- Email: [email protected]
Role of Sp1 Response Element in Transcription of the Human Transglutaminase 1 Gene
Role of Sp1 Response Element in Transcription of the Human Transglutaminase 1 Gene Bart A. Jessen, Marjorie A. Phillips, Alain Hovnanian,* and Robert H. Rice
Department Of Environmental Toxicology, University Of California, Davis, California, U.S.A.; *The Wellcome Trust Center for Human Genetics, University of Oxford, U.K.
This study addresses the contribution of an Sp1 response element in the proximal promoter of the transglutaminase 1 gene to transcription in normal epidermis and in a case of lamellar ichthyosis lacking transglutaminase 1 activity. The latter exhibited an Sp1 promoter mutation previously hypothesized to suppress transcription. In this study, several experiments indicated that the native Sp1 response element was functional, but it had only a small in¯uence on transcription, and the previously observed mutation had no effect. These experiments involved mobility shift assays and transfections of promoter constructs in which the Sp1 site was mutated or lacking altogether. In addition the proximal 1.6 kb of the promoter from the affected individual was as active in
transfections as the promoter from unaffected individuals. A search for sequence alterations in mRNA transcribed in keratinocytes from the patient revealed a novel single base mutation in codon 661 of the transglutaminase coding region predicted to result in premature termination of protein translation. The presence of this mutation in parental genomic DNA was con®rmed by restriction digestion. Thus the lamellar ichthyosis phenotype in this case is likely attributable to a novel non-sense mutation in the coding region leading to reduced transglutaminase 1 mRNA levels rather than mutation of the Sp1 site. Key words: keratinocyte transglutaminase/lamellar ichthyosis/non-sense mutation/transglutaminase 1. J Invest Dermatol 115:113±117, 2000
P
1987). Although ubiquitously expressed in mice, expression can vary by over 100-fold depending upon cell type and stage of development (Saffer et al, 1991). Sp1 typically acts in conjunction with other transcription factors to activate transcription (Lania et al, 1997). It has importance beyond maintenance of housekeeping functions in many cell types including keratinocytes. For example, an Sp1 response element in the distal region of the involucrin promoter reportedly aids in transcriptional activation by an adjoining AP1 site (Banks et al, 1998). Recent progress in analyzing the human TGM1 promoter provides an opportunity to test the hypothesis that Sp1 is important for transcription of the human gene. Deletion analysis of the 2.2 kb promoter, which is suf®cient to yield proper tissue expression in transgenic mice (Yamada et al, 1997), shows that regions 5¢ to an Sp1 site 86 bp upstream of the transcription start site account for as much as 90% of the transcriptional activity in transiently transfected keratinocytes (Mariniello et al, 1995).1 Present results demonstrate that this Sp1 site accounts for only a small portion of the overall transcriptional activity and identify a non-sense mutation in the coding region of patient LIA4 that accounts for the loss of TGM1 activity.
roper maturation of human epidermal cells involves formation of a corni®ed protein envelope stabilized by transglutaminase cross-linking (Rice and Green, 1977; Green, 1979). Transglutaminase 1 (keratinocyte transglutaminase, the product of the TGM1 gene) appears to be a major contributor to this process, as the stratum corneum of individuals lacking this activity do not form cross-linked envelopes (Jeon et al, 1998) and as a result suffer from severe lipid barrier defects. A number of mutations in the TGM1 coding region have now been identi®ed in patients with lamellar ichthyosis (Huber et al, 1995; Parmentier et al, 1995; Russell et al, 1995), and knock-out of the gene produces a similar syndrome in mice, which leads to neonatal death (Matsuki et al, 1998). A recent survey of lamellar ichthyosis patients identi®ed a case (LIA4) in which expression of TGM1 was absent from the epidermis and from cultured epidermal cells. A mutation in an Sp1 response element in the proximal promoter of the TGM1 gene was associated with profound reduction of TGM1 transcripts, leading to the hypothesis that this change could impair TGM1 transcription (Petit et al, 1997). Instances of single base pair promoter mutations leading to loss of transcription are rare and thus interesting for highlighting critical regulatory elements, prompting this study. One of the ®rst transcription factors identi®ed and molecularly cloned, Sp1 belongs to a subfamily of Cys, His, and zinc ®nger proteins (Kadonaga et al,
MATERIALS AND METHODS Cell culture Normal human epidermal cells (hEp) and the SIK spontaneously immortalized line (Rice et al, 1993) were grown in a 3:1 mixture of Dulbecco±Vogt Eagle's and Ham's F-12 media containing 5% fetal bovine serum, 5 mg insulin per ml, 5 mg transferrin per ml, 0.18 mM adenine, 20 pM triiodothyronine, 10 ng cholera toxin per ml, 10 ng epidermal growth factor per ml, antibiotics and 0.4 ng hydrocortisone per ml (Allen-Hoffman and Rheinwald, 1984). A line of rat bladder
Manuscript received November 8, 1999; revised April 7, 2000; accepted for publication April 26, 2000. Reprint requests to: Dr. R. H. Rice, Department of Environmental Toxicology, University of California, One Shields Avenue, Davis, CA 95616-8588. Email: [email protected] Abbreviations: hEp, human epidermal cells; rB, rat bladder cells; TGM1, transglutaminase 1 0022-202X/00/$15.00
1Jessen BA, Rice RH: Transcriptional regulation of keratinocyte transglutaminase (TGK). FASEB J 12:A130, 1998 (abst. 760)
´ Copyright # 2000 by The Society for Investigative Dermatology, Inc. 113
114
JESSEN ET AL
epithelial cells (rB) exhibiting keratinocyte character (Phillips and Rice, 1983) was cultured in the same way without EGF or cholera toxin. Mouse 3T3 cells were grown in Dulbecco±Vogt Eagle's medium supplemented with 10% bovine serum and lethally irradiated for use as a feeder layer (Rheinwald and Green, 1975) with hEp and SIK. Constructs An EMBL3 clone containing human TGM1 genomic DNA (Phillips et al, 1992) was used as a template for polymerase chain reaction (PCR) to generate the wild-type constructs. Genomic DNA isolated from keratinocytes cultured from patient LIA4 (Petit et al, 1997) was used as a template for PCR of the TGM1 promoter. These products were subcloned into the Promega (Madison, WI) pGL3 basic vector containing a ®re¯y luciferase reporter. Constructs with 3¢ deletions were produced by inserting the indicated region of the TGM1 promoter upstream of the TGM1 basal promoter (±70 to +70). Mutagenesis of the Sp1 site in the promoter was performed using the QuickChange site-directed mutagenesis kit from Stratagene (La Jolla, CA). A pRLCMV plasmid (Promega) containing the cytomegalovirus promoter driving transcription of the renilla luciferase gene was used to normalize for transfection ef®ciency. Large-scale plasmid preparations were puri®ed by Qiagen (Valencia, CA) plasmid kits or CsCl gradients. Transfections Transient transfections of promoter constructs were performed using calcium phosphate coprecipitation (Gorman, 1985). Of each construct, 2.5 pmol (»10 mg of DNA) were cotransfected with 0.1 mg of pRLCMV per 6 cm culture. To optimize precipitate formation, pGL3Basic was added so that each contained 20 mg of DNA. Cultures were harvested 4 d after transfection, and luciferase and renilla activities were assayed using a commercial kit (Promega). In separate experiments TGM1 promoter activities in transfections of 3T3 cells were found to be negligible (not shown).
