JBC Advanced Peptides, Inc.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


Originally published In Press as doi:10.1074/jbc.M000676200 on April 20, 2000

J. Biol. Chem., Vol. 275, Issue 25, 19324-19333, June 23, 2000
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
275/25/19324    most recent
M000676200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Johnson, R. C.
Right arrow Articles by Mains, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Johnson, R. C.
Right arrow Articles by Mains, R. E.
Social Bookmarking
 Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Isoforms of Kalirin, a Neuronal Dbl Family Member, Generated through Use of Different 5'- and 3'-Ends Along with an Internal Translational Initiation Site*

Richard C. Johnson, Peter Penzes, Betty A. Eipper, and Richard E. MainsDagger

From the Department of Neuroscience, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205

Received for publication, January 29, 2000, and in revised form, March 29, 2000

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Kalirin is a neuron-specific GDP/GTP exchange factor for Rho subfamily GTP-binding proteins. The major Kalirin transcripts in adult rat brain were identified. Most include a Sec14p-like putative lipid-binding motif followed by nine spectrin-like repeats and a Dbl homology/pleckstrin homology (DH-PH) domain. Kalirin proteins with four different NH2 termini are generated through the use of five different 5'-ends; three of the proteins differ only at the extreme NH2 terminus, and one is truncated because translation is initiated at a methionine in the 5th spectrin repeat. Four different 3'-ends yield Kalirin proteins with additional functional domains. Kalirin-7 (7-kilobase pair mRNA) terminates with a PDZ-binding motif, which in Kalirin-8 is replaced by an SH3 domain. Kalirin-9 contains another pair of DH-PH and SH3 domains. Kalirin-12 additionally encodes a putative Ser/Thr protein kinase. Antisera specific for different COOH termini established Kalirin-7 as the most abundant in cortex, with significant amounts of Kalirin-9 and Kalirin-12; Kalirin-7 was less prevalent in cerebellum and olfactory bulb. Kalirin proteins lacking the Sec14p-like domain and first four spectrin-like repeats were much less prevalent. Form-specific antisera demonstrated that different forms of Kalirin were localized to distinct subcellular regions of cultured neurons. Members of the family of Kalirin proteins may subserve different functions at these different locations.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

The Dbl family of GDP/GTP exchange factors (GEFs)1 for small GTP-binding proteins of the Rho subfamily is large and growing (1-4) (Fig. 1). Adjacent DH and PH domains are common to all family members, with various additional domains present in the NH2- and COOH-terminal ends. Dbl and many of the earliest members of this protein family were identified through their oncogenic activity; several were found to be especially prevalent in the nervous system (2). More recently, neuron-specific members of the Dbl family such as Kalirin (see below) and collybistin (5) have been identified through protein-protein interaction screens.


View larger version (29K):
[in this window]
[in a new window]
 
Fig. 1.   Relationship of known Kalirin isoforms to homologous proteins from C. elegans and man. Kalirin-7 and Kalirin-8 transcripts encoding three different NH2 termini have been identified; * denotes the unique 4-, 22-, or 24-amino acid sequences preceding the divergence point. Dbl is the founding member of this family of GEFs (1, 2, 44). Trio was identified by its ability to bind to LAR, a receptor type protein tyrosine phosphatase (13); in their overlapping regions, the amino acid sequences of rat Kalirin and human Trio are 73% identical. C. elegans has a single Kalirin/Trio-like gene, unc-73; UNC-73A is the longer of the two splice variants observed (14). Duet, identified in human skeletal muscle, encodes a protein 95% identical to rat Kalirin-8 in their region of overlap (19); the remainder of the Duet transcript encodes a protein 47% identical to the corresponding region of Trio. Sec14p-, spectrin-, SH3, PDZ-binding, Ig-, FnIII- and protein kinase-like domains are indicated, along with demonstrated DH-PH domains.

Kalirin was identified as a protein interacting with the COOH-terminal cytoplasmic tail of the peptide-processing enzyme peptidylglycine alpha -amidating monooxygenase (PAM) (6). Kalirin (Duo) was also identified as an interactor with Huntingtin-associated protein-1 (7) and the inducible form of nitric-oxide synthase (8). Kalirin is found in a large fraction of central nervous system neurons, notably in long projection neurons (9). When expressed in pituitary tumor cells, Kalirin increases the length and branching patterns of processes and facilitates regulated secretion (9, 10). By forming heterodimers with iNOS and inactivating it, Kalirin restores stimulated secretion to cells in which secretion has been inhibited by induction of iNOS (8). The role of Kalirin in the normal functioning of the nervous system has not yet been explored.

The first forms of rat Kalirin identified (called P-CIP10a and -10b) differed only at their 5'-ends, sharing a phospholipid binding motif (11), nine spectrin-like repeats, a single DH-PH domain, and an SH3 domain (9). The sequences of human Kalirin (Duo), P-CIP10a and -10b, diverged at the same point near the NH2 terminus (7). Human Kalirin (Duo) lacked an SH3 domain, terminating instead with a putative PDZ recognition motif (7, 12). A homologous rat transcript was recently identified; based on the size of this transcript, it is referred to as Kalirin-7 (12). Similarly, P-CIP10a and -10b will be referred to as Kalirin-8a and -8b, respectively, since both correspond approximately to 8-kb mRNAs.

Current data suggested that additional forms of Kalirin remain to be identified. On Northern blots, Kalirin-7 is a major transcript, with other Kalirin mRNAs ranging in size from 5 to 15 kb. When used to analyze extracts of rat cortex, antibodies to the spectrin-like region of Kalirin recognized proteins ranging in size from 115 to 420 kDa (12). Kalirin-7, a major protein of 190 kDa, was uniquely localized to the postsynaptic density fraction. Trio and UNC-73A, the closest homologues to Kalirin, are extended COOH-terminally to contain domains not present in Kalirin-7 or -8 (13, 14) (Fig. 1).

Here we report the cloning of three new forms of Kalirin, designated Kalirin-Delta 7, Kalirin-9, and Kalirin-12. The basic approach was to employ 5'- and 3'-RACE to extend known sequence and identify distinct transcripts. Form-specific cDNA probes were used for Northern analyses, and RT-PCR was used to confirm and determine 5'-splice variants. Finally, form-specific antisera were employed to identify the corresponding proteins in brain extracts and to localize the endogenous proteins in cultured neurons.

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Northern Analyses-- Total RNA from rat parietal cortex and liver was prepared using RNA Stat-60 (Tel-Test Inc). Poly(A)+ RNA was prepared using the Promega poly(A) Ttract mRNA Isolation System. Equal amounts (1 µg) of poly(A)+ RNA from liver and cortex were loaded side by side in quadruplicate on a 0.9% agarose gel containing formaldehyde (15). After electrophoresis and transfer to a Nytran+ membrane (Schleicher & Schuell), the membrane was cut into 4 equivalent strips, each containing liver and cortex RNA samples. The strips were hybridized with random primed cDNA probes using standard techniques (15), except that Ultrahyb hybridization solution (Ambion) was used to increase the signal for the Kalirin-8 COOH terminus probe. Each blot was stripped by heating briefly to 100 °C in 1% SDS, 0.01× salt sodium citrate before hybridizing to different probes. The probes correspond to the following regions of Kalirin: spectrin domains 1-3 (nt 580-1531), spectrin domains 3-8 (nt 1364-3398), DH1-PH1 (nt 3790-4635), Kalirin-8 COOH terminus (nt 5518-5728) in Kalirin-8a; Kalirin-7 3'-end (nt 4865-5087) in Kalirin-7a; DH2-PH2 (nt 5667-6689) and Kal-9 3'UTR (nt 7151-7680) in Kalirin-9a; and Kalirin kinase (nt 7223-8950) in Kalirin-12a. Kalirin is not expressed in liver (9), which served as a negative control. Nucleotide numbering is based on the 5'-end of P-CIP10a. The Kalirin-7 GenBankTM accession number is AF230644.

