Aggregation of keratin filaments into a tight, orderly array is a prominent feature of terminal differentiation of mammalian cornified epithelia, such as epidermis. Evidence is accumulating that keratin structure is intimately involved in tissue architecture and tensile strength (reviewed in (1) ). The reorganization of keratin filaments at terminal differentiation involves filaggrin, a protein that is synthesized as a large, highly phosphorylated precursor, profilaggrin (reviewed in (2) ). Profilaggrin contains multiple internal repeats (profilaggrin internal repeat (PIR)()) that are almost identical with one another in rodent, although they vary widely in sequence in human profilaggrin. PIRs vary in size (26 kDa in mouse, 42 kDa in rat) and sequence (30-42% identity between mouse and human) between different species, as well as in number, even between strains (typically 10 to 30 PIRs are present)(3, 4, 5, 6, 7, 8, 9, 10) . The size of the PIR is similar to that of filaggrin, and it is often assumed that each PIR represents a filaggrin to be released; however, the boundaries of the PIRs do not correspond to the termini of filaggrin. The offset between the PIR and filaggrin may represent an evolutionary holdover (5) or may play some unknown functional role. Profilaggrin contains a large Ca-binding amino-terminal domain (7, 8) and may also contain a short atypical sequence at the carboxyl terminus(5, 6, 9, 10) . Synthesis of profilaggrin involves phosphorylation by several kinases at up to 21 sites in each PIR(11) . Mature profilaggrin is insoluble and stored in non-membrane-bound cytosolic keratohyalin granules. At terminal differentiation, the granule disperses; at the same time, profilaggrin undergoes dephosphorylation by phosphatase 2A (PP2A) and at least one other phosphatase (12) and proteolytic processing by two endoproteinases and at least two exopeptidases (13, 14, 15) to release the soluble filaggrin. The regulation of granule dispersal and the interaction of the processing products with keratins are not understood, although it seems likely that release is tightly regulated in order to prevent premature collapse of the keratin network.

Human, rat, and mouse PIRs show limited areas of sequence identity that are most extensive around the site of proteolytic processing(5, 10) . Proteolytic processing occurs within a region referred to as the linker region and involves an initial cleavage at a hydrophobic site within this linker region, followed by carboxypeptidase and aminopeptidase cleavages at the new termini to produce the final amino and carboxyl termini of filaggrin(13, 15) . The actual amount of sequence removed during release of filaggrin varies in different situations, so that a specific linker sequence cannot be defined. Processing occurs in two independently regulated stages. Little is known about the regulation of the first stage, although the second stage is regulated by Ca influx through nifedipine-sensitive channels (14) . The first stage in mouse epidermis or cultured rat keratinocytes produces dephosphorylated intermediates consisting of several copies of filaggrin; for example, 2DI (two domain intermediate) refers to an intermediate consisting of two filaggrin domains joined by a linker segment(16) . In mouse profilaggrin, there are two different primary sequences at the linker regions; these two variants differ by the presence of an FYPVY insert (4, 13) and are referred to as and . The first stage of processing involves cleavage of only the linker regions. In rat profilaggrin, there are no differences in the primary sequence of the PIRs(15) , requiring some other explanation to account for the two-stage processing observed in cultured rat keratinocytes(14) .

Inhibitor studies with cultured rat keratinocytes indicate that the first stage endoproteinase (PEP1 = profilaggrin endoproteinase 1) is chymotryptic-like and that the second endoproteinase is a leupeptin-sensitive enzyme, most likely calpain(14) . Earlier studies identified an endoproteinase activity in epidermal extracts which cleaved mouse profilaggrin at the linkers(13) . This endoproteinase activity was inhibited by chymostatin, which also inhibited in vivo processing, indicating that this endoproteinase was indeed PEP1. We now report purification of this enzyme from mouse epidermal extracts and demonstrate that it has the properties expected for PEP1.

