THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 266, No. 1, Issue of January 5, pp. 536-539,1991 Printed in U. S A .

The Complete Amino Acid Sequence of the Human Transglutaminase K Enzyme Deduced from the Nucleic Acid Sequences of cDNA Clones*

(Received for publication, June 20, 1990)

Hee Chul Kim@YI, William W. Idler$, In Gyu Kim$, Jung Ho Ha&$, So0 I1 Chung$, and Peter M. Steinert$ll From the $Laboratory of Cellular Development and Oncology, National Institute of Dental Research and the $Laboratory of Skin Biology, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, Maryland 20892

In order to study the expression and role of transglu- taminases in the formation of the cross-linked cell envelope of human epidermis, we have used a synthetic oligonucleotide encoding the consensual active site se- quence of known transglutaminase sequences. By Northern blot analysis, newborn foreskin epidermis expresses three different mRNA species of about 3.7, 3.3, and 2.9 kilobases while normal cultured epidermal keratinocytes express only the 3.7- and 2.9-kilobase species. The largest species corresponds to a known ubiquitous tissue type I1 or transglutaminase C activ- ity, the smallest corresponds to a known type I or transglutaminase K activity, and the mid-sized com- ponent apparently encodes a transglutaminase E activ- ity that has recently been shown to be expressed in terminally differentiating epidermis (Kim, H. C., Lewis, M. S., Gorman, J. L., Park, S. C., Girard, J. E., Folk, J. E. & Chung, S. I. (1990) J. Biol. Chem., in press). Using the active site oligonucleotide as a probe, we have isolated and sequenced cDNA clones encoding the transglutaminase K enzyme. The deduced complete protein sequence has 813-amino acid residues of 89.3 kDa, has a pl of 5.7, and is likely to be an essentially globular protein, which are properties expected from the partially purified enzyme. It shares 49-53% se- quence homology with the other transglutaminases of known sequence, especially in regions carboxyl-ter- minal to the active site, and possesses sequences likely to confer its Ca2+ dependence. Interestingly, its larger size is due to extended sequences on its amino and carboxyl termini, absent on the other transglutami- nases, that may define its unique properties.

During terminal differentiation, mammalian epidermal cells acquire a deposit of protein 10-20 nm thick on the intracellular surface of the plasma membrane which is termed the cornified or cross-linked cell envelope (CE)’ (1, 2). The

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequencefs) reported in thispaper has been submitted

505 71 2. to the GenBankTM/EMBL Data Bank with accession numberfs)

TI Present address: Central Research Institute, Korean Green Cross Corporation, Seoul, Korea.

(1 To whom all correspondence should be addressed: Bldg. 10, Rm. 12N238, L. S. B., N. I. A. M. S., N. I. H., Bethesda, MD 20892.

‘The abbreviations used are: CE, cross-linked cell envelope; NHEK, normal human epidermal keratinocyte (cultured cells); TGase, transglutamase; bp, base pair(s).

CE is the most insoluble component of the epidermis due to cross-linking by disulfide bonds (2-6) as well as by “(7- glutamy1)lysine isodipeptide bonds that are formed by the action of transglutaminases (TGases) (4-7). Several proteins including involucrin (8), keratolinin (9), a cysteine-rich en- velope protein (lo), and the quantitatively major component, loricrin (11, 12), are thought to be components of the CE of the epidermis and perhaps of other stratified squamous epi- thelia, but so far, only loricrin has been shown to be cross- linked to CEs by W(y-glutamy1)lysine bonds (12).

