Abstract
To identify genes upregulated during the process of ciliated cell differentiation of airway epithelial cells, differential display was used to compare RNA from rat tracheal epithelial (RTE) cells cultured under conditions that inhibit/promote ciliated cell differentiation. Several partial complementary DNAs (cDNAs) were identified whose expression was regulated coordinately with ciliated cell differentiation. One of these, KPL2, detected a messenger RNA transcript of ∼ 6 kb when used as a probe on Northern blots of RNA from ciliated cultures but was undetectable in RNA from nonciliated cultures. Sequencing of overlapping clones obtained by a modified rapid amplification of cDNA ends procedure generated a complete cDNA sequence that exhibited no significant homology to sequences in GenBank, indicating that KPL2 is a novel gene. Southern analysis demonstrated that KPL2 exists as a single-copy gene. KPL2 contains a long open reading frame predicted to code for a protein of > 200 kD. Several putative functional motifs are present in the protein, including a calponin homology domain, three nuclear localization signals, a consensus P-loop, and a proline-rich region, suggesting that KPL2 has a unique function. KPL2 was undetectable in heart and liver samples, but was expressed in brain and testis, tissues that contain axonemal structures. In seminiferous tubules of the testis, KPL2 expression was stage-specific and appeared to be highest in spermatocytes and round spermatids. During differentiation of RTE cells, the expression of KPL2 closely paralleled that of an axonemal dynein heavy chain. These results suggest that KPL2 plays an important role in the differentiation or function of ciliated cells in the airway.
Ciliated cells line the surface of mammalian airways and provide the force necessary for effective mucociliary clearance. The importance of this defense mechanism is clearly demonstrated by the bronchitis and sinusitis that plague patients with the inherited disease primary ciliary dyskinesia. Impaired mucociliary clearance is also a component of other respiratory diseases as well (e.g., cystic fibrosis), and ciliated cells are damaged or lost from the epithelium after exposure to a variety of agents (for review, see [1]). Ciliated cells lost from the epithelium of the large airways following injury are most likely replaced through the differentiation of secretory or basal cells (2); there is little evidence that ciliated cells replicate. However, there is some evidence that ciliated cells can dedifferentiate and participate in wound-repair processes (3). It has been suggested that following severe injury to Clara cells in the bronchiolar region the remaining ciliated cells are the progenitors for repair of the epithelium (4). The pathway of ciliated cell differentiation and the fate of ciliated cells following injury have not been well studied, in part because until recently, suitable in vitro model systems were not available. One of the difficulties in studying these processes is the shortage of markers that can identify ciliated cells in the absence of visible cilia. The presence of numerous centrioles in the apical region of a cell is recognized as a marker of preciliated cells, but requires the labor-intensive technique of electron microscopy.
Recently, several laboratories have begun to clone and characterize genes expressed in ciliated epithelial cells (i.e., 5–8). These genes may be useful as markers of ciliated cell differentiation, and studies of their regulation will provide additional information on the regulation of ciliated cell differentiation. For example, we have previously identified and studied the expression of seven axonemal dynein heavy chains (DHCs) in cultures of primary rat tracheal epithelial (RTE) cells undergoing ciliogenesis (8). These studies demonstrated that the axonemal DHCs were transcriptionally regulated and expressed coordinately with ciliated cell differentiation. These findings are consistent with those obtained in other species, such as the reciliation of sea-urchin embryos or the growth of flagella by Chlamydomonas reinhardtii (9, 10). However, although there are many similarities between the structure of cilia in mammalian airways and the structure of cilia and flagella of lower species, the differentiation and function of ciliated cells in a complex eukaryotic organism will most likely require additional unique gene products. To identify additional genes that are induced during ciliogenesis of airway epithelium, we have applied the technique of differential display to cultures of RTE cells grown under conditions that increase or inhibit ciliated cell differentiation. In this report we describe the identification and partial characterization of a novel gene whose expression parallels ciliated cell differentiation. This large protein contains sequences that show homology to several functional motifs and may play an important role in the assembly or regulation of axonemal structures.
Freshly isolated RTE cells were cultured on 24-mm Transwell-clear tissue-culture inserts (Costar, Cambridge, MA) coated with rat-tail collagen (Collaborative Research, Bedford, MA) as described in detail elsewhere (11). For the experiments reported here, all cultures were submerged in complete growth media until confluent (Days 7–8). Cultures were then assigned to the various experimental groups (see Differential Display, below) and re-fed daily for the remainder of the experiment.