RT-PCR for Alternate 5'-Ends-- Oligo(dT)-primed cDNA was made from 1 µg of rat parietal cortex total RNA using Superscript II Reverse Transcriptase (Life Technologies, Inc.) in a 20-µl reaction. Negative control samples for PCR were processed as above without Superscript II. The Expand High Fidelity PCR system (Roche Molecular Biochemicals) was used for all PCRs, following the manufacturer's protocol. cDNA template (2 µl) was amplified for 30 cycles in all reactions. Approximately 1/5 of each reaction was electrophoresed on a 0.8% agarose gel and transferred to Nytran+ membrane (Schleicher & Schuell) for Southern analysis. Table I shows the primers used for PCR.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Kalirin PCR primers
Start and stop codons are underlined where appropriate.

Cloning of Delta Kalirin-7 (Kalirin-5)-- Screening of a lambda -ZAPII rat hippocampal random primed cDNA library has been described previously (9). Two clones with a unique 5'-end upstream of nt 1735 in Kalirin-7a (lambda 10/29, lambda 10/40) were rescued into pBluescript II SK- (Stratagene). pBS.10/29 had a 2.8-kb insert with 64 bp of unique 5'-sequence, and pBS.10/40 had a 2.58-kb insert that extended 30 bp further 5' than lambda 10/29. To confirm this junction, RT-PCR was performed as described above using sense primer 5'-GGGAATTCCGACTCTCTTGACTCTC-3' containing an EcoRI site (underlined) and antisense primer 5'-GCTGCTGGAGGACAGACT-3' (nt 2209-2192). cDNA products of 569 and 858 bp (containing an additional 289 bp inserted in front of nt 1735) were amplified by this reaction. Both products were sequenced using the Sequenase PCR product sequencing kit (United States Biochemical Corp.). The Delta Kalirin-7 (Kalirin-5) GenBankTM accession number is AF229255.

In Vitro Transcription/Translation-- pBS.Kal-5TnT was constructed by digesting pBS.10/29 with BamHI, which cuts at nt 2689 in Kalirin and in the 3' polylinker. After gel purification, the 4.0-kb fragment containing the vector and the first 1.0 kb of lambda 10/29 was re-ligated to create a construct with an open reading frame of 798 bp, from nt 1894 to immediately after the BamHI site at the 3' polylinker (see Fig. 3B). Since there was an EcoRI site 7 bp from the 5'-end of the 289-bp insert in the 858-bp PCR product described above, this product was digested with EcoRI and BglII (cuts at nt 1812 in Kalirin); the 359-bp fragment obtained was cloned into EcoRI (5' polylinker)-BglII-cut pBS.Kal-5TnT to create pBS.Kal-5.3TnT. The Promega TNT-coupled Reticulocyte system was used for in vitro transcription/translation reactions. Briefly, radiolabeled proteins were synthesized from 0.5 µg of plasmid DNA using 25 µCi of [35S]methionine (Amersham Pharmacia Biotech) in a 25-µl reaction; aliquots (2 µl) were analyzed by 12% SDS-PAGE. The gel was treated with Amplify (Amersham Pharmacia Biotech), dried, and exposed to film for 18 h.

Cloning of Kalirin-9 and Kalirin-12-- A 3'-RACE protocol was used to isolate the 3'-end of rat Kalirin-9 (16). cDNA was synthesized from 1 µg of rat parietal cortex total RNA using Superscript II (Life Technologies) and a RACE hybrid primer 5'-GGAATTCGAGCTCATCGAT17-3' in a 20-µl reaction. cDNA (2 µl) was initially amplified for 30 cycles using a RACE Adapter antisense primer, 5'-GGAATTCGAGCTCATCGA-3', and a sense primer, 5'-CTGCTTCTTCCCCCTGGTGA-3', located at nt 5097-5115 in Kalirin-8 thus preventing any possibility of amplifying Kalirin-7 sequences. This reaction (1 µl) was amplified for 30 cycles using the RACE Adapter primer and a nested sense primer containing an XbaI site (underlined), 5'-GGTCTAGAATGGAGGCAAGTCTGAGT-3' (nt 5143-5164). An ~2.5-kb fragment was subcloned into pBS II SK-, and sequencing revealed that this clone diverged from Kalirin-8 at nt 5508, adding another 1.55 kb of open reading frame plus about 0.5 kb of 3'-UTR (see Fig. 4A, below). Another clone diverged at nt 5601 and was otherwise the same as the initial clone, indicating that a 93-bp fragment is sometimes spliced out. The Kalirin-9 GenBankTM accession number is AF232668.

The rat Kalirin-12 3' sequence was identified by amplifying rat parietal cortex cDNA with a sense primer, 10/28-RACE 3', located at nt 4592-4611 containing an XbaI site, 5'-CCTCTAGAGCACCCCATCCTCAGACAAT-3', and an antisense primer based on the human Duet sequence (GenBankTM accession number NM_007064 (RNA) or 5902140 (protein)) containing a SalI site, 5'-GGGGTCGACACTGCACGTCTATGTGAA-3' (nt 4028-4008). An ~4.4-kb fragment was subcloned into pBS II SK-, and sequencing revealed that this fragment diverged 15 bp from the end of the coding region of Kalirin-9 and had an open reading frame that extended 1763 bp in the 3' direction (see Fig. 5A, below). PCRs were performed using the Expand High Fidelity PCR system (Roche Molecular Biochemicals). Sequencing was done using the Sequenase 7-deaza-dGTP sequencing kit (United States Biochemical Corp.). Multiple clones from independent sources of cDNA were sequenced to identify and eliminate any errors generated during cDNA amplifications. The Kalirin-12 GenBankTM accession number is AF232669.

Antisera-- Rabbit polyclonal antisera to Kalirin-spectrin and Kalirin-7 were described previously (12). Rabbit polyclonal antisera were generated at Covance (Denver, PA) by immunizing rabbits with the following antigens. Kalirin-SH3 (JH2960) was generated against the glutathione S-transferase fusion of the SH3 domain of Kalirin (amino acids 1617-1684 of Kalirin-8a); Kalirin-9 antiserum (JH3090) was generated against the synthetic peptide comprising the COOH-terminal 19 residues of rat Kalirin-9 (K2326ATAPAESSDESIKTLLKP2344; residues unique to Kalirin-9 are bold); Kalirin-kinase antiserum (JH3226) was generated against a synthetic peptide comprising the COOH-terminal 17 residues of rat Kalirin-12 (P2943IPNVKSYIVNRVNQGT2959). Both peptides were synthesized by Dr. Henry Keutmann (Endocrine Unit, Massachusetts General Hospital) and were cross-linked to keyhole limpet hemocyanin with 0.1% glutaraldehyde (2 mg of peptide/5 mg) prior to injection. Antibodies JH2958, JH2960, JH3090, and JH3226 were affinity-purified using their respective antigenic peptides linked to Affi-Gel 10. Affinity-purified antibodies were used at a final dilution of 1:100. No immunostaining was detected when the antibodies were pre-bound to their corresponding antigenic peptides.

Tissue and Cell Culture Preparations-- Parietal cortex and hippocampi were dissected from adult female Harlan Sprague-Dawley rats (Harlan), and tissues were homogenized in 10 volumes of RIPA buffer (50 mM Tris HCl (pH 8.0), 120 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate, containing phenylmethylsulfonyl fluoride (0.3 mg/ml), and a protease inhibitor mixture (lima bean trypsin inhibitor (50 µg/ml), leupeptin (2 µg/ml), benzamidine (16 µg/ml), and pepstatin (2 µg/ml)) with 10 strokes of a glass/Teflon homogenizer. Insoluble materials were removed by centrifugation at 15,000 × g for 10 min. Primary cultures of cortical neurons from E17-E18 rat pups were prepared by established methods and processed for immunocytochemistry (12).