To date, three different TGase activities are thought to be present in mammalian epidermis and other stratified squa- mous epithelial tissues. These include a ubiquitous tissue type I1 enzyme (TGase C) (13-16), a keratinocyte type I activity (TGase K) present in cultured epidermal keratinocytes (17- 21), in other epithelial tissues (20) and also known as TGase B in rat chondrosarcoma (229, and a so-called epidermal activity (TGase E) in hair follicles (23-25) and advanced differentiated epidermal cells (24-27). These three activities differ significantly in terms of their properties. TGase E is neutral (human) or basic (rodent) in charge and is enriched in glycines (27), but other TGase activities are acidic in charge and are not glycine-rich (21, 28). TGase K (90-92 kDa) is membrane-associated (3, 21), while TGase C (80 kDa, Ref. 13, 20) and E (77 kDa, Ref. 27) are soluble. TGase E is a zymogen (27) like the plasma TGase activity factor XIIIa (15). The properties and complete amino acid sequences of guinea pig TGase C (28) and human factor XIIIa (29-31) have been determined, but to date TGase K has not been purified and only partial sequence information is available (32). Interestingly, where known, the amino acid sequences of the active sites of TGases have been conserved (Fig. 1).

In order to understand its role in CE formation during normal and abnormal differentiation, we have determined the complete cDNA sequence of the TGase K enzyme and de- scribe its unique properties.

MATERIALS AND METHODS

Isolation of cDNA Clones-The 68-mer synthetic oligonucleotide identified in Fig. 1 was utilized to screen cDNA libraries in Xgtll constructed from poly(A)-enriched RNA from either human foreskin epidermis (33) or cultured normal human keratinocytes (NHEK) induced to terminally differentiate in culture (34). Positive viral plaques were purified by established procedures (33). The inserts of selected clones were excised with EcoRI and subcloned into pGEM- 32 vectors for propagation or the M13 mp18 vector for DNA sequenc- ing.

Molecular Biology Procedures-Northern blots containing 2 pg of poly(A)-enriched RNA were performed as before (33, 35) and were calibrated with standard markers (Bethesda Research Laboratories). DNA sequencing was done using Sequenase 2 (United States Bio- chemical Corp.) with either the universal or synthetic oligonucleotide

536

Human Transglutaminase K 531

H U l W l XIIIa T"

Guinea pig Ease C

Y G Q C W V F A A V A C T V L R C L G I P T R V V T N F N S T"

Rabbit Ease K

Y G Q C W V F A G V T T T V L R C L G L A T R T V T N F N S

consensus nucleic acid 5 ' " I ' A P sequence

I P T l K f X T X

Anti-sense 5"- CGG T G G

A

Y G Q C W V F A G V F N T F L R C L G I P A R I V T N Y F S

C C A Ca;

FIG. 1. Construction of an oligo- nucleotide to the conserved active site regions of sequenced TGases. The data are from human factor XIIIa (29-31); guinea pig liver TGase C (28), and TGase K from rabbit tracheobron- chial epithelial cells (32). I = deoxyino- sine.

.' I

:46 140

3.3- 3.7- 2.9'

!.37

.35

1 L . J

FIG. 2. Northern blot analysis. Strips containing 2 pg of poly(A)-enriched RNA from human foreskin epidermis (lane I ) or cultured primary human epidermal keratinocytes (lanes 2 and 3) were probed with the 68-mer active site oligonucleotide (lanes 1 and 2) and washed with 2 X SSC (standard sodium citrate) at room temperature for 20 min, and a 1.8-kilobase pair long cDNA clone encoding a portion of TGase K (lane 3) and washed with 0.2 X SSC at 65 "C for 90 min. The positions of the standard RNA markers are shown.

primers (33). Primer extension was done using a [R2P]phosphate- labeled 69-mer synthetic oligonucleotide (36).

Computer Analyses of Sequences-Nucleic acid and protein se- quence, and secondary structure prediction analyses, were performed using the software packages compiled by the Wisconsin Genetics Computer Group (37) and the IBI Pustell sequence software version 2 (International Biotechnologies, Inc).

RESULTS AND DISCUSSION

Northern Blot Analysis-A 68-mer synthetic oligonucleo- tide was constructed to the conserved sequence region of the active sites of three transglutaminases whose sequences have been determined (Fig. 1). When used as a probe on Northern blots, three bands were seen for RNA isolated from foreskins of size about 3.7, 3.3, and 2.9 kilobases of relative abundance about 25:1065 (Fig. 2, lane 1). RNA extracted from NHEK contained 2 bands of 3.7 and 2.9 kilobases (lane 2), apparently of size identical to two of the bands of lane 1, but of relative abundance about 2080 and of several-fold greater amount. This simultaneous demonstration of these species has not been reported before, and establishes that cultured epidermal keratinocytes contain a t least two different TGase activities (Tgase K and C) (17, 21), and that terminally differentiating epidermal cells contain a third activity (TGase E) (27).