Differential display was performed using a Delta RNA Fingerprinting Kit (Clontech, Palo Alto, CA) according to the manufacturer's instructions. For differential display, RTE cells were cultured in three different conditions: at an air–liquid interface (ALI) in complete medium (CM), submerged in complete medium (SUB), and at an ALI in medium without epidermal growth factor or cholera toxin (−EGF/−CT). These conditions have been shown to inhibit (SUB) or stimulate (−EGF/−CT) ciliated cell differentiation (12, 13). Total cellular RNA was isolated by the method of Chomczynski and Sacchi (14) and treated with deoxyribonuclease (DNase) as described elsewhere (15). To reduce the number of false positives, two completely independent RTE cultures were analyzed in parallel; only bands that were reproducible between the two experiments were characterized further. Amplifications were carried out using a GeneAmp polymerase chain reaction (PCR) system (Perkin–Elmer, Branchburg, NJ). Differentially expressed bands were reamplified for use as probes on Northern blots and cloned into the pCRII vector for sequencing using the TA Cloning Kit (Invitrogen, San Diego, CA).
To isolate clones covering the entire KPL2 messenger RNA (mRNA), rapid amplification of complementary DNA (cDNA) ends (RACE) procedures were used. KPL2-specific primers were designed from the sequence of the original differential display clone. These were used in conjunction with a rat brain Marathon-Ready cDNA library (Clontech) to obtain additional KPL2 clones in both the 5′ and 3′ directions. Products of nested PCR reactions were cloned into the pCRII vector as above for characterization. Each fragment was sequenced using a Sequenase Kit (Amersham, Arlington Heights, IL) or using an ABI PRISM 377 DNA sequencer and Dye Terminator Cycle Sequencing Kit (Perkin–Elmer). Additional primers were designed to the 5′ end of the new clones, and successive walk-steps were carried out to assemble the complete cDNA sequence. Fragments were also cloned from a cDNA library constructed from mRNA isolated from well-differentiated RTE cell cultures. Each fragment was confirmed to be part of the original KPL2 message by Northern analysis and reverse transcriptase (RT)–PCR using primers that crossed the junctions between fragments. The entire cDNA sequence was obtained in both directions, and in most regions multiple clones were sequenced. Additional gene-specific primers were designed for use in sequencing reactions. The sequence data was assembled using AssemblyLIGN (Oxford Molecular, Beaverton, OR) and analyzed using MacVector (Oxford Molecular) and programs in the Genetic Computer Group sequence analysis package. Analysis of the predicted protein sequence of KPL2 was also performed using the program Profilescan (Swiss Institute for Experimental Cancer Research, Epalinges, Switzerland). The sequence of rat KPL2 has been deposited in Genbank with the accession no. AF102129.
Northern and Southern analysis were carried out using standard procedures essentially as described (8, 15). RNA was isolated from cultured cells and rat tissues by homogenization in guanidinium solution according to Chomczynski and Sacchi (14) except for tracheal epithelium RNA, which was isolated by briefly flushing the tracheal lumen with the guanidinium solution. Twenty micrograms of total RNA was used for Northern analysis. After electrophoresis, nucleic acids were transferred to Nytran membranes (Schleicher & Schuell, Keene, NH) by capillary blotting and cross-linked using a Stratalinker (Stratagene, La Jolla, CA). Probes were isolated from low melting– point agarose, labeled with the Rediprime DNA Labeling System (Amersham), and purified on NucTrap columns (Stratagene). Hybridization was performed in QuickHyb solution (Stratagene), and washouts were according to the manufacturer. Blots were exposed to film at −70°C with intensifying screens for appropriate times.
For analysis of KPL2 expression in tissues, total RNA (isolated as above) was treated with DNase, extracted, and precipitated, and the concentration was determined spectrophotometrically. RNA (0.5 μg) was converted into cDNA using random primers and an RNA PCR kit (Perkin–Elmer), and amplified using specific primers for KPL2. The forward primer was 5′-GAC CTG TGG GAA GAT GAG GAA ACA AA-3′, and the reverse primer was 5′-ATG CAG GCA GAA GTG AAC CGT TCC-3′. Control reactions were performed in the absence of RT, and the experiment was repeated using RNA isolated from two different animals. For these experiments the annealing temperature was 62°C for four cycles, 60°C for four cycles, and 58°C for 27 cycles (total of 35 cycles). An aliquot of each reaction was electrophoresed on a 2% NuSieve/Seakem GTC agarose gel (FMC Bioproducts, Rockland, ME) and visualized by ethidium bromide staining.