For immunoprecipitation, homogenate supernatants (2.5 mg of protein) diluted 1:2 in wash buffer (20 mM Tris-HCl, pH8.0, 100 mM NaCl, 1% Triton X-100, and 0.3 mg/ml phenylmethylsulfonyl fluoride) were incubated with Kalirin-spectrin antibody (10 µl). After a 1 h incubation at 4 °C, 20 µl of protein-A Sepharose (Sigma) was added, and samples were tumbled for 1 h at 4 °C. The resin was collected by centrifugation at 5000 × g for 1 min and washed 4 times with wash buffer. The bound proteins were eluted by boiling the resin in 200 µl of 1× SDS-PAGE loading buffer and analyzed by SDS-PAGE and Western blotting.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Northern Blot Analysis Reveals Kalirin Transcripts with Different 5'- and 3'-Ends-- Northern blot analysis of adult rat parietal cortex RNA using probes specific to different regions of the two cloned Kalirin transcripts (Kalirin-7 and Kalirin-8) revealed a complex hybridization pattern with multiple RNA species ranging in mass from about 5 kb to greater than 15 kb (Fig. 2). Probes encompassing Kalirin spectrin-like domains 3-8 or the DH-PH domains (lanes 2 and 3, respectively) yielded very similar patterns; major transcripts of 7.3 and 5.3 kb were detected, whereas at least three larger transcripts (9, 10, and 12 kb) were present at lower levels. A probe corresponding to the COOH terminus of Kalirin-8 (lane 4) hybridized to minor transcripts of 6.8, 9, 10, 12, and greater than 15 kb. A probe specific for the 3'-end of Kalirin-7 (lane 5) hybridized only to the major transcripts (7.3 and 5.3 kb). A probe encompassing spectrin-like domains 1-3 (lane 1) hybridized to the major 7.3-kb transcript as well as to minor transcripts of 9, 10, 12, and greater than 15 kb. Even though the blot shown in lane 1 has been overexposed, it is clear that a 5.3-kb transcript was not detected by this probe. A probe specific for the extreme 5'-end of Kalirin also failed to detect a 5.3-kb transcript (data not shown). The presence of a 5.3-kb mRNA (Kalirin-5) in lanes 2, 3, and 5, but not in lane 1, suggests that this transcript is a 5'-truncated version of Kalirin-7. As expected (9), none of the probes detected cross-reactive transcripts in liver mRNA (not shown).


View larger version (71K):
[in this window]
[in a new window]
 
Fig. 2.   Northern blot analysis with probes specific for different regions of Kalirin-7 and Kalirin-8. Replicate samples of poly(A)+ RNA (1 µg) from adult rat parietal cortex were fractionated on a denaturing agarose gel and transferred to a nylon membrane. Strips of membrane were hybridized with cDNA probes encompassing the indicated regions of Kalirin (open bars): numbering from Kalirin-8a, spectrin-like domains 1-3 (nt 580-1531), spectrin-like domains 3-8 (nt 1364-3398), DH1-PH1 (nt 3790-4635), and Kalirin-8 COOH terminus (nt 5518-5728; K-8 C-ter); numbering from Kalirin-7a, Kalirin-7 3'-end (nt 4865-5087; K-7 3'-end).

Although Kalirin-8 was the first form of Kalirin identified (9), it is not an abundant transcript in the adult. The COOH-terminal end of Kalirin-7 lacks the final 283 amino acids of Kalirin-8, terminating instead with a unique 20 amino acid sequence ending in a putative PDZ recognition motif, -STYV (12). Based on hybridization to a probe specific for the 3'-end of Kalirin-7 (Fig. 2, lane 5), the two most abundant transcripts (7.3 and 5.3 kb) are the only Kalirin transcripts that have this COOH terminus. Western blot analysis of cortical extracts also demonstrated that the Kalirin-7 protein was highly expressed relative to the other Kalirin forms (Ref. 12 and Fig. 8, B and C).

Characterization of 5.3-kb Kalirin Transcript-- In RNA from adult parietal cortex, the 5.3-kb Kalirin transcript is the most prevalent form (Fig. 2). RT-PCR using a sense primer positioned at the Kalirin start ATG and an antisense primer positioned at the Kalirin-7 stop (Fig. 3A) amplified a single 5-kb fragment (12). Taken together with the Northern blotting data from Fig. 2, we speculated that the 5.3-kb Kalirin transcript possessed a unique 5'-end and was not the result of alternative splicing within the Kalirin-7-coding region. Screening a random primed rat hippocampal cDNA library with a probe to spectrin-like domains 3-8 of Kalirin yielded two clones whose sequences were identical to Kalirin-7 downstream of nt 1735 but distinct from Kalirin-7 upstream of this divergence point (Fig. 3B). Clone 10/29 had 64 bp of new 5' sequence and clone 10/40 extended 30 bp further upstream. Remarkably, this new sequence did not include an in-frame methionine codon but did contain a stop codon in-frame with the downstream Kalirin-coding sequence (Fig. 3B, *). Therefore, if the corresponding mRNAs were translated, initiation from a downstream site would be required. The first in-frame methionine codon encountered (nt 1894-1896) occurs at the start of the fifth spectrin-like domain of Kalirin-7 (Fig. 3, A and B).


View larger version (32K):
[in this window]
[in a new window]
 
Fig. 3.   Kalirin-5 has a unique 5'-end. A, the 5'-coding region of Kalirin-7a is shown along with relevant domains, the identified divergence point (nt 1735), and primers used for RT-PCR (S, sense, and AS, antisense, with arrows). B, Kalirin cDNA inserts having a unique truncated 5'-end were isolated by screening a hippocampal library as described under "Experimental Procedures"; clone 10/29 (2.8 kb) has 64 bp of unique sequence 5' to nt 1735 and clone 10/40 (2.6 kb) extends 30 bp further. The in-frame stop codon is indicated by an asterisk and the BamHI site by B. C, the indicated sense (S) and antisense (AS) primers were used for RT-PCR. Kalirin-5.3 contains an extra 289 bp (stippled region); asterisks indicate in-frame stop codons upstream of the ATG at 1894. The checkered region identifies nt 1735-1894, part of the coding region in Kalirin-7a, but part of the 5'-UTR of Kalirin-5 and -5.3. The vertical striped region indicates 5'-UTR unique to the truncated RNA forms. D, vectors used for in vitro transcription/translation. pKal-5.3TnT lacks the first 64 bp of pKal-5TnT and the first 7 bp of the retained intron (stippled box in C). E, pKal-5TnT and pKal-5.3TnT were translated in vitro using [35S]methionine. Products were separated by SDS-PAGE and analyzed by fluorography. The 32-kDa product produced by both plasmids is consistent with initiation of translation at A1894TG in Kalirin-7a. F, the domains of Kalirin-5 are drawn to scale; predicted molecular mass, 117 kDa.

To confirm the presence of this alternative 5'-end, RT-PCR was performed on RNA prepared from rat cerebral cortex using primers that span the junction at nt 1735 (Fig. 3C). In addition to the expected 569-bp product (Kalirin-5), a second product 289 bp larger in size (Kalirin-5.3) was also found (data not shown). The smaller PCR product was identical in sequence to clones 10/40 and 10/29. The larger PCR product included a 289-bp insert immediately preceding nt 1735 (Fig. 3C). This 289-bp region included no methionine codons but had two in-frame stop codons. Since the 289-bp region began with a 5' intron donor consensus site (Gggtaagg ... ) and terminated with a 3' intron acceptor consensus site ( ... ttgctggcagAA), we postulated that the insert represented a retained intron (17, 18). Consistent with this hypothesis, PCR of rat genomic DNA with primers spanning the 289-bp region yielded a single major product of the size expected for inclusion of this region (data not shown).