Isolation and Characterization of cDNA Clones Encoding Human TGase K-Following screening with the 68-mer active site oligonucleotide (Fig. l), we obtained about 100 first-round signals from the foreskin Xgtll library and more than 1000 signals from the NHEK Xgtll library. The longest clones from both were plaque-purified. Their overlapping sequences (Fig. 3) were identical and represented a total of about 2.6 kilobase pairs including the entire 3"non-coding end and a short poly(A) tail. To obtain the full-length sequence, we constructed a 69-mer oligonucleotide, rescreened the NHEK library, and found clones with an additional 250 bp that contained a likely initiation codon and 5"non-coding se-

. I

1 7 1

quences (Fig. 3). To verify the size of the full-length mRNA, we performed a primer extension experiment using the 69- mer oligonucleotide and found that reverse transcriptase ex- tended it to a total of about 290 bp toward the cap site at the 5'-end, as expected from the sequence data (Fig. 3).

The sequence information contains 2,732 bp which consists of 98 bp of 5'-non-coding sequences, a single open reading frame of 2,439 bp, 178 bp of 3'-non-coding sequences, and 17 bp of the poly(A) tail. Of two potential initiation codons, only the one illustrated in Fig. 3 conforms to a consensus sequence for a typical eukaryote mRNA species (38). The open reading frame encodes a protein of 813-amino acid residues of 89.3 kDa and pl5.7. When used as a probe on a Northern blot and washed at high stringency, the clones recognized only an RNA species of 2.9 kilobases (Fig. 2, lane 3). The likely presence of 100-150 bp of additional poly(A) tail sequences on mRNA but absent on our available cDNA clones can account for the remainder of the mRNA size. Taken together, the following data support the view that our sequence encodes the full- length human TGase K: 1) high homology to the partial rabbit type I TGase K recently reported (32); 2) the size of the protein is similar to that reported for human TGase K (90- 92 kDa) (3, 21); 3) it is a strongly acidic protein as expected (21); and 4) its level of expression is elevated in cells cultured under conditions of low vitamin A concentration that favor differentiation (18, 19, 21, 27). Since the TGase K protein has not yet been purified to permit chemical amino acid sequencing of peptides derived from it (21), we have not been able to otherwise confirm the accuracy of our deduced amino acid sequence. However, the sequence displays marked ho- mologies with other TGases that lends validity to the infor- mation we present (see Fig. 5 below).

During sequencing of our clones, we noted a discrepancy with the partial sequences of the rabbit TGase K clone (32). Fig. 4 shows a portion of a sequencing gel in which we demonstrate that these authors have omitted a single nucleo- tide which resulted in a frame-shift reading error in their sequence.

TGases Constitute a Family of Related Proteins-In Fig. 5 the complete amino acid sequences of the three TGases (our human TGase K, guinea pig liver TGase C and human factor XIIIa) available to date are aligned to maximize homology. Overall, the TGase K (813 residues) shares 34% identity and 49% homology with TGase C (690 residues) and 39% identity and 53% homology with factor XIIIa (731 residues). The alignment reveals numerous extended islands of identity, especially near the active site and in the carboxyl-terminal half of the chains. Presumably, such highly conserved regions are critical determinants of catalytic activity. The major se- quence differences occur at the amino and carboxyl termini, and in fact, most of the larger size of TGase K compared to the other TGases of known sequence is accounted for by the presence of additional sequences on its ends. The data show that the various TGases constitute a family of closely related

Human Transglutaminme K 539

1.25 1.20 1.15 z 1.10 % 1.05

5 A.8: 0.85 0.80 0.75

I I ! I I I ! I I I

I 100 200 300 100 500 600 700 800

100 260 360 460 560 600 100 860

FIG. 6. Secondary structure characteristics of the human TGase K enzyme. These were performed using the IBI Pustell sequence software packages and are based on the analytical methods of Ref. 37. Only the predicted secondary structure, flexibility, and hydrophilicity profiles are shown.