Adult rats were killed by CO2 asphyxiation, and testes were isolated and fixed for 3 h with 4% paraformaldehyde in phosphate-buffered saline, pH 7.4 (PBS). After rinsing in PBS, samples were dehydrated in a graded ethanol series, infiltrated with xylene, and embedded in paraffin. Cultures of RTE cells were fixed for 3 h with 4% paraformaldehyde in PBS, embedded in 2% agarose for ease of handling, and processed as above. In situ hybridization was performed as previously described (16) with the following minor modifications. Serial sections (7 μm) were mounted on Silane-prep slides (Sigma, St. Louis, MO), deparaffinized, acetylated, and pretreated with 2× sodium citrate–50% formamide–10 mM dithiothreitol at 50°C. Radiolabeled sense and antisense KPL2 and β-actin probes were generated with either SP6 or T7 RNA polymerase and 35S-labeled cytidine triphosphate (Amersham). Probes were subjected to limited alkaline hydrolysis to reduce the size of transcripts to about 150 bases. After hybridization at 50°C for 16 h in a humidified chamber, the sections were washed and treated with ribonuclease, as described (16). The slides were dipped in Kodak NTB-2 emulsion (Eastman Kodak, Rochester, NY) diluted 1:1.5 and exposed for 14 d at 4°C. Slides were developed and counterstained with hematoxylin.
Differential display was used to identify genes whose expression was increased during ciliogenesis of RTE cells in culture. RNA was isolated from RTE cells cultured at an ALI in CM, cells cultured in the −EGF/−CT condition, and cells cultured in the SUB condition. Fully differentiated, 15-d-old cultures were used for these experiments. These conditions have previously been shown to stimulate (−EGF/−CT) or inhibit (SUB) ciliated cell differentiation compared with control conditions (12). Comparison of the differential display patterns showed that the technique was highly reproducible; the majority of cDNAs were expressed at similar levels in all three conditions, and in two separate experiments (Figure 1A). However, several cDNA bands were clearly upregulated in the −EGF/−CT condition (arrow, Figure 1A). These cDNAs were cloned and used as probes on Northern blots to verify their differential expression pattern. One of these clones, KPL2, detected an approximately 6-kb transcript in RNA isolated from −EGF/ −CT cultures, but produced no signal in RNA isolated from control or SUB cultures (Figure 1B). Because this mRNA was expressed at higher levels in RTE cultures that contain more ciliated cells, it was chosen for further study.
Fig. 1. Identification of genes induced during ciliated cell differentiation. (A) Differential display gel showing the reproducibility of this technique applied to cultures of rat tracheal epithelial cells. Three different conditions were compared from two separate experiments. The arrow indicates a band induced in the −EGF/−CT cultures, a condition which increases ciliated cell differentiation. (B) Northern blot using an isolated clone from the band shown in A as a probe. The clone, referred to as KPL2, detects an approximately 6-kb mRNA that is induced in the −EGF/−CT cultures, confirming the results of the differential display. Ethidium bromide staining of 18S ribosomal RNA shows equal amounts of RNA loaded.
[More] [Minimize]To determine the sequence of the KPL2 mRNA, RACE procedures were used to obtain additional clones in both the 5′ and 3′ direction from the original KPL2 clone. Clones were amplified from both a rat brain cDNA library and an RTE cell cDNA library. Each additional clone was confirmed by Northern blotting and in RT–PCR reactions by using primers that spanned the junctions between fragments. Successive RACE steps were used to assemble the full-length sequence of 5,365 nucleotides. Searching of GenEMBL revealed no significant homology to any previously identified sequence, indicating that KPL2 is a novel gene. Southern analysis was performed on rat genomic DNA digested with BamHI, EcoRI, or HindIII. When the original differential display clone was used as a probe, a single hybridizing fragment was detected in each digest, indicating that KPL2 is a single-copy gene (Figure 2). One shorter 3′ clone was also obtained that contains a unique sequence and may correspond to an alternatively spliced message (data not shown).
Fig. 2. Southern analysis of KPL2. Rat genomic DNA was digested with three different restriction enzymes and, after electrophoresis and transfer, was hybridized with the KPL2 probe. A single major fragment was detected in each digest, indicating that KPL2 is a single-copy gene.
[More] [Minimize]The KPL2 sequence contains a long open reading frame that predicts a protein of 1744 amino acids (Figure 3) with a mol wt of 200,948 daltons and an isoelectric point of 5.42. The protein contains several regions that are strongly homologous to known functional domains. The amino-terminal 105 amino acids are predicted to form a calponin homology domain (CH-domain; [17]), which functions as an actin binding domain in a variety of proteins. Several stretches of basic amino acids fit the consensus sequence of a bipartite nuclear localization signal (NLS; [18]). One of these occurs at residues 238–255; the other two sequences overlap at residues 1217–1234 and 1224–1241. These two overlaps occur in a stretch of amino acids that consists of four tandem repeats of the sequence Q/KAKKEKE. KPL2 also contains an adenosine triphosphate (ATP)/guanosine triphosphate (GTP) binding site (P-loop; 509–516) and a proline-rich region (1369–1379). In addition, residues 221– 369 show a similarity to the cullin family of proteins (19).