MATERIALS AND METHODS

Profilaggrin was purified as described previously(11, 15) . PEP1 activity was detected with an SDS-PAGE assay whereby profilaggrin was deposited as a thin film. To test for evenness of the film on the surface (rather than formation of a firm pellet at the bottom of the tube), the film was stained with Coomassie Blue dye in water; after destaining in water, the evenness of the protein deposit could be judged by visual examination. Assays for protease activity were carried out in 0.5-ml Eppendorf tubes by precipitation of 5 µl of profilaggrin (stock solution in 9 M urea, 50 mM Tris, pH 8) with 95 µl of 5 mM ZnCl at 0 °C for 30 min, followed by centrifugation in an Eppendorf centrifuge for 5 min at 4 °C. After the supernatant was removed, 100 µl of reaction mixture (2 mM dithiothreitol, 0.01 mg/ml pancreatic trypsin inhibitor (Trasylol, Sigma), 50 mM Tris, pH 8.0, and various column fractions or buffer alone) was added. After a timed incubation at 38 °C, reactions were stopped by adding 33 µl of hot 4 Laemmli sample buffer and boiling for 5 min. Half of each sample was analyzed on SDS-PAGE gradient gels as described previously(13) , staining with Coomassie Blue. Products were quantified by densitometric scanning of the stained gel. Reaction rates were then determined by fitting the data to a straight-line equation using Enzfitter software (excluding the zero time point). For analysis of solubility of the products, the assay tubes were respun after the timed incubation, and the supernatants were removed. Supernatants and pellets were analyzed separately.

PEP1 was purified from an epidermal extract, prepared as described previously(13) . The extract was chromatographed in DE52 in 2 mM dithiothreitol, 0.1 mM EDTA, 50 mM Tris, pH 8.0, developed with a gradient 0-0.4 M NaCl, followed by gel filtration on Sephacryl S-300 (1.0 100 cm) in 2 mM dithiothreitol, 0.1 mM EDTA, 100 mM Tris, pH 8.0. After diluting with equal volumes of 2 mM dithiothreitol, 0.1 mM EDTA, the active fractions were applied to phenyl-agarose (Sigma P9533) in the same buffer, using a gradient up to 1 M NaCl to elute the bound proteins.

For kinetic studies with phosphorylation isoforms of P1, the unphosphorylated, monophosphorylated, and diphosphorylated P1 peptides were purified by HPLC on Synchropak C18 (4.6 mm 240 cm) in trifluoroacetic acid/acetonitrile as described previously(15) . Digestion of the profilaggrin peptide P1 by PEP1 was carried out in the same assay buffer as for digestion of profilaggrin; the reaction mixtures were analyzed by Synchropak C18 HPLC. Fractions were collected during the HPLC analyses, lyophilized, and resuspended in 0.1% formic acid, 50% MeOH for mass analysis and sequencing by tandem mass spectrometry as described previously(15) . On-line mass spectrometric analysis of sites of proteolysis of profilaggrin was accomplished by treatment of profilaggrin with PEP I as usual, except that the pancreatic trypsin inhibitor was left out of the reaction mixture; the PEP I was then inhibited with TPCK, followed by digestion of the products with 2% TPCK-treated trypsin. The resulting tryptic peptides were then analyzed by HPLC with the HPLC column directly coupled to a mass spectrometer, as described previously(11, 15) .

The effect of inhibitors was determined by preincubating the PEP1 with each inhibitor for 20 min at room temperature before assaying. Gas-phase sequence analysis was accomplished by transferring the protein to PVDF, followed by gas phase microsequencing directly from excised Coomassie-stained bands(18) .

RESULTS AND DISCUSSION

Development of an Assay for PEP1

Figure 1: Assay of PEP1 activity on Swiss-Webster mouse profilaggrin, comparing the effect of adding PP2A along with the PEP1. Reaction mixtures were analyzed by SDS-PAGE as described under ``Materials and Methods.'' The times of incubation are indicated above each lane. Activity of PEP1 produces specific cleavage of the linker regions, yielding processing products that have masses that correspond to one, two, three, etc. filaggrin units, as indicated at the left (the filaggrin unit had an apparent mass during SDS-PAGE of 30 kDa for products derived from mouse profilaggrin). The profilaggrin in this case did not behave well on SDS-PAGE, producing a smear rather than a discrete band. The specific enzymes added are indicated at the top. The arrowheads at the left indicate the number of filaggrin units in each proteolytic product. Molecular mass markers are indicated on the right (in kDa).