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
generalized lamellar ichthyosis with large, thick scales and persistent ectropion despite systemic retinoid treatment (Petit et al, 1997).
RESULTS Analysis of Sp1 in¯uence Regions of the TGM1 promoter between ±2.2 kb and either ±1.5 kb, ±1.0 kb, or ±0.5 kb were generated by PCR and subcloned upstream of the TGM1 basal promoter (±70 to +70 bp) in a luciferase reporter construct. These constructs, all lacking the Sp1 site at ±86 bp, were compared in transcriptional activity with the intact 2.2 kb promoter in transient transfections. As shown in Fig 1, the construct with the smallest deletion (containing the sequence ±2.2 through ±0.5 kb) maintained > 60% and as much as 200% of the activity of the 2.2 kb promoter in rB and SIK cultures, respectively. Even the construct containing only 0.7 kb of the distal promoter (±2.2 through ±1.5 kb) stimulated the basal promoter by » 5±15-fold and generated » 20% of the activity of the intact ±2.2 kb promoter. To study the in¯uence of Sp1 mutations on transcription, the binding of keratinocyte nuclear proteins to the TGM1 Sp1 response element was ®rst examined by electrophoretic mobility shift assays and then the in¯uence of several mutated Sp1 sequences was evaluated. The relevant sequences of the wild-type (W1) and
Electrophoretic mobility shift assays Keratinocyte nuclear extracts (3±5 mg of protein) prepared by standard methods (Chodosh, 1993) were incubated for 10 min at room temperature in buffer containing 0.1 M NaCl, 12.5% glycerol, 12.25 mM HEPES (pH 7.9), 0.01% Nonidet P-40, 0.1 mM ethylenediamine tetraacetic acid, 1 mM dithiothreitol and 0.5 mg of poly[d(I-C)]±poly[d(I-C)]. A 50-fold molar excess of unlabeled doublestranded oligonucleotide was subsequently added as competitor where appropriate and incubated for an additional 10 min at room temperature. Other competitors were a commercial consensus Sp1 site (Stratagene), 5¢GATCGATCGGGGCGGGGCGATC-3¢, and the 26 bp sequence 5¢GAACTGCAGCCTCGGGCATAGAGGCT-3¢ from the involucrin gene containing a consensus AP2 site (underlined). For antibody supershift experiments, 2 mg of anti-SP1 (Santa Cruz SC-59X) or 0.2 mg of anti-AP2 (Santa Cruz SC-184) were added and the mixture was incubated on ice for 1 h. Then 50 fmol of 32P radio-labeled doublestranded 26 bp oligonucleotide (W1) comprising the transglutaminase promoter sequence 5¢-GTCCCCATTTCCCGCCCAGAGGCCTGG-3¢ (Sp1 response element underlined) were added and incubated for 10 min at room temperature. In one experiment, a probe truncated by 5 bp at the 3¢ end (W2), removing the last base (C) of the AP2 binding site, was employed. Complexes were separated by electrophoresis through a 4% nondenaturing polyacrylamide gel in buffer containing 45 mM Tris base, 45 mM boric acid, and 1 mM ethylenediamine tetraacetic acid. Phosphorimages of the dried gels are illustrated. Sequence analysis The entire TGM1 coding region and the 5¢ and 3¢ untranslated regions were represented in a series of nine overlapping PCR products prepared using as template cDNA transcribed from RNA from patient LIA4 cultured epidermal cells (sequences available upon request). In particular, a product containing exons 13 and 14 of the TGM1 gene (bp 1939±2306 of the coding sequence) was ampli®ed by PCR from LIA4 cDNA using the forward and reverse primers (5¢-CAAGGAGACCAAGAAGGAAG-3¢ and 5¢-TGTCACTGTTTCATTGCCTC-3¢, respectively). PCR products were sequenced directly in both directions using the ABI BigDye terminator mix, a 377 XL ABI automated sequencer and ABI EditView data analysis software. Restriction-endonuclease analysis Genomic DNA was extracted from peripheral blood cells (John et al, 1991). A region containing exon 13 was ampli®ed by PCR using primers previously described (Petit et al, 1997) using as template genomic DNA from the parents and an unaffected sibling of patient LIA4. The PCR product was digested with AvaII restriction-endonuclease and analyzed by 2% agarose gel electrophoresis. Patient The patient (LIA4) is a 9 y old boy of consanguineous parents of Moroccan origin. He was born as a collodion baby and developed severe
Figure 1. Transient transfections of TGM1 deletion constructs. Transcriptional activities are shown in rB cultures (a) and SIK cultures (b) for the indicated promoter regions (2.2±0.5, 2.2±1.0, and 2.2±1.5 in order of decreasing size) joined to the basal promoter (bp ±70/+70) in PGL3 basic. All but the intact promoter (2.2), included for comparison, lacked the Sp1 site at ±86. Illustrated are the mean 6 SD of three or more independent trials.
VOL. 115, NO. 1 JULY 2000
SP1 ELEMENT IN TRANSGLUTAMINASE-1 PROMOTER
115
Figure 3. Effect of Sp1 mutations on TGM1 promoter activity. Sitedirected mutations in the promoter (2.2 kb) corresponding to those in the oligonucleotides M1, M2, and M3 as indicated were evaluated by transient transfection of SIK cultures in parallel with the native construct (normalized to 100%). The promoter (2.2 kb) from the transglutaminase gene of patient LIA4, prepared by PCR, was assayed in hEp and SIK cultures, which responded the same (the combined result is labeled hEp), in parallel with the native construct (100%). A proximal 0.2 kb promoter construct with mutation corresponding to oligonucleotide M3 was assayed in SIK and rB cultures in parallel with the native 0.2 kb construct (100%). Illustrated are the mean 6 SD of three or more independent trials. Figure 2. Binding of keratinocyte nuclear proteins to Sp1 sequences. Illustrated are the wild-type (W1) and mutant (M1, M2, M3) central sequences of the oligomers employed (a). For W1, the Sp1 core sequence is underlined and an overlapping AP2 site is overlined. The mutant Sp1 sequence M2 (indicated by · ) is that found in the TGM1 promoter in patient LIA4. Parts (b)±(d) show the electrophoretic mobility shift patterns obtained with these oligonucleotides. Assays were performed using SIK nuclear extract and 32P-labeled W1 oligomer (b, d) or W2 (c), a shorter probe missing the 3¢ C of the core AP2 site, at saturating levels, alone (±) or with a 50-fold excess of the unlabeled competitor as indicated above the lanes. Competitor (Comp) Sp1 is a commercial consensus site and the AP2 competitor is a consensus site from the involucrin promoter. Antibodies (Ab) were included (AP2 or Sp1) or not (±), also indicated above each lane. The arrow to the right of each panel shows the Sp1 complex that was disrupted by inclusion of anti-Sp1 antibody, and the asterisk indicates the position of a supershifted AP2 complex.
mutant oligonucleotides (M1, M2, M3) used are shown in Fig 2(a). An Sp1 core sequence is underlined and an overlapping AP2 site is overlined. When W1 was labeled with 32P and incubated with SIK nuclear extract, three distinguishable complexes were formed (Fig 2b, lane 1). All three complexes were competed by a 50-fold molar excess of W1 (lane 2), whereas a commercial Sp1 consensus oligonucleotide competed for the top and bottom bands and partially with the middle band (lane 3). The remainder of the middle band (not competed) was supershifted with an anti-AP2 antibody (lane 4), as indicated by the asterisk (*) to the right of Fig 2(b). Competition with an AP2 oligonucleotide from another gene (involucrin) removed the lower portion of the middle complex, leaving three distinct complexes (lane 5). The lowest mobility complex (top band) disappeared upon incubation with an anti-Sp1 antibody (lane 6), as designated by an arrow to the right of Fig 2(b). The other two bands likely represent binding of other members of the Sp1 transcription factor family (Banks et al, 1998; Medvedev et al, 1999). Lane 7 shows that all complexes were competed by the combination of AP2 and Sp1 oligonucleotides.