There are three short open reading frames upstream of the potential initiator methionine codon at nt 1894; these could reduce initiation of translation at this site (17). To determine whether Kalirin-5 and 5.3 kb were efficiently translated, we used an in vitro transcription and translation system and a plasmid that contained only the first 1.0 kb of Clone 10/29 (Fig. 3D). The first in-frame methionine codon in this plasmid (pKal-5TnT) is 225 bp from the 5'-end of the cDNA and corresponds to Met624 (nt 1894) in Kalirin-7a. The nucleotide sequence at this putative start site is GACAUGU (nt 1891-1897), which is considered an "adequate" context for initiation of translation based on the presence of a G at position -3 (17, 18). Methionine residues next occur at positions 639 and 736. A second plasmid (pKal-5.3TnT) containing the insert sequence described above was also constructed and includes 441 bp upstream of nt 1894.

Initiation of translation at Met624 would yield a protein of 31 kDa. Consistent with this, a protein of approximately 32 kDa was expressed when either plasmid was used (Fig. 3E). In this system, the plasmid with the retained intron (pKal-5.3TnT) was translated more efficiently than the plasmid lacking the 289-bp insert (pKal-5TnT). These results suggest that an internal methionine (Met624) in Kalirin-7 can act as an initiation codon when placed in the context of this shorter mRNA. The protein encoded by the 5.0- or 5.3-kb Kalirin mRNA (117 kDa) would lack the Sec14p-like domain and the first four spectrin-like repeats of Kalirin-7 (Fig. 3F).

Kalirin-9 Encodes a Protein with Two DH-PH Domains-- Western blot analysis of rat parietal cortex extracts using Kalirin-spectrin domain antibodies identified several cross-reactive proteins much larger than Kalirin-7 (190 kDa) or Kalirin-8 (220 kDa) (12). Human Trio and Caenorhabditis elegans UNC-73A are highly homologous to Kalirin-7 and -8 but are larger because they include additional COOH-terminal domains (Fig. 1) (13, 14). Kalirin transcripts of 9, 10, and 12 kb were identified with cDNA probes corresponding to the spectrin-like repeats, the DH-PH domain, or the COOH terminus of Kalirin-8 (Fig. 2, lanes 1-4). To determine the identity of these transcripts, 3'-RACE was performed using a sense primer located within the SH3 domain of Kalirin-8 (Fig. 4A); our goal was to extend the Kalirin sequence in the 3' direction without amplifying the abundant Kalirin-7 and Kalirin-5 cDNAs. After nested PCR, a 2.5-kb fragment was amplified. This cDNA diverged from Kalirin-8 at nt 5508 (after Met1828), adding another 1.55 kb of open reading frame and 0.5 kb of 3'-UTR (Fig. 4A). Another clone was found that did not diverge from Kalirin-8 until nt 5601 (Fig. 4A). When the new 0.5-kb 3'-UTR was used as a Northern blot probe, a major band at 9 kb as well as minor bands at 3.2 and 7 kb were observed; therefore, the major form has been called Kalirin-9 (Fig. 4B, lane 2).


View larger version (31K):
[in this window]
[in a new window]
 
Fig. 4.   Kalirin-9 encodes an additional DH-PH domain. A, the open reading frames of Kalirin-8 (K-8) and Kalirin-9 (K-9) are compared. The nested sense primers used to generate the 2.5-kb RACE product are indicated. The 3'-RACE product includes DH (dark gray box) and PH (diagonal striped box) domains. The RACE clone starts at nt 5143 of Kalirin-8 and diverges after nt 5508. Kalirin-8 includes 215 bp of unique 3'-coding sequence (checkered box). The inverted black triangle indicates the site at which a 93-bp fragment can be spliced out of Kalirin-9. B, rat cortex poly(A)+ RNA (1 µg) was hybridized with either the DH2-PH2 or the 3'-UTR of Kalirin-9 (K-9 3'-UTR) probe. C, the domains of Kalirin-9 are drawn to scale; predicted molecular mass, 270 kDa.

The 270-kDa protein encoded by the Kalirin-9 transcript includes a second DH domain (DH2) with an adjacent PH domain (PH2) about 1 kb downstream from the first DH (DH1) and PH (PH1) domains (Fig. 4C). When a DH2-PH2 probe was hybridized to a Northern blot, major bands of 9 and 12 kb were observed as well as well as minor bands at 3.2, 7, 10, and >15 kb (Fig. 4B, lane 1). However, only the transcripts at 3.2, 7, and 9 kb hybridized to the Kal-9 3'-UTR probe (Fig. 4B, lane 2). Interestingly, a 3.2-kb transcript is also observed in liver, suggesting that a truncated form of this protein only containing the DH2-PH2 domain of Kalirin may be expressed in neuronal and non-neuronal tissue.

Kalirin-12 Also Encodes a Putative Protein Kinase Domain-- Since larger Kalirin transcripts and proteins were detected, we searched for still longer forms of Kalirin. Kawai et al. (19) recently identified human Duet, a 1289 amino acid protein containing adjacent DH and PH domains and a serine/threonine protein kinase domain (Fig. 5A). Comparison of Duet to the DH2 and PH2 domains of Kalirin-9 demonstrated a striking degree of identity. Kalirin-9 (Cys1699 to Lys2371) is 95% identical at the amino acid level to human Duet (amino acids 29-701) (Fig. 5A). The protein kinase domain of Duet also exhibited a high degree of homology to the protein kinase domain of Trio (58% identity). Therefore, we speculated that one or more of the largest Kalirin transcripts would encode a form of Kalirin containing a kinase domain equivalent to the kinase domain of Duet.


View larger version (33K):
[in this window]
[in a new window]
 
Fig. 5.   Kalirin-12 encodes a putative Ser/Thr protein kinase domain. A, the open reading frames of Kalirin-9 (K-9), hDuet, and Kalirin-12 (K-12) are compared. K12s (from Kalirin-9) and K12as (from hDuet) (arrows) indicate the location of primers used in RT-PCR to clone a 4.4-kb fragment that diverged from Kalirin-9 15 bp upstream of the Kalirin-9 stop codon. Kalirin-12 contains the rat homolog of the hDuet Ser/Thr protein kinase (box with wavy dotted lines). The first 66 bp of coding sequence in hDuet (open checkered box) is unique. B, Northern blot analysis of rat cortex poly(A)+ RNA (1 µg) hybridized with probe specific for the DH2-PH2 or Kalirin kinase domain. See Fig. 4B for comparison with Kalirin-9 3'-UTR probe. C, the domains of Kalirin-12 are drawn to scale; predicted molecular mass, 340 kDa.

By using a Kalirin sense primer (K12s) upstream of the region of homology with Duet, and an antisense primer from the 3'-UTR of human Duet (K12as), a 4.4-kb fragment was amplified (Fig. 5A). When sequenced, this fragment was found to encode the Kalirin SH3, DH2, and PH2 domains along with a putative serine/threonine protein kinase domain. After its unique 22-amino acid NH2 terminus, human Duet is 94% identical to this new form of rat Kalirin. Northern blots hybridized with a kinase domain probe exhibited a major 12-kb band, as well as minor bands of 10 kb and >15 kb (Fig. 5B, 2nd lane). A 12-kb message also hybridized to probes specific for spectrin domains 1-3 and 3-8, DH1-PH1, Kal-8 COOH terminus, and DH2-PH2. This 340-kDa protein has been designated Kalirin-12 and contains a Sec14p-like domain, nine spectrin-like repeats, two DH-PH domains, an SH3 domain, and a putative protein kinase domain (Fig. 5C). Note that the kinase domain probe did not recognize a transcript of the size expected for Duet (5.4 kb (19)) in adult parietal cortex.