Fig. 3. Amino acid sequence of KPL2. RACE procedures were used to isolate clones covering the complete KPL2 mRNA. Sequencing revealed an opening reading frame of 1744 amino acids. Significant features identified by computer analysis include a CH domain (open box, residues 1–105), several NLS (dashed underline, residues 238–255, 1217–1234, and 1224–1241), an ATP/GTP binding site (underlined, residues 509–516), and a proline-rich region (shaded box, residues 1369–1379). In addition, the sequence contains four imperfect copies of the sequence QAKKEKE (bold boxes, residues 1201–1228).
[More] [Minimize]To examine the tissue distribution of KPL2, total RNA was isolated from tracheal epithelium and from rat brain, heart, kidney, liver, lung, spleen, and testis tissues, and analyzed by Northern blotting. KPL2 expression was detected only in the tracheal epithelium sample by this method (data not shown). To assess KPL2 expression at a more sensitive level, RT–PCR was carried out on tissue RNA samples using primers specific for KPL2. By this technique, brain, lung, testis, and trachea were all positive for KPL2 expression (Figure 4). Spleen and kidney produced weaker signals, whereas heart and liver were negative. This experiment was duplicated using RNA prepared from a second animal. Experiments using primers for other genes gave positive signals for all tissues, confirming that the RNA samples were intact (Andrews and colleagues, submitted). Although not quantitative, these results clearly suggest that KPL2 is expressed at higher levels in tissues that contain cilia or flagella (brain, lung, trachea, and testis) and is not expressed or is expressed at lower levels in the other tissues examined.
Fig. 4. Expression of KPL2 in rat tissues. RNA prepared from brain, heart, kidney, liver, lung, spleen, and testis tissue and from tracheal epithelium was analyzed by RT–PCR. After 35 cycles of amplification using KPL2-specific primers, product was clearly detected in the brain, lung, testis, and tracheal samples. Weaker signals were also present in the spleen and kidney samples, whereas heart and liver were negative. These results indicate that the expression of KPL2 is tissue-specific.
[More] [Minimize]To determine the localization of KPL2 mRNA at the cellular level, in situ hybridization was used to localize KPL2 expression in mature rat testis. Paraffin-embedded sections of rat testis (Figure 5) were hybridized with antisense or sense probes to KPL2 or β-actin. The antisense KPL2 probe produced a strong signal in a subset of the seminiferous tubules (Figure 5a). Because different tubules are at different stages of development, this suggests that KPL2 is differentially expressed during spermatogenesis. KPL2 expression appeared to occur predominantly in round spermatocytes or spermatid cells (Figure 5b). Hybridization using the sense probe as a negative control (Figure 5c) produced a low level of uniform background signal. This indicates that the antisense probe was specifically detecting KPL2 mRNA. KPL2 is therefore not only strongly expressed in tissues that contain axonemes, but its expression is also limited to specific stages or times of development.
Fig. 5. In situ analysis of KPL2 expression in testis. Paraffin-embedded sections of adult rat testis were hybridized with sense or antisense KPL2 RNA probes. (a) Darkfield photomicrograph of a section of normal rat testis hybridized with antisense KPL2 probe demonstrating high level of expression in the seminiferous tubules. (b) Brightfield photomicrograph of the same section as in a. (c) Darkfield photomicrograph of a similar section hybridized with a sense KPL2 probe (negative control). No specific signal is present. The high level of KPL2 expression in the seminiferous tubules is consistent with a role of KPL2 in the differentiation of spermatocytes. Bar = 55 × 10−6 m.