Con®rming this interpretation, Fig 2(c) shows the mobility shift using an oligonucleotide probe (W2) derived from W1 by truncation at the 3¢ end through the C at the 3¢ end of the AP2 site, which prevents binding by this transcription factor. The resulting band-shift pattern was identical to that using the longer W1 probe with an AP2 competitor oligonucleotide (lane 1). Moreover an anti-Sp1 antibody removed the complex with lowest mobility (lane 2), indicated by an arrow, whereas an anti-AP2 antibody had no effect (lane 3). The effect of mutations on protein binding to these sites is shown in Fig 2(d). Two mutant oligonucleotides with single base pair changes (M1, M2) and one with a 2 bp change (M3) were employed, where the M2 sequence corresponds to that observed in the TGM1 promoter of patient LIA4. Mutants M1 and M2 did not compete for any of the complexes that were competed by a consensus Sp1 oligonucleotide (lanes 2 and 5). The lowest mobility complex contained Sp1 as indicated by its disappearance upon inclusion of an anti-Sp1 antibody (lanes 3 and 9). Addition of an anti-AP2 antibody had no effect, as these mutants still competed for AP2 (lanes 4 and 7). By contrast, mutant M3 did not compete for any of the complexes (lane 8), and an anti-AP2 antibody produced a supershifted band (lane 10), indicated by an asterisk. Thus, this mutation destroyed both Sp1 and AP2 binding sites. The effect of Sp1 mutation on TGM1 promoter activity was then measured by transient transfection. The results with promoter constructs corresponding to mutations in the M1, M2, and M3 oligonucleotides are shown in Fig 3. In the context of the 2.2 kb promoter, the two single base pair mutations had minimal in¯uence on promoter activity, producing a » 20% decrease in transcription compared with the native sequence assayed in parallel. Consistent with this ®nding, the promoter from the LIA4 patient, which has an Sp1 mutation as appearing in the M2 oligonucleotide, was not detectably different in activity from the native sequence. In the context of the 2.2 kb promoter, the 2 bp mutation seen in oligonucleotide M3 produced a signi®cant reduction in transcrip-
116
JESSEN ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
tional activity in human cells, to » 50% that of the native sequence (Fig 3). In the context of the proximal 0.2 kb promoter, however, a less impressive reduction was evident; the mutant had » 70% or » 90% of the activity of the native 0.2 kb region in human keratinocyte or rat bladder epithelial cultures, respectively. Identi®cation of a non-sense mutation in the TGM1 coding region in patient LIA4 Although low, suf®cient TGM1 mRNA was produced by cultured epidermal cells from the patient to obtain cDNA sequences by reverse transcription±PCR. The entire coding region and nearly all the 5¢ and 3¢ ¯anking regions were included in a series of nine overlapping PCR products, each 0.3±0.5 kb. None differed in size from those obtained using keratinocyte RNA from an unaffected individual (not shown). Sequencing of the PCR products revealed a single mutation that altered the protein primary structure. As illustrated in Fig 4, the alteration was a C to T transition in codon 661, changing a CAG (glutamine) codon to a UAG (non-sense) codon in the af¯icted patient mRNA (Q661X). A premature stop at that point would cause a truncation of the protein by 20%, deleting the ®nal 155 amino acids. This mutation, abolishing an AvaII site, was con®rmed by restriction analysis of genomic DNA from the patient and his family. A 360 bp region in exon 13 (positions 8419±8777 in the gene sequence) was ampli®ed by PCR and digested with AvaII, which cuts the native sequence at codon 661 to give products (among others) of 52 and 179 bp. As shown in Fig 5, these products were indeed obtained using DNA from a healthy control and the unaffected sibling (homozygous for the native sequence), but instead the uncleaved product of 231 bp was given by DNA from the patient (homozygous for the defect). The DNA from each parent showed the cleaved and uncleaved products, consistent with each of them bearing one copy of the same TGM1 mutation.
Figure 4. Identi®cation of TGM1 mutation. Illustrated are the readouts from automated sequencing of reverse transcription±PCR products prepared using RNA from patient LIA4 (b) and of cloned cDNA from an unaffected individual (a). The ®rst base (C) of Q661 is mutated to T in the patient DNA, yielding a stop codon in the mRNA (Q661 X).
DISCUSSION This study revealed that the Sp1 element in the normal TGM1 promoter had little transcriptional activity in human epidermal cells grown by the same method as used for assaying transglutaminase expression from patients. A 2 bp mutation that destroys both Sp1 and AP2 binding to overlapping sites appeared to have a small effect in a truncated promoter and to contribute modestly to transcription in the context of an intact promoter, where these proteins likely interact with transcription factors bound to other response elements. The magnitude of the effect, however, was considerably less impressive than that reported for this site in the rabbit TGM1 promoter (Medvedev et al, 1999). This contrast is possibly due to species or cell type differences, which may also be the reason for the reduced effect of the 2 bp mutation observed in rB compared with SIK cultures. Alternatively, differences in activity could be ascribed to positional differences in human and rabbit Sp1 sites within the TGM1 promoter, located at different distances from the transcription start site (5¢ base of the consensus at ±86 and ±46, respectively). It is unlikely that Sp1 and AP2 proteins bind simultaneously to the TGM1 composite site as competition with single recognition site oligonucleotides and antibody experiments revealed no bands affected in common. This would strongly suggest that the greater effect of the 2 bp mutation on transcriptional activity, compared with single mutations affecting only Sp1 binding, is due to the disruption of AP2 binding. It is interesting to note that not only is the rabbit TGM1 site in a different location than the site in the human promoter, but it also lacks an overlapping AP2 site. Perhaps the positive transcriptional effect of this AP2 site compensates for the diminished activity of the Sp1 site in the human proximal promoter. Consistent with these ®ndings, transfection assays indicated that the single base pair alteration in the Sp1 site of the promoter in patient LIA4 had no detectable effect on transcription. Further sequence analysis of the promoter did not reveal any differences from the wild type other than the observed Sp1 mutation. Thus, no deletions were evident, and the sequences of the regions surrounding the CATAA box and the critical AP1 and CRE
Figure 5. Con®rmation of the mutation site by restriction digestion. A region of exon 13 from a control (C), the LI patients' father (F), mother (M), and unaffected sibling (S), and from the LIA4 (A4) patient was ampli®ed from genomic DNA by PCR. The products were digested with AvaII restriction endonuclease and analyzed by 2% agarose gel electrophoresis. Arrows indicate the major restriction fragments, where the restriction fragment resulting from the mutation is denoted by an asterisk. Numbers at the left edge refer to lengths in base pairs of the size markers (lane L).
