and resuspended in DMEM containing 10% fetal calf serum (FCS). Aliquots of the final suspension were cytospun onto Superfrost plus slides (Fisher) and the number of Clara cells were determined by nitroblue tetrazolium (NBT) staining. The enrichment of expression of rat CCSP was followed by Northern blotting of total RNA hybridized with a probe to rat CCSP (7). Production of BE cell cDNA library. BE cell poly A/ RNA was purified directly from fresh cells using the MicroFast Track kit (Invitrogen). A size selected directionally cloned library was produced from 3.6mg of poly A/ RNA
1
Fax: 0171-6066300. E-mail: [email protected]. 877 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5238
/
6907$$$441
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. RNA blot analysis of Clara cell secretory protein expression (CCSP) in a population of BE cells. Total RNA was isolated from the indicated sources and 15mg (lane 1) and 1mg (lanes 2-4) was electrophoresed on a 1% agarose/formaldehyde gel, transferred to nylon and hybridized with a 32P-labelled antisense cRNA probe to rat CCSP. The blot was washed and exposed to film for 10 minutes.
primed with oligo dT primers, using the Uni-Zap XR cDNA vector kit (Stratagene). The library was packaged and amplified once using XL1-Blue MRF1 cells. Differential screening of the BE cell library. 1mg of poly A/ RNA from isolated BE cells and rat liver was used as a template for random primed first strand cDNA probe production using AMV reverse transcriptase and [32P] dCTP. The probes were used to screen duplicate filters each of which had õ 2,000 plaque forming units (pfu) per plate. 20,000pfu were screened in total. The filters were hybridized at 427C overnight using standard buffers and washed at 507C in 0.21SSC/0.1% SDS. Positive BE cell plaques were picked for further analysis. Inserts were recovered into pBS by excision rescue. Clones were arrayed on filters and were hybridized with cDNA probes to CCSP, SP-A, -B, -C and -D. Clones which failed hybridize were sequenced, using T3 and Sequenase 2 (Amersham). One of the novel sequences (CC-4) was chosen for further study. To obtain full length clones, 300,000 pfus of the BE cell library were screened using antisense cRNA probes using standard conditions. These clones were fully sequenced in both directions by the dideoxychain termination method. Nucleotide and protein sequence analysis was performed using the PC/GENE software system. RNA isolation and analysis. Total RNA from primary cells and rat liver was isolated by Caesium chloride centrifugation following dissolution in guanidinium isothiocyanate. RNA samples were electrophoresed, transferred to nylon membranes and analyzed by hybridization with 32P-labelled cRNA antisense probes as described (6).
RESULTS AND DISCUSSION
Isolation and characterization of rat BE cells. Using the method outlined above we were able to isolate ú11107 BE cells per rat. Differential staining with nitro blue tetrazolium revealed that 38{2 % (nÅ22) of the cells were classified as Clara cells (results not shown). To confirm that we were isolating a population of cells highly enriched for Clara cells, total RNA prepared from whole lung, the initial cell digest and the isolated cell population was hybridized with a rat CCSP cRNA probe. The enrichment of CCSP expressing cells compared to total lung is clearly shown in both the primary cell digest and the final cell isolate (Fig. 1). By re-hybridizing the blot with a probe to rat SP-C (a typeII cell marker), the final cell isolate was also shown to be relatively denuded in type II cells compared to the primary cell digest (results not shown). Construction and differential screening of the BE cell cDNA library. The initial titer of the library was 4.9 1 106 pfu and random amplification of individual plaques suggested that the average insert size was ú1 Kb (results not shown). Using a simple differential screening protocol, we picked 47 plaques which gave a strong signal with the BE cell cDNA probe but which gave little detectable signal with the liver cDNA probe. Although the protocol was intended to isolate novel BE cell enriched gene products it also allowed us to confirm that the library contained transcripts whose expression is enriched in BE cells. To identify such clones we initially screened them with probes for CCSP and the surfactant apoproteins A, B, C and D. 23 of the clones hybridized to these probes representing 18 which hybridized with CCSP, 878
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. RNA blot analysis of CC-4 expression in BE cells and liver. 5mg of total RNA from Clara cells (lane 1) and rat liver (lane 2) was electrophoresed on a 1% agarose/formaldehyde gel transferred to nylon and hybridized with a 32P-labelled antisense cDNA probe to CC-4. The blot was washed and exposed to film at 0707C for 2 hours.
