The human homologue of the yeast CHL1 gene is a novel keratinocyte growth factor-regulated gene.

Keratinocyte growth factor (KGF) is a potent and specific mitogen for different types of epithelial cells, including keratinocytes of the skin. To gain insight into the mechanisms of KGF action in this tissue, we attempted to identify genes that are regulated by KGF in keratinocytes. Using the differential display reverse transcription polymerase chain reaction technology, a gene was identified which was strongly induced in these cells by treatment with KGF but not with serum growth factors or pro-inflammatory cytokines. This gene seems to be part of a multigene family as assessed by Southern blot analysis. Molecular cloning and sequencing of the full-length cDNA revealed a strong homology with the yeast CHL1 gene. The latter encodes a putative helicase, which is involved in correct chromosome transmission and cell cycle progression. Furthermore, the CHL1 gene product and the protein encoded by the novel KGF-regulated gene were identical in size, indicating that we had cloned the human CHL1 homologue. This finding suggests a novel and specific role of KGF in correct chromosome segregation and/or cell cycle progression.

in these respects from other epithelial cell mitogens such as epidermal growth factor (EGF), suggesting the existence of KGF-specific target genes, which are responsible for these effects. To gain insight into the molecular mechanisms of KGF action, we have used the differential display RT-PCR (DDRT-PCR) technology to identify and clone new genes that are specifically regulated by KGF. Using this strategy, we have recently identified a non-selenium glutathione peroxidase as a KGF-regulated gene, suggesting a role of KGF in the detoxification of reactive oxygen species. 2 Here we report the identification and cDNA cloning of a novel KGF-regulated gene from cultured keratinocytes. Interestingly the novel cDNA revealed a striking homology to the yeast CHL-1 gene that encodes a putative helicase involved in chromosome transmission and normal cell cycle progression. This finding suggests a novel role of KGF in these processes.
RNA Isolation and RNase Protection Assays-Total cellular RNA and RNase protection assays were carried out as described (13,14). Probe DNAs: (i) 297-base pair cDNA fragment corresponding to the 3Ј-end of the KRG-2 cDNA. This fragment had been obtained by DDRT-PCR (ii); 275-base pair probe corresponding to amino acids 664 -756 of the KRG-2 open reading frame.
Northern Blot Analysis-Isolation of polyadenylated RNA was performed by affinity chromatography on oligo(dT)-cellulose, using a Pharmacia RNA isolation kit. 3 g of polyadenylated RNA from quiescent HaCaT cells and KGF-treated HaCaT cells were used for Northern blot analysis as described (14), using the 32 P-labeled DDRT-PCR cDNA fragment as a probe.
Southern Blot Analysis-Chromosomal DNA from human fibroblasts was digested with HindIII, EcoRI, or BamHI restriction endonucleases, respectively, fractionated by agarose gel electrophoresis and transferred to nitrocellulose filters. Filters were hybridized under high stringency conditions with the 32 P-labeled cDNA probe described above.
Cloning of a Full-length KRG-2 cDNA-To obtain a full-length KRG-2 cDNA, a cDNA library was generated from 6 g of polyadenylated RNA of KGF-stimulated HaCaT cells using the Uni-Zap XR cDNA cloning kit (Stratagene) and the Gigapack Gold III packaging extract (Stratagene) as described by the manufacturer. Filters were hybridized and washed at high stringency using standard methods (15). The DDRT-PCR cDNA fragment was used as a probe.
In Vitro Transcription/Translation-For in vitro transcription/ translation studies the complete KRG-2 cDNA was subcloned into the transcription vector pBluescript KSIIϩ (Stratagene). 1 g of this plasmid was used as a template in a coupled reticulocyte lysate in vitro transcription/translation system (Promega) using T3 RNA polymerase and [ 35 S]methionine (1000 Ci/mmol) (Amersham Corp.). Proteins were * This work was supported by grants from the Deutsche Forschungsgemeinschaft (WE 1983/1-1 and WE 1983/1-2) and a Hermann and Lilly Schilling Award (to S. W.). 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 sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) X99583 (hCHL1).

RESULTS AND DISCUSSION
Identification of a KGF-regulated Gene by DDRT-PCR-To elucidate the mechanisms of KGF action in keratinocytes, we attempted to identify genes that are regulated by this growth factor in keratinocytes. For this purpose we used the DDRT-PCR technology, which allows the identification of differentially expressed genes within different cell populations. The HaCaT keratinocyte cell line, which is known to express functional KGF receptors, 3 was used for these studies. Quiescent cells were treated for 1.5, 5, and 8 h with purified KGF. RNAs from three independent experiments were analyzed by DDRT-PCR for differentially expressed genes in KGF-treated and nontreated cells. One of the fragments was exclusively detected after amplification of cDNA from KGF-stimulated cells but not from quiescent or FCS-treated keratinocytes (data not shown). It was therefore isolated from the gel, reamplified, and cloned. Induction of the corresponding gene by KGF was confirmed by RNase protection assay (Fig. 1A) and Northern blot analysis (Fig. 1B). The gene was expressed at low levels in quiescent keratinocytes but was strongly induced within 5-8 h after addition of KGF. In contrast to KGF, induction of this gene was not seen with FCS (Fig. 1A). Since this is the second KGFregulated gene that we identified, it was designated KRG-2. Northern blot analysis demonstrated the presence of a single 4.3-kilobase mRNA (Fig. 1B).