The linker region of rodent profilaggrin contains several adjacent tyrosines that might be cleaved nonspecifically by a number of known proteinase contaminants of epidermal extracts. To determine the amount of nonspecific cleavage, the differential cleavage of and linker regions by PEP1 was exploited. Balb/c mouse profilaggrin has a distribution of linkers which should not produce filaggrin-sized (30-kDa) PEP1 products. This allowed us to distinguish PEP1 from contaminating endoproteinases that would produce filaggrin-sized fragments (see detailed discussion below). A comparison of the results with the two profilaggrin substrates is shown in Fig. 1(using Swiss-Webster profilaggrin) and Fig. 2(using Balb/c profilaggrin). Swiss-Webster profilaggrin was much larger than Balb/c profilaggrin (we estimate that there are more than 30 PIRs in Swiss-Webster profilaggrin, and 13 or 14 in Balb/c profilaggrin), leading to difficulty in resolubilizing the Swiss-Webster profilaggrin after precipitation. To test for recovery of the proteolytic products, profilaggrin which had not been precipitated was analyzed on the same gel. Quantitation of the stained bands by densitometric scanning showed mass recoveries of 85-103% (5 trials), when comparing a nearly complete digest (such as that produced at 60 min in Fig. 1) with the intact, unprecipitated profilaggrin.

Figure 2: Activity of PEP1 from the phenyl agarose activity peak (Fig. 3B) toward Balb/c mouse profilaggrin. Assays were carried out as described in Fig. 1. The time of incubation in minutes is shown above each lane. The numbers at the right indicate the number of PIRs in each product (the unfilled arrows indicate missing components that would be expected if there was random cleavage of the linkers; these sizes were determined by chymotryptic cleavage of the Swiss-Webster profilaggrin). The masses of the products labeled 2 and 3 agree with those predicted for 2 or 3 filaggrin units. However, no product is observed for the mass predicted for the 4 or 5 filaggrin units. Instead, a product that corresponds to an abnormally large 5 is observed. In the higher mass range, only a few of the possible products are seen, suggesting an ordered disassembly of this profilaggrin. The 5b and 6b products from the Balb/c profilaggrin are not multiples of the PIRs; they probably reflect the presence of the amino and carboxyl termini of profilaggrin on those products. Note that at early times there appears to be specific cleavage of only a few of the linker sites, suggesting that PEP1 is discriminating among the linkers.

Figure 3: Purification of PEP1 from rat epidermis. Silver-stained gels of column fractions from DE52 chromatography (A) and phenyl-agarose (B) chromatography. Chromatography conditions are given under ``Materials and Methods.'' The activity toward Swiss-Webster profilaggrin and Balb/c profilaggrin is indicated below the fraction (only those having significant activity are shown). The bands at 50-65 kDa in B are contaminants of the buffers.

Once a relatively pure PEP1 was available, kinetic studies were undertaken to find a way to quantify activity. The kinetic complexity for production of the smaller intermediates (which would require at least two proteolytic events for their release) precluded use of normal enzyme units; however, densitometry indicated that the product containing two filaggrin units (2DI) gave the most reproducible recovery in assays conducted on several days using both Swiss-Webster and Balb/c profilaggrin. When the amount of 2DI was determined by densitometry of the SDS-PAGE gel, a short lag in the production of the smaller products (consisting of one, two, or three filaggrin domains) was revealed, as was expected because two cleavage events are required. To quantify activity, the enzyme solution was diluted so that the rates of 2DI release after 5 min was linear with enzyme concentration, although normal Michaelis-Menten kinetics were not obtained when varying the amount of profilaggrin in the precipitate. This assay was then used to quantify the activity of the protease for estimation of yield during the enzyme purification (Table 1).