3 with SP-A, 1 with SP-B and 1 with SP-C. We then subjected 24 of the remaining hybridization negative clones to a single run of sequencing using the T3 primer. The resulting DNA sequences were searched against available databases and nine were found to correspond to known sequences. There were an additional 4 CCSP clones, 2 were Lysozyme, 1 was ubiquitin, 1 was ribosomal protein L-34 and 1 was ribosomal protein S-3. Characterization of CC4 cDNA. We chose to further characterise one of the 15 unknown clones, CC-4, based on the observation that its mRNA was highly enriched in the BE cell population compared to the liver (Fig. 2). The northern blotting data also suggested that the CC-4 cDNA was not full length and so the library was rescreened to isolate additional clones. From 300,000 pfus screened ú30 positive clones were isolated. Sequencing the largest insert resulted in the identification of a 1612bp cDNA (Fig. 3) (Acession number U61729) which contained a single open reading frame coding for a protein of 271 amino acids with an estimated molecular weight of 29,502. The ATG was not found to be in a particularly favourable context for initiation. A G was present in position /4 which is commonly found but no purine was present at position 03, which occurs in 97% of vertebrate mRNAs (13). mRNAs with highly unfavourable initiation sites have been found to encode regulatory proteins such as growth factors and cytokines and it has been suggested that a weak context may be a deliberated ploy to regulate expression of proteins whose over expression may be harmful to the cell (13). The 5* untranslated region of the cDNA is 176 nucleotides long which is longer than the majority of mRNAs, which are between 20-100 nucleotides (14). A major exception are proto-oncogenes, many of which have been shown to have longer leader sequences than normal. There is no evidence that longer than normal leader sequences have any effect on translational efficiency (13). The protein as a whole contains 14.4% proline with the majority being located in the N-terminal half of the protein. Proline rich proteins (PRP) occur widely in both prokaryotes and eukaryotes. Many are found to contain repetitive proline rich regions (XP)n or (XPX)n which form extended structures or longer proline rich sequences of 5-8 residues in length which are tandemly repeated many times (15). In other PRP, the proline rich regions are arranged in a non-repetative manner, as is the case with proteins which bind to SH3 domains (16). The exact function of proline rich regions in proteins remains unclear, however in many proteins it appears that they mediate functionally important binding reactions (15). The predictated amino acid sequence of CC4 contains no signal peptide suggesting that the protein product is not secreted, however, the region of highly basic amino acids from residues 60-66 (KKRRKKK, underlined in Figure 4) has the characteristics of a potential nuclear localization signal (17). The presence of the helix-breaking amino acid proline to the N-terminal side of the motif at position 59 allows this region to be putatively identified as a class A NLS (18). 879
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. The complete nucleotide and deduced amino acid sequence of CC-4. Numbers refer to the nucleotides of the full length cDNA. The initiation codon is shown in bold. The putative nuclear localization signal is underlined and the shaded areas represent sequence not present in the human homologue B4-2.
Absolute identification of this motif as a functional NLS requires further studies with chimeric cytosolic proteins. Searching of both cDNA and protein databases revealed no significant homology with any previously published rodent sequences. Searching of the human cDNA data base revealed that CC-4 is highly homologous to a recently published cDNA sequence, B4-2 (accession number U03105) isolated by differential screening of a human Natural Killer cell line library (19). Close alignment of the translated open reading frames reveal (Fig.4) that 880
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. A comparison of the deduced amino acid sequence of CC-4 with its human homologue. Identical amino acids are shaded and the dashes indicate regions of no homology.