[More] [Minimize]To examine further the expression of KPL2 during differentiation of RTE cells, parallel cultures of RTE cells were grown in four different conditions. All cultures were grown to confluence submerged in CM until Day 7. On Day 7, an ALI was created and the cultures were divided into four groups. One group of cultures was re-fed basally with CM containing both retinoic acid (RA) and EGF (+EGF/+RA). Another was re-fed with medium without EGF (−EGF/+RA), a condition that increases ciliated cell differentiation. Although RTE cells cultured continuously in the absence of RA undergo squamous differentiation, preliminary studies demonstrated an unexpected increase in ciliated cell differentiation in RTE cultures acutely deprived of RA after reaching confluence. Therefore, one group of cells was re-fed with medium without RA (+EGF/−RA) and another group was re-fed with medium without either EGF or RA (−EGF/−RA). Cultures were re-fed daily with the appropriate media from Days 7 to 16. RNA was isolated from cultures at various time points and analyzed by Northern blotting for markers of differentiation and KPL2 expression. RTE cells grown in CM for the entire experiment showed a low level of ciliated cell differentiation, with mostly single scattered ciliated cells visible. Previous studies have shown that in CM (+EGF/+RA) ciliated cells begin to appear about Day 10 and cover 10 to 20% of the culture surface by Day 14, after which the percentage of ciliated cells remains relatively constant (12, 13). Northern analysis with a probe for axonemal dynein, a marker of ciliated cell differentiation (8), demonstrated that axonemal dynein was first expressed at detectable levels at Day 14 and increased at Day 16 (Figure 6). KPL2 expression, in agreement with the differential display results, was very low in these cultures. Northern analysis produced only a faint KPL2 signal in Day-16 RNA even after extended exposure to film. In contrast, Northern analysis showed that muc 5 RNA, a marker for mucous cell differentiation (20), was detectable by Day 10 of culture and increased until Day 16. Cornifin, a marker of squamous differentiation (21), was not highly expressed in these cultures. These results are consistent with previous studies from our laboratory demonstrating that mucous differentiation preceded ciliated cell differentiation under these conditions (22). In RTE cells cultured in the absence of exogenous EGF, ciliated cell differentiation was increased. The percentage of ciliated cells in −EGF/ +RA cultures was found to be 2- to 4-fold higher than in +EGF/+RA cultures in our previous studies (12). Axonemal dynein was also expressed earlier (detectable at Day 10) and at higher levels. KPL2 expression paralleled that of axonemal dynein, increasing from Days 10 to 12 and decreasing thereafter. The decrease in axonemal dynein and KPL2 expression may occur as a result of apoptosis, which increases in the absence of EGF (23). Muc 5 RNA was expressed in the −EGF cultures, similarly to the cultures grown in CM. Cornifin was again not detected under these conditions. RTE cell cultures re-fed with medium lacking both EGF and RA showed more ciliated cells than the other groups when examined by light microscopy. Axonemal dynein was expressed early in these cultures, as was KPL2. Muc 5 RNA was again expressed at a relatively constant level. Cornifin expression increased with increased time in the absence of RA. RTE cells grown in the presence of EGF but in the absence of exogenous RA also showed an increase in ciliated cells compared with the cultures grown in CM; however, by the end of the culture period there were also patches of squamous cells visible. RNA prepared from these cultures showed expression of axonemal dynein and KPL2 at late time points, with a peak at Day 14. In contrast, muc 5 RNA was highest at Day 10 and appeared to decrease by Day 16. Cornifin RNA was highly expressed in these cultures, especially at the later time points. Importantly, the expression pattern of KPL2 under all these conditions most closely resembles that of axonemal dynein. This data suggests that KPL2 expression is closely linked to the ciliated cell phenotype.
Fig. 6. Northern analysis of KPL2 and markers of mucociliary and squamous differentiation in cultures of RTE cells grown under four different conditions. Parallel cultures of RTE cells were grown under four different conditions as described in text. RNA was isolated on the indicated days and analyzed by Northern blotting for the expression of KPL2; axonemal dynein b (axo b), a marker of ciliated cell differentiation; rat mucin 5 (muc 5), a marker of mucous differentiation; and cornifin, a marker of squamous differentiation. Under all conditions, the expression of KPL2 most closely resembled that of axonemal dynein.
[More] [Minimize]To determine which cell type expressed KPL2 in cultures of RTE cells grown in the absence of RA from Days 7 to 16, in situ hybridization was performed. As mentioned previously, these cultures consisted of a mixed epithelium with areas of both mucociliary and squamous differentiation present. Small groups of ciliated cells were frequently located above a layer of squamous cells (Figure 7a), suggesting that as the culture underwent squamous differentiation, cells committed to ciliated cell differentiation were forced upward by the newly developing squamous cells. In situ hybridization with an antisense KPL2 probe produced a weak signal in these cultures, in agreement with the results obtained by Northern blot analysis. Cilia in these sections were difficult to visualize after the in situ procedure, which included treatment with proteinase K. However, the KPL2 signal was predominantly localized over the ciliated cell clusters and not the squamous or basal cell layers (Figure 7b). Hybridization with a sense KPL2 probe (negative control) produced a low level of nonspecific background (Figure 7c), whereas hybridization with an antisense β-actin probe (positive control) produced a strong signal (Figure 7d), demonstrating that these sections contained intact RNA. These results show that in RTE cell cultures grown under conditions in which both squamous and ciliary differentiation is present, KPL2 expression is restricted to the ciliated epithelium.
Fig. 7. In situ analysis of KPL2 expression in RTE cultures of mixed phenotype. Cultures of RTE cells grown in RA-containing media for 7 d were re-fed with media without RA from Days 8 to 15 and processed for in situ hybridization. (a) Under these conditions, ciliated cells appeared in clusters above areas of squamous metaplasia. (b) Similar section as in a hybridized with an antisense KPL2 probe demonstrating KPL2 expression in the cluster of ciliated cells. Individual cilia were difficult to visualize after the in situ procedure. (c) Hybridization with a sense KPL2 probe (negative control) produced no specific signal. (d) Hybridization with an antisense β-actin probe (positive control) showed intense labeling throughout the epithelium, demonstrating the intactness of the RNA in these samples. In cultures containing ciliated, basal, and squamous cells, KPL2 expression is restricted to the ciliated epithelium. Bar = 16 × 10−6 m.