PEP1 Purification

The usefulness of the two profilaggrin substrates in evaluating the presence of a contaminant is demonstrated in the DE52 chromatography. Below each lane of the DE52 profile, the optical density of several products is compared. Examination of 2DI produced from either Swiss/Webster or Balb/c profilaggrin indicated the presence of a broad activity peak 10 fractions wide. However, only the first half of this broad peak yielded a 30-kDa product from Balb/c profilaggrin. These data indicated that a contaminating proteinase eluted just before PEP1, but could not be resolved completely until the phenyl-agarose step; the contaminating proteinase did not bind to this resin. Inhibitor sensitivity of the contaminating proteinase indicated it was similar to, if not identical with, mast cell chymase; this identity was supported by amino-terminal microsequencing after electroblotting to polyvinylidene difluoride(17) .

In the most highly purified PEP1 fractions, a 29.5-kDa band could be correlated with activity (Fig. 3B). Two 18-19-kDa contaminants were present in amounts that varied in different preparations. The elution of PEP1 from Sepharose S-300 indicated a mass of about 30 to 35 kDa (assuming a globular configuration), suggesting that the 29.5-kDa PEP1 did not have associated subunits. Besides the small contaminants, some minor bands could be detected after longer silver stain development, so that this PEP1 preparation appeared to be about 50% pure. Comparison of the 29.5-kDa band with the silver-stained standards indicated that approximately 1.5 µg of PEP1 was obtained from the epidermis of 40 newborn mice. Several attempts at further purification resulted in complete loss of the protein, which was not surprising in view of the small amount present. An attempt to microsequence PEP1 after electroeluting to polyvinylidene difluoride did not give any signal above background; this may reflect a blocked amino terminus, although it is more likely that there was simply too little material for sequencing.

PEP1 Inhibitor Profile, pH Optimum, and Lack of Activity toward Small Substrates

Conditions that solubilize profilaggrin, such as high salt(13) , also inhibited PEP1. The pH profile of PEP1 showed activities that were fairly constant from pH 6.5 to 8.5, with a rapid decrease in activity below pH 6.5 and complete inhibition by pH 5.0. A small decrease in activity (about 20%) was seen at pH 9.0. This pH profile is consistent with identification of PEP1 as a cytoplasmic enzyme.

Identification of the Sites of Proteolysis

Figure 4: Sequence of proteolytic products produced by action of PEP1 on the a linker mouse profilaggrin (A) or peptide P1 from rat profilaggrin (B). Each panel includes the sequence of the linker region, where each amino acid is represented by the one-letter code; * indicates a phosphorylated residue. The partial sequences below the main sequence in A are the amino-terminal sequences of the intact two-filaggrin unit product and three-filaggrin unit product, as determined by gas-phase microsequencing after transfer to polyvinylidene difluoride. The sequences in B were determined by MS/MS analysis of HPLC-resolved peptide subfragments produced after incubation of peptide P1 with PEP1. Both the amino and carboxyl termini of the products could be identified, providing clearer evidence of processing events than that provided by the amino-terminal sequencing.

In the second experiment, a peptide containing the proteolytic processing sites of rat profilaggrin (P1, see (15) ) was used as a substrate for PEP1. The products were resolved by reverse-phase HPLC, their masses were determined by mass spectrometry (MS), and their sequence was determined by tandem mass spectroscopy (MS/MS) as described previously(11, 15) . Five products were recovered from the reverse-phase HPLC chromatography; their observed masses were 3089.8, 2763.6, 1165.6, 1328.5, 1491.5 Da/e. Cleavage of P1 at three sites as shown in Fig. 4would give predicted masses of 3090.1, 2763.8, 1165.5, 1328.6, 1491.7, using monoisotopic masses below 1500 Da/e and average masses above 1500 Da/e. The MS/MS spectra resembled those previously reported for some of the amino termini of filaggrin (15) and confirmed the cleavage site. In two cases, both of the products of the cleavage were observed; in the third case, only one product was observed (this was a low yield product, and the other subfragment was probably lost in handling). The cleavage of rat P1 peptide occurred at sites similar to those observed with mouse profilaggrin; however, it was now clear that some cleavage occurred after the YYY as well (Fig. 4B). An adjacent valine was not cleaved, nor was the only other tyrosine in the PIR, which is 35 residues away from the multiple tyrosine site (both of these sites were cleaved by the contaminating proteinase). It thus appears that PEP1 can cleave at any tyrosine in the linker peptide.