response elements were identical to normal.1 Moreover, no diminution of transcription was observed using a 1.6 kb promoter derived from the patient DNA. Sequencing of the coding region, however, did identify a premature stop that can account for the lack of enzyme activity in the patient's epidermal cells. Assuming the structure of TGM1 is homologous to that deduced by X-ray crystallography for factor XIII (Yee et al, 1994), the truncated patient protein lacks the Cterminal barrel domain and most of the adjoining barrel domain, likely destabilizing the conformation and/or enhancing proteolysis. Destabilizing alterations downstream of residue 661 that have been reported include an R687C mutation and a deletion of residues 679±696 (Huber et al, 1997). Although enough mutant TGM1 mRNA was synthesized to permit its reverse transcription±PCR ampli®cation, the levels were clearly below normal (Petit et al, 1997). The basis for this reduction, probably involving RNA processing, transport, or stability, is unclear, but reduction in mRNA levels due to non-sense mutation is often observed (Frischmeyer and Dietz, 1999). We thank Dr Dominique Teillac-Hamel for referring patient LIA4 and Drs Yann Barrandon and Ariane Rochat for providing cultured LIA4 epidermal cells. This work was supported by USPHS grants AR27130, ES07059, ES05707, and ES04699.
VOL. 115, NO. 1 JULY 2000
SP1 ELEMENT IN TRANSGLUTAMINASE-1 PROMOTER
REFERENCES Allen-Hoffman BL, Rheinwald JG: Polycyclic aromatic hydrocarbon mutagenesis of human epidermal keratinocytes in culture. Proc Natl Acad Sci USA 81:7802± 7806, 1984 Banks EB, Crish JF, Welter JF, Eckert RL: Characterization of human involucrin promoter distal regulatory region transcriptional activator elementsÐa role for Sp1 and AP1 binding sites. Biochem J 331:61±68, 1998 Chodosh LA. Mobility shift DNA-binding assay using gel electrophoresis. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (eds). Current Protocols in Molecular Biology. New York: Current Protocols, 1993, pp 12.12.11±12.12.10 Frischmeyer PA, Dietz HC: Nonsense-mediated mRNA decay in health and disease. Hum Mol Genet 8:1873±1900, 1999 Gorman CM. High ef®ciency gene transfer into mammalian cells. In: Glover DM (ed.). DNA Cloning, a Practical Approach. Oxford: IRL Press, 1985, pp 143±190 Green H: The keratinocyte as differentiated cell type. Harvey Lect 74:101±138, 1979 Huber M, Rettler I, Bernasconi K, et al: Mutations of keratinocyte transglutaminase in lamellar ichthyosis. Science 267:525±528, 1995 Huber M, Yee VC, Burri N, Vikerfors E, Lavrijsen APM, Paller AS, Hohl D: Consequences of seven novel mutations on the expression and structure of keratinocyte transglutaminase. J Biol Chem 272:21018±21026, 1997 Jeon S, Djian P, Green H: Inability of keratinocytes lacking their speci®c transglutaminase to form cross-linked envelopes: absence of envelopes as a simple diagnostic test for lamellar ichthyosis. Proc Natl Acad Sci USA 95:687± 690, 1998 John SWM, Weitzner G, Scriver CR: A rapid procedure for extracting genomic DNA from leucocytes. Nucleic Acids Res 19:408, 1991 Kadonaga JT, Carner KR, Masiarz FR, Tjian R: Isolation of cDNA encoding transcription factor Sp1 and functional analysis of the DNA binding domain. Cell 51:1079±1090, 1987 Lania L, De Majello B, De Luca P: Transcriptional regulation by the Sp1 family proteins. Int J Biochem Cell Biol 29:1313±1323, 1997 Mariniello L, Qin Q, Jessen BA, Rice RH: Keratinocyte transglutaminase promoter analysis. Identi®cation of a functional response element. J Biol Chem 270:31358±31363, 1995 Matsuki M, Yamashita F, Ishida-Yamamoto A, et al: Defective stratum corneum and early neonatal death in mice lacking the gene for transglutaminase 1 (keratinocyte transglutaminase). Proc Natl Acad Sci USA 95:1044±1049, 1998
117
Medvedev A, Saunders N, Matsuura H, Chistokhina A, Jetten AM: Regulation of the transglutaminase 1 gene. Identi®cation of DNA elements involved in its transcriptional control in tracheobronchial epithelial cells. J Biol Chem 274:3887±3896, 1999 Parmentier L, Blanchet-Bardon C, Nguyen S, Prud'homme J-F, Dubertret L, Weissenbach J: Autosomal recessive lamellar ichthyosis: Identi®cation of a new mutation in transglutaminase 1 and evidence for genetic heterogeneity. Hum Mol Genet 4:1391±1395, 1995 Petit E, Huber M, Rochat A, et al: Three novel point mutations in the keratinocyte transglutaminase (TGK) gene in lamellar ichthyosis: Signi®cance for mutant transcript level, TGK immunodetection and activity. Eur J Hum Genet 5:218± 228, 1997 Phillips MA, Rice RH: Convergent differentiation in cultured rat cells from nonkeratinizing epithelia: Keratinocyte character and intrinsic differences. J Cell Biol 97:686±691, 1983 Phillips MA, Stewart BE, Rice RH: Genomic structure of keratinocyte transglutaminase: Recruitment of new exon for modi®ed function. J Biol Chem 267:2282±2286, 1992 Rheinwald JG, Green H: Serial cultivation of strains of human epidermal keratinocytes: The formation of keratinizing colonies from single cells. Cell 6:331±344, 1975 Rice RH, Green H: The corni®ed envelope of terminally differentiated human epidermal cells consists of cross-linked protein. Cell 11:417±422, 1977 Rice RH, Steinmann KE, deGraffenried LA, Qin Q, Taylor N, Schlegel R: Elevation of cell cycle control proteins during spontaneous immortalization of human keratinocytes. Mol Biol Cell 4:185±194, 1993 Russell LJ, DiGiovanna JJ, Rogers GR, Steinert PM, Hashem N, Compton JG, Bale SJ: Mutations in the gene for transglutaminase 1 in autosomal recessive lamellar ichthyosis. Nature Genet 9:279±283, 1995 Saffer JD, Jackson SP, Annarella MB: Developmental expression of Sp1 in the mouse. Mol Cell Biol 11:2189±2199, 1991 Yamada K, Matsuki M, Morishima Y, et al: Activation of the human transglutaminase 1 promoter in transgenic mice: Terminal differentiation-speci®c expression of the TGM-lacZ transgene in keratinized strati®ed squamous epithelia. Hum Mol Genet 6:2223±2231, 1997 Yee VC, Pedersen LC, Trong IL, Bishop PD, Stenkamp RE, Teller DC: Threedimensional structure of a transglutaminase: Human blood coagulation factor XIII. Proc Natl Acad Sci USA 91:7296±7300, 1994
Department Of Environmental Toxicology, University Of California, Davis, California, U.S.A.; *The Wellcome Trust Center for Human Genetics, University of Oxford, U.K.