PCR Analysis of Alternate Kalirin 5' Sequences-- Three different Kalirin 5'-ends (a-c) encoding unique NH2 termini have been identified (Fig. 6A). The a and b forms are designated based on the original cloning of P-CIP10a and -10b, respectively. The rat c form corresponds to human Duo. Each NH2 terminus diverges immediately upstream of the amino acid sequence GSFR (Fig. 6A), suggesting that there are at least three alternatively spliced Kalirin 5'-ends. We previously used RT-PCR to demonstrate that Kalirin-7 utilized each of these 5'-ends when examined at the level of transcription (12).


View larger version (42K):
[in this window]
[in a new window]
 
Fig. 6.   Kalirin transcripts have multiple 5'-ends. A, Kalirin-a, -b, and -c denote three different 3'-ends; unique amino acid sequences are underlined and bold. B, Kalirin-7, -8, -9, and -12 have distinct 3'-ends. RT-PCR analysis of rat parietal cortex RNA was carried out using sense primers Kal-start a, b, and c, and antisense primers Kal-7/3', Kal-8/3', Kal-9/3', and Kal-12/3' (arrows) in all possible combinations. C, cortex total RNA (1 µg) was reverse-transcribed and amplified as described under "Eperimental Procedures" using the indicated primers. Equal aliquots of each PCR reaction were electrophoresed on a 0.8% agarose gel, denatured, and transferred to nitrocellulose for Southern analysis using a probe to Kalirin spectrin domains 1-3. Reverse transcriptase was not added to the reaction (-) to control for any contaminating DNA.

To determine which of these 5'-ends were used in the expression of Kalirin-8, -9, and -12, RT-PCR was performed using sense primers specific for each 5'-end (Kal-start a, b, c) and antisense primers (Kal-7/3', Kal-8/3', Kal-9/3', and Kal-12/3') specific for each 3'-end (Fig. 6B). In the case of Kalirin-12, the antisense primer (Kal-12/3') was placed upstream of the putative kinase domain to minimize the length of the product to be amplified. Rat parietal cortex cDNA was amplified using all primer combinations, and aliquots were electrophoresed on agarose gels and transferred to membranes for Southern analysis (Fig. 6C).

All three 5'-ends could be detected for Kalirin-7, -9, and -12. In contrast, although Kalirin-8 transcripts containing the b and c 5'-ends could be amplified, no Kalirin-8 transcripts having the a 5'-end were detected. A control PCR using Kalirin-8a and -8b templates (2.1 × 10-9 pmol of template/reaction or ~1300 molecules of starting template) showed that the Kal-start a + Kal-8/3' primer pair actually amplified more efficiently than the Kal-start b + Kal-8/3' primer pair (not shown). Since the Kal-start a primer works well on control templates, yet the amount of product generated using this primer in RT-PCR is low, it is likely that relatively few Kalirin transcripts have the Kal-start a 5'-end.

PCR Analysis of Delta Kalirin Forms-- Northern blot analysis with probes specific for Kalirin-9 or Kalirin-12 revealed the presence of minor mRNAs about 2 kb smaller than the major transcripts identified by both probes (Fig. 4B, lane 2; Fig. 5B, lane 2). These hybridization patterns were similar to that observed on blots hybridized with a Kalirin-7-specific probe (Fig. 2, lane 5) in which the smaller Kalirin-5 message had its own unique 5'-end encoding a truncated protein. This consistent pattern suggested that these minor RNA transcripts (7 kb in Fig. 4B, lane 2; 10 kb in Fig. 5B, lane 2) might have the 5'-end identified in Kalirin-5.

To confirm the existence of these transcripts, RT-PCR was again carried out using antisense primers specific for Kalirin-7, -8, -9. or -12 and a sense primer specific for the 5'-end of the Kalirin-5 message (Kal-start Delta ) (Fig. 7A). Products of the expected size were observed by Southern analysis using all four sets of primers (Fig. 7B), consistent with the suggestion that the minor transcripts described above have the predicted short 5'-end. Southern analysis gave similar results for blots probed with either the Kalirin DH1-PH1 region or the 289 bp retained intron from the Kalirin-5 5'-UTR. In an attempt to simplify the naming of the growing number of Kalirin forms, transcripts having the truncated 5'-end have been designated as delta (Delta ) forms. Therefore, Kalirin-5 will be called Delta Kalirin-7, and the forms shown in Fig. 7A will be called Delta Kalirin-8, -9, and -12, respectively.


View larger version (21K):
[in this window]
[in a new window]
 
Fig. 7.   Multiple forms of Kalirin have a truncated 5'-end. A, sense primer (Delta ) specific to the 5'-end of Kalirin-5 was used for RT-PCR in combination with antisense primers Kal-7/3', Kal-8/3', Kal-9/3', and Kal-12/3' (arrows). Delta Kalirin-7 is identical to Kalirin-5 or -5.3 described in Fig. 3. The stippled boxes indicate the 289-bp retained intron. The checkered boxes indicate the region of Kalirin (nts 1735-1894) which becomes 5'-noncoding sequence in the truncated message forms. The bent arrow is located at the putative start site (ATG) at the beginning of spectrin repeat 5. B, total cortex RNA (1 µg) was reverse transcribed, amplified and analyzed as described in Fig. 6; the blot was hybridized with a probe specific for Kalirin DH1-PH1. A cDNA probe specific for the 289 bp retained intron (stippled box in A) hybridized with similar intensities as the DH1-PH1 probe in each case, suggesting that a significant amount of PCR product observed for each primer pair is derived from cDNAs with the retained intron.

Identification of Major Kalirin Proteins in Adult Rat Brain-- To appreciate the functional significance of the many Kalirin transcripts identified, we developed form-specific antisera. Anti-peptide sera specific for the COOH termini of Kalirin-7, Kalirin-9, and Kalirin-12 were produced. In addition, we generated an antiserum specific for the SH3 domain, which should detect Kalirin-9 and Kalirin-12 (and Kalirin-8) but not Kalirin-7 (Fig. 8A). The Kalirin-spectrin antisera generated previously would detect all forms of Kalirin except Duet.


View larger version (59K):
[in this window]
[in a new window]
 
Fig. 8.   Identification of major Kalirin proteins using form-specific antisera. A, the synthetic peptides and recombinant proteins used to generate form-specific antisera are indicated. B, Western blot analysis of Kalirin proteins in extracts of adult rat cortex/hippocampus. Samples were immunoprecipitated with Kalirin-spectrin antibody ("Experimental Procedures"). Homogenate (Input, 50 µg) and immunoprecipitated (IP) proteins (5, 5, 20, 20, and 15% of the total) were detected by Western blotting with Kalirin-spectrin antibody (K-spec) or affinity-purified antibodies specific for individual Kalirin isoforms. Apparent molecular masses of the various isoforms are indicated. C, Western blot analyses of Kalirin proteins in extracts of adult rat cortex, olfactory bulb, and cerebellum (each 40 µg of protein) were performed using the Kalirin-spectrin antibody. The bands marked Kalirin-7 and -12 were identified when the blots were stripped and reprobed with form-specific antisera. The results shown are representative of three or more blots from each tissue.

Kalirin proteins were immunoisolated from extracts of rat cortex and hippocampus using a Kalirin-spectrin antibody. Immune complexes were then fractionated by SDS-PAGE, and proteins were visualized with form-specific Kalirin antisera (Fig. 8B). The Kalirin-spectrin antibody detected Kalirin proteins with apparent molecular masses of 115, 190, 370 and 470 kDa; the most prevalent Kalirin protein had a mass of 190 kDa and was previously identified as Kalirin-7 (12). As expected, the Kalirin forms detected by the Kalirin-spectrin antibody in the immune complexes were the same forms detected in the input homogenate. In addition to the 190-kDa Kalirin-7 protein, the antibody specific for the COOH terminus of Kalirin-7 detected a minor protein of 115 kDa, presumably representing the protein encoded by the Delta Kalirin-7 transcript (predicted mass 117 kDa). Although Delta Kalirin-7 transcripts were more prevalent than Kalirin-7 transcripts, the Delta Kalirin-7 protein was less prevalent.