Analysis of filaggrin amino and carboxyl termini produced in vivo had previously revealed that processing involved exopeptidase action after an initial endoproteinase cleavage(15) . In order to eliminate the possibility that PEP1 had exopeptidase activity, a time course of the reaction was examined for recovery of the various products (which were separated on reverse phase-HPLC). Product peak heights increased linearly with no lag until the P1 peptide was completely digested. There was no evidence for conversion of the larger fragments to smaller fragments, even up to 20 min after the complete digestion of the P1 peptide and after addition of a second aliquot of PEP1 (not shown). Because an exopeptidase should convert larger fragments to smaller, these data indicate that PEP1 did not have exopeptidase activity.

Effect of Phosphate on PEP1 Activity

Because the results with the P1 peptide indicated that PEP1 activity might be modulated by the presence of phosphate in the linker region, activity of PEP1 on profilaggrin treated with phosphatase 2A (PP2A) was examined, which is capable of nearly complete removal of both phosphates on P1. ()Both mouse and rat profilaggrin were examined. Addition of PP2A and PEP1 together had no effect on proteolytic processing of mouse profilaggrin, compared to PEP1 alone (Fig. 1). To ensure that the phosphatase was active under these conditions, parallel experiments with P-labeled mouse profilaggrin showed that 45% of the radiolabel was removed by the end of the experiment. In fact, when PEP1 plus PP2A was compared with PP2A alone, there was a 2-fold enhancement of release of radiolabel from mouse profilaggrin (not shown). These data indicate that PEP1 processing of mouse profilaggrin is directed exclusively by the primary sequence, although variable phosphorylation in the PIRs may explain the ability of PEP1 to discriminate among the apparently identical linkers observed with Balb/c profilaggrin (Fig. 2).

On the other hand, PP2A added with PEP1 markedly enhanced the rate of proteolytic cleavage of rat profilaggrin, compared to PEP1 alone (particularly obvious at 7 and 15 min in Fig. 5), although the extent of proteolysis at longer times was the same in both cases. In this case, the phosphate content of the peptides derived from rat profilaggrin was determined by directly coupling the HPLC eluate to a mass spectrometer(17) , rather than radiolabeling. No significant difference in the phosphate content of the profilaggrin peptides was detected when comparing PP2A plus PEP1 to PP2A alone (not shown). However, PP2A efficiently removed all phosphate from the P1 peptide derived from rat profilaggrin under these conditions.

Figure 5: Effect of PP2A on PEP1 processing of rat profilaggrin. The wells marked + have PP2A added, while those marked - have only PEP1. The numbers above each pair of lanes indicate the time of incubation. This profilaggrin is particularly hard to resolubilize (see ``Results''). However, the proteolytic products clearly show that the presence of PP2A markedly enhances profilaggrin processing by PEP1. Furthermore, PP2A affects the mobility of the products as well. Parallel experiments indicated that the maximum extent of dephosphorylation was no more than 45% at the longest time point. Molecular mass markers are at the left and arrows at the right indicate the number of filaggrin units (rat filaggrin unit is 42 kDa).

The difference in behavior of rat and mouse profilaggrin toward PP2A and PEP1 may be due to extended structural features of the substrate. A well characterized example of this is found in peptide hormone processing, which occurs at dibasic sites situated in, or next to, -turns(18) . Although structure prediction programs show that profilaggrin has little structure(5, 9, 10) , there is predicted structure in the linker region, with a structure just before and an helix following (the residues predicted to have these structures are all included within the sequence of the P1 peptide). The two phosphorylation sites of P1 occur four residues apart at one end of the predicted helix. It seems plausible that in rat profilaggrin, where the structure is not as hydrophobic, perturbation of the helix by these adjacent phosphates would have more of an effect than in mouse profilaggrin. Indeed, dephosphorylation in rat profilaggrin has a more profound effect on rat profilaggrin structure than on mouse profilaggrin structure, as reflected in the altered mobility of the PEP1 rat profilaggrin products on SDS-PAGE (Fig. 5), while no alteration in mobility was observed with mouse profilaggrin PEP1 products (Fig. 1).