This study addresses the contribution of an Sp1 response element in the proximal promoter of the transglutaminase 1 gene to transcription in normal epidermis and in a case of lamellar ichthyosis lacking transglutaminase 1 activity. The latter exhibited an Sp1 promoter mutation previously hypothesized to suppress transcription. In this study, several experiments indicated that the native Sp1 response element was functional, but it had only a small in¯uence on transcription, and the previously observed mutation had no effect. These experiments involved mobility shift assays and transfections of promoter constructs in which the Sp1 site was mutated or lacking altogether. In addition the proximal 1.6 kb of the promoter from the affected individual was as active in
transfections as the promoter from unaffected individuals. A search for sequence alterations in mRNA transcribed in keratinocytes from the patient revealed a novel single base mutation in codon 661 of the transglutaminase coding region predicted to result in premature termination of protein translation. The presence of this mutation in parental genomic DNA was con®rmed by restriction digestion. Thus the lamellar ichthyosis phenotype in this case is likely attributable to a novel non-sense mutation in the coding region leading to reduced transglutaminase 1 mRNA levels rather than mutation of the Sp1 site. Key words: keratinocyte transglutaminase/lamellar ichthyosis/non-sense mutation/transglutaminase 1. J Invest Dermatol 115:113±117, 2000
P
1987). Although ubiquitously expressed in mice, expression can vary by over 100-fold depending upon cell type and stage of development (Saffer et al, 1991). Sp1 typically acts in conjunction with other transcription factors to activate transcription (Lania et al, 1997). It has importance beyond maintenance of housekeeping functions in many cell types including keratinocytes. For example, an Sp1 response element in the distal region of the involucrin promoter reportedly aids in transcriptional activation by an adjoining AP1 site (Banks et al, 1998). Recent progress in analyzing the human TGM1 promoter provides an opportunity to test the hypothesis that Sp1 is important for transcription of the human gene. Deletion analysis of the 2.2 kb promoter, which is suf®cient to yield proper tissue expression in transgenic mice (Yamada et al, 1997), shows that regions 5¢ to an Sp1 site 86 bp upstream of the transcription start site account for as much as 90% of the transcriptional activity in transiently transfected keratinocytes (Mariniello et al, 1995).1 Present results demonstrate that this Sp1 site accounts for only a small portion of the overall transcriptional activity and identify a non-sense mutation in the coding region of patient LIA4 that accounts for the loss of TGM1 activity.
roper maturation of human epidermal cells involves formation of a corni®ed protein envelope stabilized by transglutaminase cross-linking (Rice and Green, 1977; Green, 1979). Transglutaminase 1 (keratinocyte transglutaminase, the product of the TGM1 gene) appears to be a major contributor to this process, as the stratum corneum of individuals lacking this activity do not form cross-linked envelopes (Jeon et al, 1998) and as a result suffer from severe lipid barrier defects. A number of mutations in the TGM1 coding region have now been identi®ed in patients with lamellar ichthyosis (Huber et al, 1995; Parmentier et al, 1995; Russell et al, 1995), and knock-out of the gene produces a similar syndrome in mice, which leads to neonatal death (Matsuki et al, 1998). A recent survey of lamellar ichthyosis patients identi®ed a case (LIA4) in which expression of TGM1 was absent from the epidermis and from cultured epidermal cells. A mutation in an Sp1 response element in the proximal promoter of the TGM1 gene was associated with profound reduction of TGM1 transcripts, leading to the hypothesis that this change could impair TGM1 transcription (Petit et al, 1997). Instances of single base pair promoter mutations leading to loss of transcription are rare and thus interesting for highlighting critical regulatory elements, prompting this study. One of the ®rst transcription factors identi®ed and molecularly cloned, Sp1 belongs to a subfamily of Cys, His, and zinc ®nger proteins (Kadonaga et al,
MATERIALS AND METHODS Cell culture Normal human epidermal cells (hEp) and the SIK spontaneously immortalized line (Rice et al, 1993) were grown in a 3:1 mixture of Dulbecco±Vogt Eagle's and Ham's F-12 media containing 5% fetal bovine serum, 5 mg insulin per ml, 5 mg transferrin per ml, 0.18 mM adenine, 20 pM triiodothyronine, 10 ng cholera toxin per ml, 10 ng epidermal growth factor per ml, antibiotics and 0.4 ng hydrocortisone per ml (Allen-Hoffman and Rheinwald, 1984). A line of rat bladder
Manuscript received November 8, 1999; revised April 7, 2000; accepted for publication April 26, 2000. Reprint requests to: Dr. R. H. Rice, Department of Environmental Toxicology, University of California, One Shields Avenue, Davis, CA 95616-8588. Email: [email protected] Abbreviations: hEp, human epidermal cells; rB, rat bladder cells; TGM1, transglutaminase 1 0022-202X/00/$15.00
1Jessen BA, Rice RH: Transcriptional regulation of keratinocyte transglutaminase (TGK). FASEB J 12:A130, 1998 (abst. 760)
´ Copyright # 2000 by The Society for Investigative Dermatology, Inc. 113
114
JESSEN ET AL
epithelial cells (rB) exhibiting keratinocyte character (Phillips and Rice, 1983) was cultured in the same way without EGF or cholera toxin. Mouse 3T3 cells were grown in Dulbecco±Vogt Eagle's medium supplemented with 10% bovine serum and lethally irradiated for use as a feeder layer (Rheinwald and Green, 1975) with hEp and SIK. Constructs An EMBL3 clone containing human TGM1 genomic DNA (Phillips et al, 1992) was used as a template for polymerase chain reaction (PCR) to generate the wild-type constructs. Genomic DNA isolated from keratinocytes cultured from patient LIA4 (Petit et al, 1997) was used as a template for PCR of the TGM1 promoter. These products were subcloned into the Promega (Madison, WI) pGL3 basic vector containing a ®re¯y luciferase reporter. Constructs with 3¢ deletions were produced by inserting the indicated region of the TGM1 promoter upstream of the TGM1 basal promoter (±70 to +70). Mutagenesis of the Sp1 site in the promoter was performed using the QuickChange site-directed mutagenesis kit from Stratagene (La Jolla, CA). A pRLCMV plasmid (Promega) containing the cytomegalovirus promoter driving transcription of the renilla luciferase gene was used to normalize for transfection ef®ciency. Large-scale plasmid preparations were puri®ed by Qiagen (Valencia, CA) plasmid kits or CsCl gradients. Transfections Transient transfections of promoter constructs were performed using calcium phosphate coprecipitation (Gorman, 1985). Of each construct, 2.5 pmol (»10 mg of DNA) were cotransfected with 0.1 mg of pRLCMV per 6 cm culture. To optimize precipitate formation, pGL3Basic was added so that each contained 20 mg of DNA. Cultures were harvested 4 d after transfection, and luciferase and renilla activities were assayed using a commercial kit (Promega). In separate experiments TGM1 promoter activities in transfections of 3T3 cells were found to be negligible (not shown).