the two proteins are highly homologous over large regions but they also contain two highly divergent areas. Between amino acids 35-56 and 109-116 of the rat sequence there is no homology (indicated by the shaded areas in Fig. 4). Additionally in the same regions of the human protein there are inserted areas of sequence which are not present in the rat protein (Fig. 4). Such regions of lack of homology in an otherwise highly homologous protein suggest that the rat protein may be alternatively spliced. However sequencing of multiple clones isolated in the second screen suggested that they all contained the same sequence. In addition RT-PCR performed using a series of oligonucleotides surrounding the divergent regions produced only bands of the expected size (results not shown). The observation that CC-4 is highly similar to a human cDNA recently isolated by differential screening of a NK cell line library, suggests that CC-4 isoforms may be expressed in multiple cell types in a cell specific altenatively-spliced fashion. Consistant with this notion the human homologue is expressed in multiple cell types by RNA blot analysis (19). The exact cellular localization of CC4 within the bronchiolar epithelium and identification of its function awaits further experiments and the nature of the specific isoforms will require isolation of the corresponding gene. In conclusion, we have constructed a cDNA library from an enriched population of rat BE cells. We have shown that this library contains abundant clones known to be expressed in BE epithelial cells and shown that it may also be a valuable tool for the identification of novel BE cell enriched gene products. ACKNOWLEDGMENTS The work described in this paper was undertaken whilst C.D.B. was a Post-doctoral fellow in the laboratory of Dr. Jonathan D. Gitlin at Washington University School of Medicine, St. Louis, Missouri. The BE cell populations were provided by Dr. Mike Moxley at St. Louis University.
REFERENCES 1. Phelps, D. S., and Floros, J. (1991). Exp. Lung Res. 17, 985–995. 2. Crouch, E., Purghi, D., Kuan, S. F., and Persson, A. (1992). Am. J. Physiol. 263, L60–L66. 881
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
3. Kido, H., Yokogoshi, Y., Saki, K., Tashiro, M., Fukutomi, A., and Katunuma, N. (1992). J. Biol. Chem. 267, 13573. 4. Mooren, H. W. D., Kramps, J. A., Franken, C., Meijer, C. J. L. M., and Dijkman. (1983). Thorax 38, 180–183. 5. Bedetti, C. D., Singh, J., Sing, G., Katyal, S. L. O., and Wong-chong, M-L. (1987) J. Histochem. Cytochem. 35, 789–794. 6. Hackett, B. P., Shimizu, N., and Gitlin, J. D. (1992). Am. J. Physiol. 262, L399–L406. 7. Chichester, C. H., Philpot, R. M., Weir, A. J., Buckpitt, A. R., and Plopper, C. G. (1991). Am. J. Respir. Cell. Mol. Biol. 4, 179–186. 8. Massaro, G. D., Sing, G., Mason, R., Plopper, C. G., Malkinson, A. M., and Gail, D. B. (1994). Am. J. Physiol. 266, L101–L106. 9. Stripp, B. R., Maxson, K., Mera, R., and Singh, G. (1995). Am. J. Physiol. 267, L791–L799. 10. Oreffo, V. I., Morgan, A., and Richards, R. J. (1990). Environ. Health. Persp. 85, 51–64. 11. Jones, K. G., Holland, J. F., and Fouts, J. R. (1982). Cancer Res. 42, 4658–4663. 12. Dobbs, L. G., Gonzalez, R., and Williams, M. C. (1986). Am. Rev. Respir. Dis. 134, 141–145. 13. Kozak, M. (1991). J. Cell. Biol. 115, 887–903. 14. Kozak, M. (1987). Nuc. Acid. Res. 15, 8125–8148. 15. Williamson, M. P. (1994). Biochem. J. 297, 249–260. 16. Ren, R., Mayer, B. J., Cicchetti, P., and Baltimore, D. (1993). Science 259, 1157–1161. 17. Boulikas, T. (1993). CRC. Crit. Rev. Euk. Gene. Exp. 3, 193–227. 18. Dang, C. V., and Lee, M. W. (1989). J. Biol. Chem. 264, 18019– 19. Chen, J., Lui, L., and Pohajdak, B. (1995). Biochem. Biophys. Acta. 1264, 19–22.
882
AID
BBRC 5238
/
6907$$$442
08-06-96 09:09:50
bbrca
AP: BBRC