[More] [Minimize]Ciliated cells are well recognized for their important role in mucociliary clearance. Ciliated cells have also been suggested to perform several other functions critical to normal airway physiology. For example, ciliated cells produce tracheal antimicrobial peptide, a defensin molecule that may increase resistance to infection (24). Ciliated cells express the cystic fibrosis transmembrane conductance regulator protein, which suggests that they contribute to the regulation of the composition of the airway surface fluid (25). Ciliated cells may also contribute to normal airway function in other, as yet unknown ways. We have begun to study this important cell type, with an emphasis on ciliated cell differentiation and gene expression. Although studies on the repair of airway epithelium following injury have suggested that ciliated cells regenerate through the further differentiation of secretory cells (2), other studies have provided evidence that ciliated cells dedifferentiate and participate in the repair process (3). Recently, several laboratories have reported the cloning of genes preferentially expressed in ciliated cells (5-8). Some of these were isolated on the basis of their homology to proteins of lower organisms known to be structural components of cilia or flagella (e.g., heavy- and light-chain dyneins). However, only a fraction of the estimated 200 proteins that constitute a cilium have been identified. In addition, it is likely that ciliated cells of the mammalian airway express proteins not found in simpler organisms. In this study, differential display was chosen to identify additional ciliated cell– specific genes. RTE cells were cultured under previously characterized conditions to increase or inhibit ciliated cell differentiation (12, 13). When applied to these cultures, the differential display technique proved to be very reproducible and allowed the identification of several differentially expressed genes, including KPL2.
KPL2 expression was closely correlated with ciliated cell differentiation of RTE cells in vitro. When confluent cultures of RTE cells were deprived of EGF and CT, a condition that increases the extent of ciliated cell differentiation, KPL2 was clearly induced compared with the cultures maintained in CM or SUB cultures (Figure 1B). In RTE cell cultures acutely deprived of RA after 7 d of culture, a surprising increase in the number of ciliated cells was observed. In these cultures simultaneously undergoing both mucociliary and squamous differentiation, KPL2 expression again paralleled both ciliated cell differentiation and axonemal dynein expression. In situ analysis performed at a time when both ciliated and squamous cells were present demonstrated that KPL2 expression was restricted to the ciliated epithelium. This demonstrates that KPL2 expression is not simply induced by the removal of EGF, but instead is related to the cell phenotype. Thus, under all conditions examined, KPL2 expression correlates with the expression of axonemal dynein and the appearance of ciliated cells in this model. Additional experiments will be required to determine if KPL2 is expressed during ciliogenesis, in fully differentiated ciliated cells, or both.
In vivo, KPL2 was expressed in a tissue-specific pattern. RT–PCR experiments demonstrated KPL2 expression in lung, testis, and brain, and at lower levels in kidney and spleen. No signal was detected in heart and liver. Lung and brain tissue contain ciliated cells, sperm cells in the testis contain flagella, and kidney is known to contain many primary cilia (26). Thus, the tissue distribution of KPL2 is consistent with a role in axoneme-containing cells, although it may also function in other cell types. Other components of cilia have recently been shown to be present in nonciliated tissues. This pattern of expression is also similar to that of the axonemal DHCs, which are known components of cilia and flagella (8). In situ analysis of KPL2 expression in the testis demonstrated that KPL2 was expressed at specific stages in the developing sperm cells, again consistent with a role for KPL2 in the differentiation of an axoneme-containing cell.
Sequence analysis indicates that KPL2 is a novel protein that contains several domains that have been reported to play a role in protein:protein interactions. The amino terminus of KPL2 is predicted to form a CH domain, which binds to filamentous actin (17, 27). Two CH domains are present in spectrin, which cross-links actin filaments, whereas single CH domains are present in signaling proteins such as Vav. The presence of a CH domain suggests that KPL2 interacts with the cytoskeleton. KPL2 possesses three consensus bipartite NLSs, which have been shown to be sufficient to target many proteins to the nucleus (18). The effect of multiple NLSs is cumulative, increasing the likelihood that KPL2 performs at least part of its function in the nucleus. However, until it is demonstrated that the KPL2 protein is actually localized in the nucleus, other possibilities also need to be considered. Two of the NLSs occur as part of a highly charged stretch of amino acids consisting of four tandem repeats of the sequence Q/KAKKEKE. Interestingly, this sequence shares a weak similarity with a highly repetitive 16–amino acid sequence (KKKCAEAAKKEKEAAE) found in the Drosophila hydei sperm tail–specific proteins of the Dhmst101 family (28, 29). However, the function of this unique sequence in KPL2 is unknown at present. In addition, KPL2 contains a proline-rich region, which may function as a site of interaction with SH3 domains (30, 31). Finally, KPL2 contains an ATP/GTP binding site (32) at residues 509– 516. The presence of sequences homologous to these functional motifs allows some speculation as to the function of KPL2. KPL2 may bind to actin filaments through its N-terminal CH domain and interact with SH3-containing proteins through its proline-rich region. Alternatively, KPL2 may be involved in transmitting signals from the cytoskeleton to the nucleus. The P-loop suggests that KPL2 possesses catalytic activity or its activity is regulated by the presence of nucleotide triphosphates.