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
generalized lamellar ichthyosis with large, thick scales and persistent ectropion despite systemic retinoid treatment (Petit et al, 1997).
RESULTS Analysis of Sp1 in¯uence Regions of the TGM1 promoter between ±2.2 kb and either ±1.5 kb, ±1.0 kb, or ±0.5 kb were generated by PCR and subcloned upstream of the TGM1 basal promoter (±70 to +70 bp) in a luciferase reporter construct. These constructs, all lacking the Sp1 site at ±86 bp, were compared in transcriptional activity with the intact 2.2 kb promoter in transient transfections. As shown in Fig 1, the construct with the smallest deletion (containing the sequence ±2.2 through ±0.5 kb) maintained > 60% and as much as 200% of the activity of the 2.2 kb promoter in rB and SIK cultures, respectively. Even the construct containing only 0.7 kb of the distal promoter (±2.2 through ±1.5 kb) stimulated the basal promoter by » 5±15-fold and generated » 20% of the activity of the intact ±2.2 kb promoter. To study the in¯uence of Sp1 mutations on transcription, the binding of keratinocyte nuclear proteins to the TGM1 Sp1 response element was ®rst examined by electrophoretic mobility shift assays and then the in¯uence of several mutated Sp1 sequences was evaluated. The relevant sequences of the wild-type (W1) and
Electrophoretic mobility shift assays Keratinocyte nuclear extracts (3±5 mg of protein) prepared by standard methods (Chodosh, 1993) were incubated for 10 min at room temperature in buffer containing 0.1 M NaCl, 12.5% glycerol, 12.25 mM HEPES (pH 7.9), 0.01% Nonidet P-40, 0.1 mM ethylenediamine tetraacetic acid, 1 mM dithiothreitol and 0.5 mg of poly[d(I-C)]±poly[d(I-C)]. A 50-fold molar excess of unlabeled doublestranded oligonucleotide was subsequently added as competitor where appropriate and incubated for an additional 10 min at room temperature. Other competitors were a commercial consensus Sp1 site (Stratagene), 5¢GATCGATCGGGGCGGGGCGATC-3¢, and the 26 bp sequence 5¢GAACTGCAGCCTCGGGCATAGAGGCT-3¢ from the involucrin gene containing a consensus AP2 site (underlined). For antibody supershift experiments, 2 mg of anti-SP1 (Santa Cruz SC-59X) or 0.2 mg of anti-AP2 (Santa Cruz SC-184) were added and the mixture was incubated on ice for 1 h. Then 50 fmol of 32P radio-labeled doublestranded 26 bp oligonucleotide (W1) comprising the transglutaminase promoter sequence 5¢-GTCCCCATTTCCCGCCCAGAGGCCTGG-3¢ (Sp1 response element underlined) were added and incubated for 10 min at room temperature. In one experiment, a probe truncated by 5 bp at the 3¢ end (W2), removing the last base (C) of the AP2 binding site, was employed. Complexes were separated by electrophoresis through a 4% nondenaturing polyacrylamide gel in buffer containing 45 mM Tris base, 45 mM boric acid, and 1 mM ethylenediamine tetraacetic acid. Phosphorimages of the dried gels are illustrated. Sequence analysis The entire TGM1 coding region and the 5¢ and 3¢ untranslated regions were represented in a series of nine overlapping PCR products prepared using as template cDNA transcribed from RNA from patient LIA4 cultured epidermal cells (sequences available upon request). In particular, a product containing exons 13 and 14 of the TGM1 gene (bp 1939±2306 of the coding sequence) was ampli®ed by PCR from LIA4 cDNA using the forward and reverse primers (5¢-CAAGGAGACCAAGAAGGAAG-3¢ and 5¢-TGTCACTGTTTCATTGCCTC-3¢, respectively). PCR products were sequenced directly in both directions using the ABI BigDye terminator mix, a 377 XL ABI automated sequencer and ABI EditView data analysis software. Restriction-endonuclease analysis Genomic DNA was extracted from peripheral blood cells (John et al, 1991). A region containing exon 13 was ampli®ed by PCR using primers previously described (Petit et al, 1997) using as template genomic DNA from the parents and an unaffected sibling of patient LIA4. The PCR product was digested with AvaII restriction-endonuclease and analyzed by 2% agarose gel electrophoresis. Patient The patient (LIA4) is a 9 y old boy of consanguineous parents of Moroccan origin. He was born as a collodion baby and developed severe
Figure 1. Transient transfections of TGM1 deletion constructs. Transcriptional activities are shown in rB cultures (a) and SIK cultures (b) for the indicated promoter regions (2.2±0.5, 2.2±1.0, and 2.2±1.5 in order of decreasing size) joined to the basal promoter (bp ±70/+70) in PGL3 basic. All but the intact promoter (2.2), included for comparison, lacked the Sp1 site at ±86. Illustrated are the mean 6 SD of three or more independent trials.
VOL. 115, NO. 1 JULY 2000
SP1 ELEMENT IN TRANSGLUTAMINASE-1 PROMOTER
115
Figure 3. Effect of Sp1 mutations on TGM1 promoter activity. Sitedirected mutations in the promoter (2.2 kb) corresponding to those in the oligonucleotides M1, M2, and M3 as indicated were evaluated by transient transfection of SIK cultures in parallel with the native construct (normalized to 100%). The promoter (2.2 kb) from the transglutaminase gene of patient LIA4, prepared by PCR, was assayed in hEp and SIK cultures, which responded the same (the combined result is labeled hEp), in parallel with the native construct (100%). A proximal 0.2 kb promoter construct with mutation corresponding to oligonucleotide M3 was assayed in SIK and rB cultures in parallel with the native 0.2 kb construct (100%). Illustrated are the mean 6 SD of three or more independent trials. Figure 2. Binding of keratinocyte nuclear proteins to Sp1 sequences. Illustrated are the wild-type (W1) and mutant (M1, M2, M3) central sequences of the oligomers employed (a). For W1, the Sp1 core sequence is underlined and an overlapping AP2 site is overlined. The mutant Sp1 sequence M2 (indicated by · ) is that found in the TGM1 promoter in patient LIA4. Parts (b)±(d) show the electrophoretic mobility shift patterns obtained with these oligonucleotides. Assays were performed using SIK nuclear extract and 32P-labeled W1 oligomer (b, d) or W2 (c), a shorter probe missing the 3¢ C of the core AP2 site, at saturating levels, alone (±) or with a 50-fold excess of the unlabeled competitor as indicated above the lanes. Competitor (Comp) Sp1 is a commercial consensus site and the AP2 competitor is a consensus site from the involucrin promoter. Antibodies (Ab) were included (AP2 or Sp1) or not (±), also indicated above each lane. The arrow to the right of each panel shows the Sp1 complex that was disrupted by inclusion of anti-Sp1 antibody, and the asterisk indicates the position of a supershifted AP2 complex.
mutant oligonucleotides (M1, M2, M3) used are shown in Fig 2(a). An Sp1 core sequence is underlined and an overlapping AP2 site is overlined. When W1 was labeled with 32P and incubated with SIK nuclear extract, three distinguishable complexes were formed (Fig 2b, lane 1). All three complexes were competed by a 50-fold molar excess of W1 (lane 2), whereas a commercial Sp1 consensus oligonucleotide competed for the top and bottom bands and partially with the middle band (lane 3). The remainder of the middle band (not competed) was supershifted with an anti-AP2 antibody (lane 4), as indicated by the asterisk (*) to the right of Fig 2(b). Competition with an AP2 oligonucleotide from another gene (involucrin) removed the lower portion of the middle complex, leaving three distinct complexes (lane 5). The lowest mobility complex (top band) disappeared upon incubation with an anti-Sp1 antibody (lane 6), as designated by an arrow to the right of Fig 2(b). The other two bands likely represent binding of other members of the Sp1 transcription factor family (Banks et al, 1998; Medvedev et al, 1999). Lane 7 shows that all complexes were competed by the combination of AP2 and Sp1 oligonucleotides.