many of the unique properties of the TGase K enzyme. TGase K is a membrane-bound enzyme (3, 21). Searches

for conventional membrane-binding or membrane-spanning regions (37) revealed only one potential membrane-binding site, between residues 576-600. However, since this also cor- responds to a region of high sequence homology with TGase C and factor XIIIa (Fig. 5), both of which are soluble enzymes, it is unlikely that this region is involved in membrane asso- ciation. However, we cannot exclude the possibility of other less regular membrane association or anchoring regions that are known to occur in other proteins (39). Recent reports have shown that the TGase K enzyme is acylated in uiuo by both palmitic and myristic acids through thiol-ester linkages near its amino terminus (40, 41). I t is possible that one or more of the residues in the cysteine-rich cluster at 43-49 are involved in the acylation.

TGases are Ca2+-binding and -dependent enzymes (7). There are two sequence regions that are strikingly acidic (residues 22-28 and 54-63) in the unique amino terminus that may be involved in binding of Ca2+, but further work will be necessary to confirm this possibility.

We note that the TGase K enzyme contains six potential asparagine-linked glycosylation sites (43) and 50 threonine -t- serine residues that are potential phosphorylation sites by one or more of several known protein kinases (42). The presence of these recognition sequences raises the exciting possibility of regulation of TGase K activity in cells by a variety of post-translational modification events.

In conclusion, these specific clones will allow further anal- yses on the gene structure, expression, function(s), and regu- lation of function of the TGase K enzyme system in CE formation in the epidermis and related epithelia.

Acknowledgments-We thank Drs. Wright Caughman, Jeff Gor- man, and Sang Chul Park for their numerous helpful discussions during the course of this work. Dr. Toshihiro Tanaka generously provided several fig of NHEK poly(A)-enriched RNA. Dr. John Stanley kindly provided access to the NHEK-derived XgtII cDNA library.

REFERENCES 1. Odland, G. F. & Reed, T. H. (1967) in Ultrastructure of Normal

and Abnormal Skin (Zelickson, A. S., ed) pp. 54-75, Lea and Febiger, Philadelphia, PA

2. Matoltsy, A. G. & Matoltsy, M. N. (1966) J. Invest. Dermatol.

3. Sun, T.-T. & Green, H. (1976) Cell 9, 511-521 46,127-129

4. 5. 6.

7.

8. 9.

10. 11.

12.

13. 14. 15.

16.

17. 18.

19.

20.

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38. 39.

40.

41.

42.

43.

Rice, R. H. & Green, H. (1977) Cell 11, 417-422 Rice, R. H. & Green, H. (1978) J. Cell Biol. 76, 705-711 Hennings, H., Steinert, P. & Buxman, M. M. (1981) Biochem.

Biophys. Res. Commun. 102,739-745 Goldsmith, L. A. (1983) in Biochemistry and Physiology of the

S k i n (Goldsmith, L. A,, ed) pp. 184-196, Oxford University Press Inc., New York

Rice, R. H. & Green, H. (1979) Cell 18, 681-694 Zettergren, J. G., Peterson, L. L. & Wuepper, K. D. (1984) Proc.

Tezuka, T. & Masae, T . (1987) J . Znuest. Dermatol. 88,47-51 Mehrel, T., Hohl, D., Nakazawa, H., Rothnagel, J . A., Longley,

M. A., Bundman, D., Cheng, C. K., Lichti, U., Bisher, M. E., Steven, A. C., Steinert, P. M., Yuspa, S. H. & Roop, D. R. (1990) Cell 61, 1103-1112

Hohl, D., Mehrel, T., Lichti, U., Yuspa, S. H., Roop, D. R. & Steinert, P. M. (1989) J. Znuest. Dermatol. 92, 445