However, regulation of proteolysis by phosphorylation at the P1 sites does not explain adequately the complete absence of filaggrin-sized products in first-stage processing in cultured rat keratinocytes. It is tempting to speculate that more extensive structural features may be relevant to the specific proteolysis of a subset of the linker regions in rat profilaggrin, possibly induced by association with keratin, while mouse profilaggrin cleavage is nearly completely determined by the primary structure adjacent to the proteolysis sites.

Solubility of Profilaggrin Products and Effects of Treatment with Phosphatase 2A

Figure 6: Solubilization of smaller profilaggrin processing products. PEP1 was incubated with Swiss-Webster profilaggrin for the times indicated at the top of each pair of lanes. The enzyme reaction mixtures were removed, and the reaction was stopped by boiling in Laemmli sample buffer. The protein remaining bound to the tube was solubilized with boiling Laemmli sample buffer. Both fractions were analyzed by SDS-PAGE. P indicates the material remaining in the tube, while S indicates the proteins solubilized into the reaction mixture. The molecular mass markers are on the right, and the number of filaggrin units in each band is indicated on the left.

Conclusion

FOOTNOTES

*

This work was supported by National Institutes of Health Grants AR39730 (to K. A. R.) and RR05543 (to Kenneth A. Walsh, University of Washington). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§

To whom correspondence should be addressed: Dept. of Chem. and Biochemistry, Campus Box 215, University of Colorado, Boulder, CO 80303. Tel.: 303-492-4604; Fax: 303-492-5894.

()

The abbreviations used are: PIR, profilaggrin internal repeat; 2DI and 3DI, two- and three-domain intermediate; ESI/MS, electrospray ionization mass spectrometry; Da/e, daltons per unit charge; HPLC, high pressure liquid chromatography; MS/MS, tandem mass spectrometry; PP2A, phosphatase 2A; PEP1, profilaggrin endoproteinase 1; PAGE, polyacrylamide gel electrophoresis; TPCK, tosylphenylalanyl chloromethyl ketone.

()