Con®rming this interpretation, Fig 2(c) shows the mobility shift using an oligonucleotide probe (W2) derived from W1 by truncation at the 3¢ end through the C at the 3¢ end of the AP2 site, which prevents binding by this transcription factor. The resulting band-shift pattern was identical to that using the longer W1 probe with an AP2 competitor oligonucleotide (lane 1). Moreover an anti-Sp1 antibody removed the complex with lowest mobility (lane 2), indicated by an arrow, whereas an anti-AP2 antibody had no effect (lane 3). The effect of mutations on protein binding to these sites is shown in Fig 2(d). Two mutant oligonucleotides with single base pair changes (M1, M2) and one with a 2 bp change (M3) were employed, where the M2 sequence corresponds to that observed in the TGM1 promoter of patient LIA4. Mutants M1 and M2 did not compete for any of the complexes that were competed by a consensus Sp1 oligonucleotide (lanes 2 and 5). The lowest mobility complex contained Sp1 as indicated by its disappearance upon inclusion of an anti-Sp1 antibody (lanes 3 and 9). Addition of an anti-AP2 antibody had no effect, as these mutants still competed for AP2 (lanes 4 and 7). By contrast, mutant M3 did not compete for any of the complexes (lane 8), and an anti-AP2 antibody produced a supershifted band (lane 10), indicated by an asterisk. Thus, this mutation destroyed both Sp1 and AP2 binding sites. The effect of Sp1 mutation on TGM1 promoter activity was then measured by transient transfection. The results with promoter constructs corresponding to mutations in the M1, M2, and M3 oligonucleotides are shown in Fig 3. In the context of the 2.2 kb promoter, the two single base pair mutations had minimal in¯uence on promoter activity, producing a » 20% decrease in transcription compared with the native sequence assayed in parallel. Consistent with this ®nding, the promoter from the LIA4 patient, which has an Sp1 mutation as appearing in the M2 oligonucleotide, was not detectably different in activity from the native sequence. In the context of the 2.2 kb promoter, the 2 bp mutation seen in oligonucleotide M3 produced a signi®cant reduction in transcrip-
116
JESSEN ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
tional activity in human cells, to » 50% that of the native sequence (Fig 3). In the context of the proximal 0.2 kb promoter, however, a less impressive reduction was evident; the mutant had » 70% or » 90% of the activity of the native 0.2 kb region in human keratinocyte or rat bladder epithelial cultures, respectively. Identi®cation of a non-sense mutation in the TGM1 coding region in patient LIA4 Although low, suf®cient TGM1 mRNA was produced by cultured epidermal cells from the patient to obtain cDNA sequences by reverse transcription±PCR. The entire coding region and nearly all the 5¢ and 3¢ ¯anking regions were included in a series of nine overlapping PCR products, each 0.3±0.5 kb. None differed in size from those obtained using keratinocyte RNA from an unaffected individual (not shown). Sequencing of the PCR products revealed a single mutation that altered the protein primary structure. As illustrated in Fig 4, the alteration was a C to T transition in codon 661, changing a CAG (glutamine) codon to a UAG (non-sense) codon in the af¯icted patient mRNA (Q661X). A premature stop at that point would cause a truncation of the protein by 20%, deleting the ®nal 155 amino acids. This mutation, abolishing an AvaII site, was con®rmed by restriction analysis of genomic DNA from the patient and his family. A 360 bp region in exon 13 (positions 8419±8777 in the gene sequence) was ampli®ed by PCR and digested with AvaII, which cuts the native sequence at codon 661 to give products (among others) of 52 and 179 bp. As shown in Fig 5, these products were indeed obtained using DNA from a healthy control and the unaffected sibling (homozygous for the native sequence), but instead the uncleaved product of 231 bp was given by DNA from the patient (homozygous for the defect). The DNA from each parent showed the cleaved and uncleaved products, consistent with each of them bearing one copy of the same TGM1 mutation.
Figure 4. Identi®cation of TGM1 mutation. Illustrated are the readouts from automated sequencing of reverse transcription±PCR products prepared using RNA from patient LIA4 (b) and of cloned cDNA from an unaffected individual (a). The ®rst base (C) of Q661 is mutated to T in the patient DNA, yielding a stop codon in the mRNA (Q661 X).
DISCUSSION This study revealed that the Sp1 element in the normal TGM1 promoter had little transcriptional activity in human epidermal cells grown by the same method as used for assaying transglutaminase expression from patients. A 2 bp mutation that destroys both Sp1 and AP2 binding to overlapping sites appeared to have a small effect in a truncated promoter and to contribute modestly to transcription in the context of an intact promoter, where these proteins likely interact with transcription factors bound to other response elements. The magnitude of the effect, however, was considerably less impressive than that reported for this site in the rabbit TGM1 promoter (Medvedev et al, 1999). This contrast is possibly due to species or cell type differences, which may also be the reason for the reduced effect of the 2 bp mutation observed in rB compared with SIK cultures. Alternatively, differences in activity could be ascribed to positional differences in human and rabbit Sp1 sites within the TGM1 promoter, located at different distances from the transcription start site (5¢ base of the consensus at ±86 and ±46, respectively). It is unlikely that Sp1 and AP2 proteins bind simultaneously to the TGM1 composite site as competition with single recognition site oligonucleotides and antibody experiments revealed no bands affected in common. This would strongly suggest that the greater effect of the 2 bp mutation on transcriptional activity, compared with single mutations affecting only Sp1 binding, is due to the disruption of AP2 binding. It is interesting to note that not only is the rabbit TGM1 site in a different location than the site in the human promoter, but it also lacks an overlapping AP2 site. Perhaps the positive transcriptional effect of this AP2 site compensates for the diminished activity of the Sp1 site in the human proximal promoter. Consistent with these ®ndings, transfection assays indicated that the single base pair alteration in the Sp1 site of the promoter in patient LIA4 had no detectable effect on transcription. Further sequence analysis of the promoter did not reveal any differences from the wild type other than the observed Sp1 mutation. Thus, no deletions were evident, and the sequences of the regions surrounding the CATAA box and the critical AP1 and CRE
Figure 5. Con®rmation of the mutation site by restriction digestion. A region of exon 13 from a control (C), the LI patients' father (F), mother (M), and unaffected sibling (S), and from the LIA4 (A4) patient was ampli®ed from genomic DNA by PCR. The products were digested with AvaII restriction endonuclease and analyzed by 2% agarose gel electrophoresis. Arrows indicate the major restriction fragments, where the restriction fragment resulting from the mutation is denoted by an asterisk. Numbers at the left edge refer to lengths in base pairs of the size markers (lane L).