The major proteins detected by the antibody specific for the SH3 domain corresponded to the 370- and 470-kDa proteins detected by the Kalirin-spectrin antiserum. The antibody to the COOH terminus of Kalirin-9 detected a major 370-kDa protein, supporting its identification as Kalirin-9. The antibody specific for the COOH terminus of Kalirin-12 detected a major 470-kDa protein band, supporting its identification as Kalirin-12. Since the COOH terminus of Kalirin-9 contains only 5 unique amino acids and a 19-amino acid peptide was used to generate the Kalirin-9 antiserum, the minor band at 470 kDa is thought to represent cross-reactivity of the polyclonal antiserum with sequences common to Kalirin-12. Identification of other minor bands detected by the various antisera will require further purification and characterization. Smaller proteins generated by endoproteolysis of Kalirin-7, -9, or -12 would not be identified by this screening method since rabbit antisera were used both for immunoprecipitation and detection.

The nervous system does not show uniform expression of all forms of Kalirin (Fig. 8C). In the cortex, Kalirin-7 is the major form, and Delta Kalirin-7 is detectable. By comparison, in the olfactory bulb, Kalirin-7, -9, and -12 are roughly equally represented. For the same amount of protein loaded onto the gel, there is much less Kalirin in the cerebellum, and the pattern is yet another mixture of Kalirin-related proteins.

Differential Localization of Major Kalirin Proteins in Cortical Neurons-- The Kalirin-7, Kalirin-9, and Kalirin-12 antisera were used to compare the localization of these major Kalirin proteins in primary cultures of cortical neurons. Kalirin-7 was primarily localized to punctate structures distributed along neuronal processes (Fig. 9). By contrast, Kalirin-9 was observed both in neuronal cell bodies and in neuronal processes; the staining pattern observed with the Kalirin-9 antiserum was more uniform within the processes than the punctate staining pattern observed with antiserum to Kalirin-7. Strikingly, Kalirin-12 was largely excluded from neuronal processes and was found primarily in cell bodies; occasional bright, vesicular structures were observed, with many neurons exhibiting a concentration of staining in the perinuclear region.


View larger version (38K):
[in this window]
[in a new window]
 
Fig. 9.   Kalirin isoforms have distinct subcellular localizations in neurons. Seven- and two-day-old cultures of dissociated cerebral cortical neurons were immunostained separately with affinity-purified antibodies specific for Kalirin-7, -9, and -12; all staining was eliminated when the antibodies were pre-bound to their antigenic peptides. Scale bar, 50 µm.

The Multiple Forms of Kalirin-- The Kalirin transcripts identified appear to arise through the combined use of five different 5'-ends, four different 3'-ends, and alternative splicing of internal exons (Fig. 10A). The exon/intron structure of the unc-73 gene has been determined (14). Comparison of the sequences of rat Kalirin and unc-73 indicates that several of the sites where the Kalirin splice variants diverge can be aligned with exon/intron junctions in unc-73. Although almost every combination of 5'-end with 3'-end can be detected by RT-PCR, the most prevalent 5'-ends are Kal-start b and c along with Kal-start Delta  with the 289 bp retained intron; these three 5'-ends are of roughly equal prevalence. The unique sequences of Kal-start b and c do not show significant homology to any sequences in the data base.


View larger version (34K):
[in this window]
[in a new window]
 
Fig. 10.   Summary diagram of forms and how they relate to each other. A, the Kalirin core, common to all Kalirin transcript (except a rat equivalent of human Duet), is shown in the center of the diagram. Alternative splice sites are indicated by dashed lines. Translational initiation sites are indicated by a bent arrow. Known protein domains are indicated. Translated regions are shown by bold lines, with 5'- and 3'-untranslated regions shown by dotted lines. B, the major Kalirin proteins, with their catalytic activities and domain structures, are drawn to scale.

The Delta -Kalirin forms, lacking the Sec14p-like domain, are of special interest. Translation of these transcripts requires initiation at a Met residue located at the start of the 5th spectrin-like repeat. During in vitro transcription/translation of Kalirin-7, there was no indication of internal initiation of translation. The 289-bp retained intron could be involved in regulating Delta -Kalirin expression. Introns near the 5'-ends of mRNAs are often poorly spliced from mRNA precursors (17, 18).

Expression of the Delta -Kalirin forms is likely to be of functional significance. The region of Kalirin preceding the spectrin-like repeats (rat Kalirin-(15-162)) resembles the COOH-terminal, lipid interacting part of a Sec14p domain (11). By mediating the binding of Kalirin to membranes, the Sec14p-like domain could affect its ability to act on its target Rho family members. This same region of Kalirin (rat Kalirin-(1-231)) and homologous regions of Dbl and Ost bind the beta 1 subunit of heterotrimeric G proteins (20). Interestingly, Delta -Kalirin was identified as an iNOS interactor in a yeast two-hybrid screen (8); our subsequent studies indicated that iNOS also interacted with NH2-terminally extended forms of Kalirin.

The genes most closely related to Kalirin also yield multiple transcripts. UNC-73A and -73B correspond to Kalirin-12 and Kalirin-8; UNC-73B lacks the second DH-PH domain along with the Ig and fibronectin type III domains (14). Only two splice variants of Trio have been observed; they differ in the region immediately following the second PH domain (gi 3644048 and gi 3544970). Splicing of Ost (Dbs), a closely related neuronally expressed Dbl family member, is also complex, involving multiple NH2- and COOH-terminal regions with distinct functional domains (21).

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

With the identification of Kalirin-12, the domain structures of Kalirin and Trio (13) can be seen to be similar from NH2- to COOH-terminal end (Figs. 1 and 10B). UNC-73A, a C. elegans protein that plays a cell autonomous role in axon guidance, shares all but the protein kinase domain (14). In the adult rat, expression of Kalirin is limited to the nervous system (9), whereas expression of Trio is widespread (13). UNC-73, the only member of the Trio/Kalirin family identified in C. elegans, is ubiquitously expressed in the early embryo, with expression growing more limited in late embryogenesis and after hatching (14).

The Catalytic Domains of Kalirin-- The second DH domains of rat and human Kalirin (Duet) differ at only 1 of the 198 residues. This second DH domain of Kalirin is most similar to the second DH domains of Trio (68% identity) and UNC-73A (49% identity). The homology of the first and second DH domains of Kalirin (40% identity) is similar to the homology of Kalirin DH-2 to Dbl (39% identity) or to Ost (38% identity). Based on sequence homology, the second DH domain of Kalirin is likely to catalyze GDP/GTP exchange on RhoA (13).

The COOH-terminal kinase domain of Kalirin has the 8 invariant residues common to protein kinases; unlike most protein kinases, the sequence predicted for the Mg2+-binding loop (DFG) connecting beta -strands 8 and 9 has the sequence DLE (22, 23). The kinase domains of rat and human Kalirin (Duet) are 99% identical, whereas the kinase domains of rat Kalirin and human Trio are 58% identical. Duet was autophosphorylated in an in vitro kinase assay, but no exogenous substrate has been identified (19). The kinase domain of Trio has not yet been shown to have catalytic activity (13, 24). If the kinase domain of Kalirin exhibits substrate specificity as high as that of another PAM-CD interactor, P-CIP2 (25), identification of the endogenous substrate may be difficult.