Induction of the Novel Gene in Keratinocytes Is a Specific
Effect of KGF-Since expression of KRG-2 was only detected in KGF-stimulated cells but not in serum-treated keratinocytes, we further determined the specificity of this effect for KGF. For this purpose, quiescent HaCaT cells were treated with different growth factors and cytokines and analyzed by RNase protection assay for KRG-2 expression. As shown in Fig. 1, A and B, KGF strongly induced expression of the novel gene, whereas EGF, another potent keratinocyte mitogen, as well as the proinflammatory cytokine IL-1␤, had no significant effect (Fig. 1C). By contrast, inhibitors of keratinocyte proliferation, such as TGF-␤1 and TNF-␣, caused a slight reduction of KRG-2 expression (Fig. 1C). Taken together, these findings suggest that induction of KRG-2 is a novel and specific effect of KGF.
Multiple KRG-2-like Genes Exist in the Human Genome-To determine if KRG-2 is a unique gene or a member of a multigene family, human chromosomal DNA was digested with different restriction enzymes and used for Southern blot analysis. The fragment that had been obtained by DDRT-PCR was used as a probe. As shown in Fig. 2, multiple bands of variable intensities were obtained after a high stringency wash of the filter, suggesting the existence of several closely related genes in the human genome.
KRG-2 Is Highly Homologous to the Yeast CHL-1 Gene-To further characterize the KRG-2 gene and its product, we isolated the full-length cDNA from a cDNA library of KGF-treated keratinocytes. The longest insert (3.8 kilobase) was sequenced from both strands. As shown in Fig. 3A, the cDNA consists of a 212-nucleotide 5Ј-noncoding region, a 2568-nucleotide open reading frame, and a 975-nucleotide 3Ј-noncoding region, whereby the last 297 nucleotides are identical to the fragment obtained by DDRT-PCR. The cDNA encodes a protein of approximately 100 kDa as determined by in vitro transcription/ translation studies (Fig. 4). Using a fragment from the open reading frame as a template for RNase protection assays, we confirmed the results obtained with the probe from the 3Ј-end, which are shown in Fig. 1. Sequence comparison of the nucleotide sequence with known sequences from the EMBL data base demonstrated a significant homology with the yeast CHL1 gene. The latter had been identified in a screen for yeast mutants with decreased chromosome transmission fidelity (16). Mutants lacking the CHL1 gene exhibit a 200-fold increase in the rate of chromosome III missegregation per cell division due to sister chromatid loss and sister chromatid nondisjunction (17). Furthermore, these mutants display a signif-  2. KRG-2 is a member of a multigene family. 10 g of human chromosomal DNA were digested with HindIII (H), EcoRI (E), or BamHI (B) restriction endonucleases, respectively, and analyzed by Southern blot analysis using a 32 P-labeled cDNA corresponding to the DDRT-PCR fragment. Hybridization was performed at high stringency conditions. icant delay in cell cycle progression in G 2 /M (17). The CHL1 gene encodes an 861-amino acid protein, which is similar in size to the KRG-2 protein. Furthermore, both proteins revealed a 32% identity and a 55% similarity at the amino acid level (Fig. 3B), suggesting that KRG-2 is the human homologue of CHL1. The most striking homologies between the yeast and human proteins were observed in regions which represent functional elements. These include the A and B motifs of ATP binding proteins, whereby the latter resembles the modified B motif present in proteins with helicase activity. These enzymes are involved in many biological processes that require unwinding of double-stranded DNA and RNA, such as DNA replication and repair, transcription, splicing, translation, and also the segregation of chromosomes at mitosis (for review, see Refs. 18 and 19). In addition, the CHL1 product and the KRG-2 proteins contain a highly conserved helix-turn-helix motif (for review, see Ref. 20), suggesting that they bind to DNA.
Recently, two closely related human genes have been identified that might encode the KRG-2 protein. Thus, the published sequence of the carboxyl-terminal region of one of these genes was almost identical to our cloned cDNA (21). A large portion of these genes has been duplicated as part of a larger human telomeric repeat sequence found on many chromosomes. These results might provide an explanation for the multiple bands that we detected by Southern blot analysis.
The function of the KRG-2 gene product is presently unknown. However, the striking similarity with the CHL1 product suggests that the novel cDNA encodes an ATP-binding protein with helicase activity, which might play a role in chromosome segregation and cell cycle progression. The latter hypothesis is supported by the detection of significantly higher levels of KRG-2 mRNA in exponentially growing keratinocytes compared with resting keratinocytes (data not shown). Furthermore, preliminary data from our laboratory suggest that KRG-2 is only expressed at extremely low levels in normal human skin, which includes very few proliferating cells. Interestingly, both the CHL1 and KRG-2 proteins contain two PEST sequences (22), which are found in proteins with short halflives such as many cell cycle proteins, thus providing further evidence for a role of these genes in cell cycle regulation. The exclusive induction of KRG-2 expression by KGF suggests a specific role of KGF for cell cycle progression and possibly chromosome segregation in keratinocytes. FIG. 3. cDNA organization and sequence of the KRG-2 protein. A full-length cDNA was isolated from a cDNA library of KGFstimulated keratinocytes using a 32 P-labeled cDNA fragment corresponding to the DDRT-PCR fragment. A, organization of the KRG-2 cDNA. The location of the fragment obtained by DDRT-PCR is indicated with a black bar at the 3Ј-end of the cDNA. B, sequence comparison of the KRG-2 protein with the yeast CHL1 protein. Amino acids that are identical in these proteins are marked with gray boxes. Sequences that are marked with triangles in A and with black bars in B represent the helicase domain (1), an additional ATP-binding site (2) and a helixturn-helix motif (3) .   FIG. 4. CHL-1 encodes a 100-kDa protein. The full-length KRG-2 cDNA was used as a template for in vitro transcription/translation using a coupled reticulocyte lysate system. Translation products were labeled with [ 35 S]methionine, separated on a 7.5% polyacrylamide gel, and analyzed by autoradiography (lane 2). Lane 1 represents the negative control where the reactions were carried out without a template.