response elements were identical to normal.1 Moreover, no diminution of transcription was observed using a 1.6 kb promoter derived from the patient DNA. Sequencing of the coding region, however, did identify a premature stop that can account for the lack of enzyme activity in the patient's epidermal cells. Assuming the structure of TGM1 is homologous to that deduced by X-ray crystallography for factor XIII (Yee et al, 1994), the truncated patient protein lacks the Cterminal barrel domain and most of the adjoining barrel domain, likely destabilizing the conformation and/or enhancing proteolysis. Destabilizing alterations downstream of residue 661 that have been reported include an R687C mutation and a deletion of residues 679±696 (Huber et al, 1997). Although enough mutant TGM1 mRNA was synthesized to permit its reverse transcription±PCR ampli®cation, the levels were clearly below normal (Petit et al, 1997). The basis for this reduction, probably involving RNA processing, transport, or stability, is unclear, but reduction in mRNA levels due to non-sense mutation is often observed (Frischmeyer and Dietz, 1999). We thank Dr Dominique Teillac-Hamel for referring patient LIA4 and Drs Yann Barrandon and Ariane Rochat for providing cultured LIA4 epidermal cells. This work was supported by USPHS grants AR27130, ES07059, ES05707, and ES04699.
VOL. 115, NO. 1 JULY 2000
SP1 ELEMENT IN TRANSGLUTAMINASE-1 PROMOTER
REFERENCES Allen-Hoffman BL, Rheinwald JG: Polycyclic aromatic hydrocarbon mutagenesis of human epidermal keratinocytes in culture. Proc Natl Acad Sci USA 81:7802± 7806, 1984 Banks EB, Crish JF, Welter JF, Eckert RL: Characterization of human involucrin promoter distal regulatory region transcriptional activator elementsÐa role for Sp1 and AP1 binding sites. Biochem J 331:61±68, 1998 Chodosh LA. Mobility shift DNA-binding assay using gel electrophoresis. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (eds). Current Protocols in Molecular Biology. New York: Current Protocols, 1993, pp 12.12.11±12.12.10 Frischmeyer PA, Dietz HC: Nonsense-mediated mRNA decay in health and disease. Hum Mol Genet 8:1873±1900, 1999 Gorman CM. High ef®ciency gene transfer into mammalian cells. In: Glover DM (ed.). DNA Cloning, a Practical Approach. Oxford: IRL Press, 1985, pp 143±190 Green H: The keratinocyte as differentiated cell type. Harvey Lect 74:101±138, 1979 Huber M, Rettler I, Bernasconi K, et al: Mutations of keratinocyte transglutaminase in lamellar ichthyosis. Science 267:525±528, 1995 Huber M, Yee VC, Burri N, Vikerfors E, Lavrijsen APM, Paller AS, Hohl D: Consequences of seven novel mutations on the expression and structure of keratinocyte transglutaminase. J Biol Chem 272:21018±21026, 1997 Jeon S, Djian P, Green H: Inability of keratinocytes lacking their speci®c transglutaminase to form cross-linked envelopes: absence of envelopes as a simple diagnostic test for lamellar ichthyosis. Proc Natl Acad Sci USA 95:687± 690, 1998 John SWM, Weitzner G, Scriver CR: A rapid procedure for extracting genomic DNA from leucocytes. Nucleic Acids Res 19:408, 1991 Kadonaga JT, Carner KR, Masiarz FR, Tjian R: Isolation of cDNA encoding transcription factor Sp1 and functional analysis of the DNA binding domain. Cell 51:1079±1090, 1987 Lania L, De Majello B, De Luca P: Transcriptional regulation by the Sp1 family proteins. Int J Biochem Cell Biol 29:1313±1323, 1997 Mariniello L, Qin Q, Jessen BA, Rice RH: Keratinocyte transglutaminase promoter analysis. Identi®cation of a functional response element. J Biol Chem 270:31358±31363, 1995 Matsuki M, Yamashita F, Ishida-Yamamoto A, et al: Defective stratum corneum and early neonatal death in mice lacking the gene for transglutaminase 1 (keratinocyte transglutaminase). Proc Natl Acad Sci USA 95:1044±1049, 1998
117
Medvedev A, Saunders N, Matsuura H, Chistokhina A, Jetten AM: Regulation of the transglutaminase 1 gene. Identi®cation of DNA elements involved in its transcriptional control in tracheobronchial epithelial cells. J Biol Chem 274:3887±3896, 1999 Parmentier L, Blanchet-Bardon C, Nguyen S, Prud'homme J-F, Dubertret L, Weissenbach J: Autosomal recessive lamellar ichthyosis: Identi®cation of a new mutation in transglutaminase 1 and evidence for genetic heterogeneity. Hum Mol Genet 4:1391±1395, 1995 Petit E, Huber M, Rochat A, et al: Three novel point mutations in the keratinocyte transglutaminase (TGK) gene in lamellar ichthyosis: Signi®cance for mutant transcript level, TGK immunodetection and activity. Eur J Hum Genet 5:218± 228, 1997 Phillips MA, Rice RH: Convergent differentiation in cultured rat cells from nonkeratinizing epithelia: Keratinocyte character and intrinsic differences. J Cell Biol 97:686±691, 1983 Phillips MA, Stewart BE, Rice RH: Genomic structure of keratinocyte transglutaminase: Recruitment of new exon for modi®ed function. J Biol Chem 267:2282±2286, 1992 Rheinwald JG, Green H: Serial cultivation of strains of human epidermal keratinocytes: The formation of keratinizing colonies from single cells. Cell 6:331±344, 1975 Rice RH, Green H: The corni®ed envelope of terminally differentiated human epidermal cells consists of cross-linked protein. Cell 11:417±422, 1977 Rice RH, Steinmann KE, deGraffenried LA, Qin Q, Taylor N, Schlegel R: Elevation of cell cycle control proteins during spontaneous immortalization of human keratinocytes. Mol Biol Cell 4:185±194, 1993 Russell LJ, DiGiovanna JJ, Rogers GR, Steinert PM, Hashem N, Compton JG, Bale SJ: Mutations in the gene for transglutaminase 1 in autosomal recessive lamellar ichthyosis. Nature Genet 9:279±283, 1995 Saffer JD, Jackson SP, Annarella MB: Developmental expression of Sp1 in the mouse. Mol Cell Biol 11:2189±2199, 1991 Yamada K, Matsuki M, Morishima Y, et al: Activation of the human transglutaminase 1 promoter in transgenic mice: Terminal differentiation-speci®c expression of the TGM-lacZ transgene in keratinized strati®ed squamous epithelia. Hum Mol Genet 6:2223±2231, 1997 Yee VC, Pedersen LC, Trong IL, Bishop PD, Stenkamp RE, Teller DC: Threedimensional structure of a transglutaminase: Human blood coagulation factor XIII. Proc Natl Acad Sci USA 91:7296±7300, 1994