G. Thulin, E. Kam, B. A. Dale, K. A. Walsh, and K. A. Resing, unpublished data.

ACKNOWLEDGEMENTS

REFERENCES

1. Steinert, P. M. (1993) J. Invest. Dermatol. 100, 729-734 [CrossRef][Medline] [Order article via Infotrieve]
2. Dale, B. A., Presland, R. B., P., F., Kam, E., and Resing, K. A. (1993) Molecular Biology of the Skin , pp. 79-106, Academic Press, New York
3. Resing, K. A., Dale, B. A., and Walsh, K. A. (1985) Biochemistry 24, 4167-4175 [CrossRef][Medline] [Order article via Infotrieve]
4. Rothnagel, J. A., and Steinert, P. M. (1990) J. Biol. Chem. 265, 1862-1865 [Abstract/Free Full Text]
5. Haydock, P. V., and Dale, B. A. (1990) DNA Cell Biol. 9, 251-261 [Medline] [Order article via Infotrieve]
6. Gan, S.-Q., McBride, O. W., Idler, W. W., Markova, N., and Steinert, P. M. (1990) Biochemistry 29, 9432-9440 [CrossRef][Medline] [Order article via Infotrieve]
7. Presland, R. B., Haydock, P. V., Fleckman, P., Nirunsuksiri, W., and Dale, B. A. (1992) J. Biol. Chem. 267, 23772-23781 [Abstract/Free Full Text]
8. Markova, N. G., Marekov, L., Chipev, D. C., Gan, S.-Q., Idler, W. W., and Steinert, P. M. (1993) Mol. Cell. Biol. 13, 613-625 [Abstract/Free Full Text]
9. Rothnagel, J. A., Mehrel, T., Idler, W. W., Roop, D. R., and Steinert, P. M. (1987) J. Biol. Chem. 262, 15643-15648 [Abstract/Free Full Text]
10. McKinley-Grant, L. J., Idler, W. W., Bernstein, I. A., Parry, D. A., Cannizzaro, L., Croce, C. M., Huebner, K., Lessin, S. R., and Steinert, P. M. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 4848-4852 [Abstract/Free Full Text]
11. Resing, K. A., Johnson, R. S., and Walsh, K. A. (1995) Biochemistry 34, 9477-9487 [CrossRef][Medline] [Order article via Infotrieve]
12. Kam, E., Resing, K. A., Lim, S. K., and Dale, B. A. (1992) J. Cell Sci. 106, 219-226 [Abstract]
13. Resing, K. A., Walsh, K. A., Haugen-Scofield, J., and Dale, B. A. (1989) J. Biol. Chem. 264, 1837-1846 [Abstract/Free Full Text]
14. Resing, K. A., Al-Alawi, N., Blomquist, C., Fleckman, P., and Dale, B. A. (1993) J. Biol. Chem. 268, 25139-25145 [Abstract/Free Full Text]
15. Resing, K. A., Johnson, R. S., and Walsh, K. A. (1993) Biochemistry 32, 10036-10045 [CrossRef][Medline] [Order article via Infotrieve]
16. Resing, K. A., Walsh, K. A., and Dale, B. A. (1984) J. Cell Biol. 99, 1372-1378 [Abstract/Free Full Text]
17. Matsudaira, P. (1987) J. Biol. Chem. 262, 10035-10038 [Abstract/Free Full Text]
18. Brakch, N., Rholam, M., Boussetta, H., and Cohen P. (1993) Biochemistry 32, 4925-4930 [CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Kamata, A. Taniguchi, M. Yamamoto, J. Nomura, K. Ishihara, H. Takahara, T. Hibino, and A. Takeda Neutral Cysteine Protease Bleomycin Hydrolase Is Essential for the Breakdown of Deiminated Filaggrin into Amino Acids J. Biol. Chem., May 8, 2009; 284(19): 12829 - 12836. [Abstract] [Full Text] [PDF]

				A. Sandilands, C. Sutherland, A. D. Irvine, and W. H. I. McLean Filaggrin in the frontline: role in skin barrier function and disease J. Cell Sci., May 1, 2009; 122(9): 1285 - 1294. [Abstract] [Full Text] [PDF]

				S. Netzel-Arnett, B. M. Currie, R. Szabo, C.-Y. Lin, L.-M. Chen, K. X. Chai, T. M. Antalis, T. H. Bugge, and K. List Evidence for a Matriptase-Prostasin Proteolytic Cascade Regulating Terminal Epidermal Differentiation J. Biol. Chem., November 3, 2006; 281(44): 32941 - 32945. [Abstract] [Full Text] [PDF]

				C. Leyvraz, R.-P. Charles, I. Rubera, M. Guitard, S. Rotman, B. Breiden, K. Sandhoff, and E. Hummler The epidermal barrier function is dependent on the serine protease CAP1/Prss8 J. Cell Biol., August 1, 2005; 170(3): 487 - 496. [Abstract] [Full Text] [PDF]

				D. R. Hewett, A. L. Simons, N. E. Mangan, H. E. Jolin, S. M. Green, P. G. Fallon, and A. N.J. McKenzie Lethal, neonatal ichthyosis with increased proteolytic processing of filaggrin in a mouse model of Netherton syndrome Hum. Mol. Genet., January 15, 2005; 14(2): 335 - 346. [Abstract] [Full Text] [PDF]

				K. List, R. Szabo, P. W. Wertz, J. Segre, C. C. Haudenschild, S.-Y. Kim, and T. H. Bugge Loss of proteolytically processed filaggrin caused by epidermal deletion of Matriptase/MT-SP1 J. Cell Biol., November 24, 2003; 163(4): 901 - 910. [Abstract] [Full Text] [PDF]

				R. B. Presland and B. A. Dale Epithelial Structural Proteins of the Skin and Oral Cavity: Function in Health and Disease Critical Reviews in Oral Biology & Medicine, January 1, 2000; 11(4): 383 - 408. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Resing, K. A. Articles by Mostad, S. Search for Related Content PubMed PubMed Citation Articles by Resing, K. A. Articles by Mostad, S. Social Bookmarking                      What's this?