The protein kinases exhibiting the greatest homology to Kalirin kinase are the death-associated protein kinases (40% identity; GenBankTM accession number X76104) and the myosin light chain kinases (36% identity; GenBankTM accession number Q15746). Death-associated protein kinase was identified as a mediator of apoptosis induced by gamma -interferon (26); it is interesting to note that expression of a fragment of Kalirin reversed the inhibitory effect of gamma -interferon/lipopolysaccharide treatment on secretion from corticotrope tumor cells (8). Conventional myosin light chain kinase is a Ca2+/calmodulin-dependent protein kinase that binds tightly to actomyosin-containing filaments (27).

Non-catalytic Domains in Kalirin-- Several of the domains identified in Kalirin may allow it to interact with lipids or other proteins. The second PH domains of rat and human Kalirin are 95% identical. Kalirin PH-2 is 51% identical to Trio PH-2 and 29-33% identical to the first PH domains of Kalirin and Trio, the PH domain of Ost and the two PH domains of UNC-73A. PH domains have been implicated in binding to phosphoinositides, to the beta gamma subunits of heterotrimeric G proteins, and to cytoskeletal components and are essential to the normal function of Dbl family members (1, 28, 29). The first PH domain of Trio targets the protein to the cytoskeleton (30).

When the region between the first and second DH-PH domains of Kalirin was analyzed, two interesting homologies were revealed. First the SH3-like domain is homologous to the third SH3 domain in Dock, a Drosophila SH2/SH3 adapter protein involved in photoreceptor cell axon guidance and targeting (32% identity; GenBankTM accession number U57816) (31). This region is immediately followed by a 194-amino acid region exhibiting 23% identity to ABP1 (GenBankTM accession number X59720), a yeast actin-binding protein (Fig. 10B).

The region between the second DH-PH domain and the Ig-like domain that precedes Kalirin-kinase has a Pro-rich region that shares homology with a region of N-WASP (43% identity over a 44-amino acid region with 10 of 13 Pro residues conserved). In N-WASP, a protein that regulates actin polymerization, this proline-rich region binds to profilin and Ash/Grb2 (32).

An Ig-like region precedes the Kalirin-kinase domain; this 214-amino acid region, comprising two Ig repeats, is 96% identical in rat and human Kalirin. Human Trio contains one Ig repeat that is 55% identical to Kalirin. Other intracellular molecules containing Ig repeats include giant kinases like titin (GenBankTM accession number I38344) and twitchin (GenBank accession number Z73897); several of the Ig repeats in titin exhibit 30% identity to this region of Kalirin. These independently folded globular domains are generally protease-resistant and facilitate specific protein-protein interactions (33, 34). In both titin and myosin light chain kinase, the region homologous to Kalirin includes the kinase domain and the preceding Ig domain. The Ig motif at the NH2 terminus of myosin light chain kinase is involved in its binding to filaments (27). Interestingly, Duet, which includes the entire Ig domain, localized to actin-associated cytoskeletal elements following transient expression in NIH-3T3 cells (19).

The Major Kalirin Proteins-- In order to relate the Kalirin proteins detected on Western blots to the major Kalirin transcripts, we used immunoprecipitation with an antibody to the spectrin-like repeats followed by Western blot analysis with more selective antisera. The data presented are all from adult rat cortex; our preliminary data indicate that dramatic changes in Kalirin expression occur during early postnatal development. Many factors make it difficult to relate mRNA prevalence to protein levels, but Kalirin-7 is identified as a major form by both methods. Although the Delta -Kalirin-7 mRNA is prevalent, the corresponding 115-kDa protein is barely detectable; diminished translational efficiency or rapid protein turnover could contribute to the pattern observed. At present we do not have reagents to distinguish Kalirin proteins with the a, b, or c sequences at their NH2 terminus. Based on both Northern and Western blot analysis, levels of Kalirin-9 and Kalirin-12 expression are similar. Although these Kalirin proteins are not expressed at levels as high as Kalirin-7, the fact that they are localized differently in cultured neurons suggests that they subserve specific functions.

At least in adult rat cortex, we see no evidence for expression of a protein equivalent to human Duet (19). Northern blots hybridized with probes that would identify a Duet-like transcript lack a band in the appropriate size range. A Duet-like protein would have a mass of about 144 kDa and be visualized by the Kalirin-kinase antiserum. No protein of this mass was observed. Our initial studies indicate that Kalirin is expressed in many non-neuronal sites during embryogenesis, and a Duet-like protein may be identified. Alternatively, expression of a Duet-like protein may reflect species-specific splicing of this complex gene. Comparison of rat, mouse, and human genomic DNA indicates that the 289-bp retained intron in Delta -Kalirin is unique to rat (data not shown).

The Biology of Dual Function GEFs-- In considering the functions of Kalirin, it is useful to estimate its size in comparison to the components with which it might interact. Crystal structures have been determined for many of the domains found in Kalirin. The partial Sec14p-like domain should span approximately 3 nm (35). The three-helix bundle of each spectrin-like repeat would be approximately 5 nm long (36). The first DH-PH domain of Trio is 7 nm long (37). An SH3 motif is another 3 nm. The two Ig repeats could span about 8 nm (33). The protein kinase domain should form an ellipsoid about 8 nm in length (38). Stretched out in a linear form, the length spanned by one Kalirin molecule could reach 85 nm or more. The vesicles used for trafficking are 40-200 nm in diameter, and a single actin filament is about 7 nm in diameter. Thus, Kalirin is large enough that one of its functions may be tethering together components whose functions must be coordinately regulated, much as spectrin and ankyrin skeletons function in networks near organelles (39-41).

Although the Dbl family is large, Kalirin, Trio and unc-73 are the only members with multiple Dbl domains. As such, they can be expected to coordinate the actions of the different Rho family members with which their two catalytic domains interact. Trio, identified by its interaction with LAR, a receptor-like protein tyrosine phosphatase, is thought to coordinate spatial and temporal changes in the actin cytoskeleton with cell-matrix interactions that are important in cell growth and migration (42). C. elegans with the most severe unc-73 mutations are nearly paralyzed and have withered tails, ectopic subventral vulvae, and defects in egg laying (14). Overexpression of Kalirin-8 in corticotrope tumor cells results in altered metabolism of PAM and marked extension of branched neuritic processes (9), and Kalirin-7 is highly localized to the post-synaptic density fraction in adult rat brain. The activity of Dbl-family members can be affected by phosphorylation, binding to membranes, and by the non-catalytic NH2-terminal domain (1). A key to understanding the functions of Kalirin will be identification of inputs regulating its localization and its many potential activities. The distinctly different localization of Kalirin-7, Kalirin-9, and Kalirin-12 in cortical neurons suggests different roles for each protein. Kalirin, through its several catalytic domains and multiple protein and lipid interacting regions, may coordinate changes in the actin cytoskeleton and the secretory machinery with peptide processing events in the lumen of the secretory pathway (43).

    ACKNOWLEDGEMENTS

We thank Henry Keutmann (Massachusetts General Hospital) for synthesis of the peptides used to generate antisera; Kate Deanehan for affinity purification of antisera; and Marie Bell and Lixian Jin for general laboratory assistance.

    FOOTNOTES

* This work was supported by National Institutes of Health Grant DK-32948.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Dagger Current address and to whom correspondence should be addressed: Dept. of Neuroscience, University of Connecticut Health Center, MC3401, 263 Farmington Ave., Farmington, CT 06030-3401. Tel.: 860-679-8894; Fax: 860-679-8766; E-mail: mains@nso.uchc.edu.