In these studies, differential display was used to identify a novel gene induced during ciliated cell differentiation of airway epithelial cells. KPL2 is expressed in several tissues that contain axonemes, and its expression pattern in RTE cultures most closely resembles that of an axonemal DHC, a marker of ciliogenesis. The KPL2 sequence predicts a large protein with the potential for many intersting protein–protein interactions. Studies of the KPL2 protein, both in vivo and in vitro, will need to be carried out to determine the function of this unique protein in ciliated cells.
| 1. | Wanner A., Salathe M., O'Riordan T. G.Mucociliary clearance in the airways. Am. J. Respir. Crit. Care Med.154199618681902 |
| 2. | Keenan K. P., Combs J. W., McDowell E. M.Regeneration of hamster tracheal epithelium after mechanical injury. Virchows Arch.411982193214 |
| 3. | Erjefalt J. S., Erjefalt I., Sundler F., Persson C. G. A.In vivo restitution of airway epithelium. Cell Tissue Res.2811995305316 |
| 4. | van Winkle L. S., Buckpitt A. R., Nishio S. J., Isaac J. M., Plopper C. G.Cellular response in naphthalene-induced Clara cell injury and bronchiolar epithelial repair in mice. Am. J. Physiol.2691995L800L818 |
| 5. | Kastury K., Taylor W. E., Shen R., Arver S., Gutierrez M., Fischer C. E., Coucke P. J., Van Hauwe P., Van Camp G., Bhasin S.Complementary deoxyribonucleic acid cloning and characterization of a putative human axonemal dynein light chain gene. J. Clin. Endocrinol. Metab.82199730473053 |
| 6. | Chapelin C., Duriez B., Magnino F., Goossens M., Escudier E., Amselem S.Isolation of several human axonemal dynein heavy chain genes: genomic structure of the catalytic site, phylogenetic analysis and chromosomal assignment. FEBS Lett.4121997325330 |
| 7. | Vaughan K. T., Mikami A., Paschal B. M., Holzbaur E. L. F., Hughes S. M., Echeverri C. J., Moore K. J., Gilbert D. J., Copeland N. G., Jenkins N. A., Vallee R.Multiple mouse chromosomal loci for dynein-based motility. Genomics3619962938 |
| 8. | Andrews K. L., Nettesheim P., Asai D. J., Ostrowski L. E.Identification of seven rat axonemal dynein heavy chain genes: expression during ciliated cell differentiation. Mol. Biol. Cell719967179 |
| 9. | Keller, L. R. 1995. Enrichment of mRNA encoding flagellar proteins. In Cilia and Flagella: Methods in Cell Biology, Vol. 47. W. Dentler and G. Witman, editors. Academic Press, Inc., San Diego. 551–557. |
| 10. | Gibbons B. H., Asai D. J., Tang W.-J. Y., Hays T. S., Gibbons I. R.Phylogeny and expression of axonemal and cytoplasmic dynein genes in sea urchins. Mol. Biol. Cell519945770 |
| 11. | Ostrowski, L. E., S. H. Randell, A. B. Clark, T. E. Gray, and P. Nettesheim. 1995. Ciliogenesis of rat tracheal epithelial cells in vitro. In Cilia and Eukaryotic Flagella: Methods in Cell Biology. W. L. Dentler and G. B. Witman, editors. Academic Press, Orlando, FL. 147:53–63. |
| 12. | Clark A. B., Randell S. H., Nettesheim P., Gray T. E., Bagnell B., Ostrowski L. E.Regulation of ciliated cell differentiation in cultures of rat tracheal epithelial cells. Am. J. Respir. Cell Mol. Biol.121995329338 |
| 13. | Ostrowski L. E., Nettesheim P.Inhibition of ciliated cell differentiation by fluid submersion. Exp. Lung Res.211995957970 |
| 14. | Chomczynski P., Sacchi N.Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem.1621987156159 |
| 15. | Potdar P. D., Andrews K. L., Nettesheim P., Ostrowski L. E.Expression and regulation of γ-glutamyl transpeptidase-related enzyme in tracheal cells. Am. J. Physiol.2731997L1082L1089 |
| 16. | Matsuura H., Myokai F., Arata J., Noji S., Taniguchi S.Expression of type II transforming growth factor-beta receptor mRNA in human skin, as revealed by in situ hybridization. J. Dermatol. Sci.819942532 |