Chung, S. I. & Folk, J . E. (1972) J. Biol. Chem. 247, 2798-2807 Chung, S.-I. (1972) Ann. N. Y. Acad. Sci. 202, 240-255. Curtis, C. G., Stenberg, P., Brown, K. L., Baron, A., Chen, K.,

Grav, A.. Simpson. I. & Lorand, L. (1974) Biochemistry 13,

Natl. Acad. Sci. U. S. A. 81, 238-242

. - . 3247-3262

Schwarz. M. L.. Pizzo. S. V.. Hill. R. L. & McKee. P. A. (1971) J. Bioi Chem: 246,'5851-5854 '

Thatcher, S. M. & Rice, R. H. (1985) Cell 40,685-695 Lichti, U., Ben, T . & Yuspa, S. H. (1985) J. Biol. Chem. 260,

Park, S. C., Kim, S. Y., Kim, H. C., Thacher, S. & Chung, S. I.

Schmidt, R., Michel, S., Shroot, B. & Reichert, U. (1988) J .

Thacher, S. M. (1989) J. Invest. Dermatol. 92, 578-584 Chang, S.-K. & Chung, S. I. (1986) J . Biol. Chem. 261, 8112-

Chung, S. I. & Folk, J. E. (1972) Proc. Natl. Acad. Sci. U. S. A.

Buxman, M. M. & Wuepper, K. D. (1976) Biochim. Biophys. Acta

Ogawa, H. & Goldsmith, L. A. (1976) J. Biol. Chem. 251, 7281-

Peterson, L. L. & Buxman, M. M. (1981) Biochim. Biophys. Acta

Kim, H. C., Lewis, M. S., Gorman, J . L., Park, S. C., Girard, J . E., Folk, J . E. & Chung, S. I. (1990) J. Bid. Chem. 265,21971- 21978

Ikura, K., Nasu, T., Yokuta, H., Tsuchiya, Y., Sasaki, R. & Chiba, H. (1988) Biochemistry 27, 2898-2905

Grundmann, U., Amann, E., Zettllemeissl, G. & Kupper, H. A. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 8024-8028

Ichinose, A., Hendrickson, L. E., Fujikawa, K. & Davie, E. W. (1986) Biochemistry 25, 6900-6906

Takahashi, N., Takahashi, Y. & Putman, F. W. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 8019-8023

Floyd, E. E. & Jetten, A. M. (1988) Mol. Cell Biol. 9, 4846-4851 McKinely-Grant, L. G., Idler, W. W., Bernstein, I. A,, Parry, D.

A. D., Cannizzaro, L., Croce, C. M., Huebner, K., Lessin, S. R. & Steinert, P. M. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 4848-4852

Stanley, J . R., Tanaka, T., Mueller, S., Klaus-Kovtun, V. & Roop, D. R. (1988) J. Clin. Invest. 82, 1864-1870

Steinert, P. M., Parry, D. A. D., Idler, W. W., Johnson, L. D., Steven, A. C. & Roop, D. R. (1985) J. Biol. Chem. 260, 7142- 7149

Steinert, P. M., Rice, R. H., Roop, D. R., Trus, B. L. & Steven, A. C. (1983) Nature 302, 794-800

Devereux, J., Haeberli, P. & Smithies, 0. (1984) Nucleic Acids Res. 12, 387-395

Kozak, M. (1989) J . Cell Biol. 108, 229-241 Klein, P., Kanehisa, M. & DeLisi, C. (1985) EMBO J. 4, 3625-

Chakravarty, R. & Rice, R. H. (1989) J. Biol. Chem. 264, 625-

Rice, R. H., Rong, X. & Chakravarty, R. (1990) Biochem. J . 265,

Edelman, A. M., Blumenthal, D. K. & Krebs, W. G. (1987) Annu.

Snider, M. D. (1984) in Biology of Carbohydrates (Ginsberg, V. &

1422-1426

(1988) J. Cell Biol. 107, 139 (abstr.)

Invest. Dermatol. 90, 475-497

8121

69,303-308

452,356-369

7288

657,268-276

3631

629

351-357

Reu. Biochen. 56, 567-614

Robbins, P. W.. ed) Vol. 2, pp. 169-213, Wiley, New York