Published, JBC Papers in Press, April 20, 2000, DOI 10.1074/jbc.M000676200

    ABBREVIATIONS

The abbreviations used are: GEFs, GDP/GTP exchange factors; DH, Dbl (deleted in B-cell lymphoma) homology; PAM, peptidylglycine alpha -amidating monooxygenase; pBS, pBluescript; P-CIP, PAM-COOH-terminal interactor protein; PH, pleckstrin homology; RACE, rapid amplification of cDNA ends; RT-PCR, reverse transcription-polymerase chain reaction; PAGE, polyacrylamide gel electrophoresis; SH3, src homology 3; iNOS, inducible nitric-oxide synthase; nt, nucleotides; kb, kilobase pair; bp, base pair; UTR, untranslated region; RACE, rapid amplification of cDNA ends.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

1. Cerione, R. A., and Zheng, Y. (1996) Curr. Opin Cell Biol. 8, 216-222
2. Whitehead, I. P., Campbell, S., Rossman, K. L., and Der, C. J. (1997) Biochim. Biophys. Acta 1332, F1-F23
3. Kjoller, L., and Hall, A. (1999) Exp. Cell Res. 253, 166-179
4. Zohn, I. M., Campbell, S. L., Khosravi-Far, R., Rossman, K. L., and Der, C. J. (1998) Oncogene 17, 1415-1438
5. Kins, S., Betz, H., and Kirsch, J. (2000) Nat. Neurosci. 3, 22-29
6. Alam, M. R., Caldwell, B. D., Johnson, R. C., Darlington, D. N., Mains, R. E., and Eipper, B. A. (1996) J. Biol. Chem. 271, 28636-28640
7. Colomer, V., Engelender, S., Sharp, A. H., Duan, K., Cooper, J. K., Lanahan, A., Lyford, G., Worley, P., and Ross, C. A. (1997) Hum. Mol. Genet. 6, 1519-1525
8. Ratovitski, E. A., Alam, M. R., Quick, R. A., McMillan, A., Bao, C., Hand, T. A., Johnson, R. C., Mains, R. E., Eipper, B. A., and Lowenstein, C. J. (1999) J. Biol. Chem. 274, 993-999
9. Alam, M. R., Johnson, R. C., Darlington, D. N., Hand, T. A., Mains, R. E., and Eipper, B. A. (1997) J. Biol. Chem. 272, 12667-12675
10. Mains, R. E., Alam, M. R., Johnson, R. C., Darlington, D. N., Back, N., Hand, T. A., and Eipper, B. A. (1999) J. Biol. Chem. 274, 2929-2937
11. Aravind, L., Neuwald, A. F., and Ponting, C. P. (1999) Current Biology 9, R195-R197
12. Penzes, P., Johnson, R. C., Alam, M. R., Kambampati, V., Mains, R. E., and Eipper, B. A. (2000) J. Biol. Chem. 275, 6395-6403
13. Debant, A., Serra-Pages, C., Seipel, K., O'Brien, S., Tang, M., Park, S. H., and Streuli, M. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 5466-5471
14. Steven, R., Kubiseski, T. J., Zheng, H., Kulkarni, S., Mancillas, J., Morales, A. R., Hogue, C. W. V., Pawson, T., and Culotti, J. G. (1998) Cell 92, 785-795
15. Bloomquist, B. T., Eipper, B. A., and Mains, R. E. (1991) Mol. Endocrinol. 5, 2014-2024
16. Johnson, R. C., Darlington, D. N., Hand, T. A., Bloomquist, B. T., and Mains, R. E. (1994) Endocrinology 135, 1178-1185
17. Kozak, M. (1996) Mamm. Genome 7, 563-574
18. Kozak, M. (1999) Gene (Amst.) 234, 187-208
19. Kawai, T., Sanjo, H., and Akira, S. (1999) Gene (Amst.) 227, 249-255
20. Nishida, K., Kaziro, Y., and Satoh, T. (1999) FEBS Lett. 459, 186-190
21. Lorenzi, M. V., Castagnino, P., Chen, Q., Hori, Y., and Miki, T. (1999) Oncogene 18, 4742-4755
22. Hanks, S. K., and Quinn, A. M. (1998) in Protein Phosphorylation (Sefton, B. M. , and Hunter, T., eds), Vol. 2 , pp. 39-63, Academic Press, Boston
23. Taylor, S. S., and Radzio-Andzelm, E. (1994) in Cyclic AMP-dependent Protein Kinase (Woodgett, J. R., ed) , pp. 1-29, IRL Press at Oxford University Press, Oxford
24. Bellanger, J. M., Lazaro, J. B., Diriong, S., Fernandez, A., Lamb, N., and Debant, A. (1998) Oncogene 16, 147-152
25. Caldwell, B. D., Darlington, D. N., Penzes, P., Johnson, R. C., Eipper, B. A., and Mains, R. E. (1999) J. Biol. Chem. 274, 34646-34656
26. Deiss, L. P., Feinstein, E., Berissi, H., Cohen, O., and Kimchi, A. (1994) Genes Dev. 9, 15-30
27. Stull, J. T., Lin, P. J., Krueger, J. K., Trewhella, J., and Zhi, G. (1998) Acta Physiol. Scand. 164, 471-482
28. Wang, D. S., Deng, T., and Shaw, G. (1997) Biochem. Biophys. Res. Commun. 234, 183-189
29. Michiels, F., Stam, J. C., Hordijk, P. L., van der Kammen, R. A., Ruuls-Van Stalle, L., Feltkamp, C. A., and Collard, J. G. (1997) J. Cell Biol. 137, 387-398
30. Bellanger, J. M., Zugasti, O., Lazaro, J. B., Diriong, S., Lamb, N., Sardet, C., and Debant, A. (1998) C. R. Soc. Biol. 192, 367-374
31. Garrity, P. A., Rao, Y., Salecker, I., McGlade, J., Pawson, T., and Zipursky, S. L. (1996) Cell 85, 639-650
32. Suetsugu, S., Miki, H., and Takenawa, T. (1998) EMBO J. 17, 6516-6526
33. Kenny, P. A., Liston, E. M., and Higgins, D. G. (1999) Gene (Amst.) 232, 11-23
34. Gregorio, C. C., Granzier, H., Sorimachi, H., and Labeit, S. (1999) Curr. Opin. Cell Biol. 11, 18-25
35. Sha, B., Phillips, S. E., Bankaitis, V. A., and Luo, M. (1998) Nature 391, 506-510
36. Yan, Y., Winograd, E., Viel, A., Cronin, T., Harrison, S. C., and Branton, D. (1993) Science 262, 2027-2030
37. Liu, X., Wang, H., Eberstadt, M., Schnuchel, A., Olejniczak, E. T., Meadows, R. P., Schkeryantz, J. M., Janowick, D. A., Harlan, J. E., Harris, E. A., Staunton, D. E., and Fesik, S. W. (1998) Cell 95, 269-277
38. Zheng, J., Knighton, D. R., Xuong, N. H., Parello, J., Taylor, S. S., and Sowadski, J. M. (1992) Acta Crystallogr. Sect. B Struct. Sci. 48, 241-244
39. DeMatteis, M. A., and Morrow, J. S. (1998) Curr. Opin. Cell Biol. 10, 542-529
40. Holleran, E. A., and Holzbaur, E. L. (1998) Trends Cell Biol. 8, 26-29
41. Stankewich, M. C., Tse, W. T., Peters, L. L., Ch'ng, Y., John, K. M., Stabach, P. R., Devarajan, P., Morrow, J. S., and Lux, S. E. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 14158-14163
42. Seipel, K., Medley, Q. G., Kedersha, N. L., Zhang, X. A., O'Brien, S. P., Serra-Pages, C., Hemler, M. E., and Streuli, M. (2000) J. Cell Sci. 112, 1825-1834
43. Zhou, A., Webb, G., Zhu, X., and Steiner, D. F. (1999) J. Biol. Chem. 274, 20745-20748
44. Graziani, G., Ron, D., Eva, A., and Srivastava, S. K. (1989) Oncogene 4, 823-829


Copyright © 2000 by The American Society for Biochemistry and Molecular Biology, Inc.
Add to CiteULike CiteULike