| 17. | Castresana J., Saraste M.Does Vav bind to F-actin through a CH domain? FEBS Lett.3741995149151 |
| 18. | Dingwall C., Laskey R. A.Nuclear targeting sequences—a consensus? Trends Biochem. Sci.161991478481 |
| 19. | Kipreos E. T., Lander L. E., Wing J. P., He W. W., Hedgecock E. M.cul-1 is required for cell cycle exit in C. elegans and identifies a novel gene family. Cell851996829839 |
| 20. | Guzman K., Randell S. H., Nettesheim P.Epidermal growth factor regulates expression of the mucous phenotype of rat tracheal epithelial cells. Biochem. Biophys. Res. Commum.2171995412418 |
| 21. | Fujimoto W., Marvin K. W., George M. D., Celli G., Darwiche N., DeLuca L. M., Jetten A. M.Expression of cornifin in squamous differentiating epithelial tissues, including psoriatic and retinoic acid-treated skin. J. Invest. Dermatol.1011993268274 |
| 22. | Kaartinen, L., P. Nettesheim, K. B. Adler, and S. H. Randell. 1993. Rat tracheal epithelial cell differentiation in vitro. In Vitro Cell. Dev. Biol. 29A: 481–492. |
| 23. | Antoshina E., Ostrowski L. E.TGFβ1 induces growth arrest and apoptosis but not ciliated cell differentiation in rat tracheal epithelial cell cultures. In Vitro Cell. Dev. Biol.331997212217 |
| 24. | Diamond G., Jones D. E., Bevins C. L.Airway epithelial cells are the site of expression of a mammalian antimicrobial peptide gene. Proc. Natl. Acad. Sci. USA90199345964600 |
| 25. | Puchelle E., Gaillard D., Ploton D., Hinnrasky J., Fuchey C., Bontterin M.-C., Jacquot J., Dreyer D., Pavirani A., Dalemans W.Differential localization of the cystic fibrosis transmembrane conductance regulator in normal and cystic fibrosis airway epithelium. Am. J. Respir. Cell Mol. Biol.71992485491 |
| 26. | Wheatley D. N.Primary cilia in normal and pathological tissues. Pathobiology631995222238 |
| 27. | Carugo K. D., Banuelos S., Saraste M.Crystal structure of a calponin homology domain. Nat. Struc. Biol.41997175179 |
| 28. | Neesen J., Bunemann H., Heinlein U. A. O.The Drosophila hydei gene Dhmst101(1) encodes a testis-specific, repetitive, axoneme-associated protein with differentialP abundance in Y chromosomal deletion mutant flies. Dev. Biol.1621994414425 |
| 29. | Neesen J., Padmanabhan S., Bunemann H.Tandemly arranged repeats of a novel highly charged 16-amino-acid motif representing the major component of the sperm-tail-specific axoneme-associated protein family Dhmst101 form extended a-helical rods within the extremely elongated spermatozoa of Drosophila hydei. Eur. J. Biochem.225199410891095 |
| 30. | Ren R., Mayer B. J., Cicchetti P., Baltimore D.Identification of a ten-amino acid proline-rich SH3 binding site. Science259199311571161 |
| 31. | Yu H., Chen J. K., Feng S., Dalgarno D. C., Brauer A. W., Schreiber S. L.Structural basis for the binding of proline-rich peptides to SH3 domains. Cell761994933945 |
| 32. | Saraste M., Sibbald P. R., Wittinghofer A.The P-loop—a common motif in ATP- and GTP-binding proteins. Trends Biochem. Sci.151990430434 |
* Present addresses: Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC;
† Department of Internal Medicine, Washington University, St. Louis, MO;
‡ Cancer Research Institute, Tata Memorial Centre, Parel, Bombay, India 400012.
Abbreviations: air–liquid interface, ALI; adenosine triphosphate, ATP; complementary DNA, cDNA; calponin homology, CH; cholera toxin, CT; dynein heavy chain, DHC; epidermal growth factor, EGF; guanosine triphosphate, GTP; messenger RNA, mRNA; nuclear localization signal, NLS; phosphate-buffered saline, PBS; polymerase chain reaction, PCR; retinoic acid, RA; rapid amplification of cDNA ends, RACE; reverse transcriptase, RT; rat tracheal epithelial, RTE; submerged in CM, SUB.
