Nucleotide sequence analysis of the human KCNJ1 potassium channel locus

Gene 188 (1997) 9–16

Nucleotide sequence analysis of the human KCNJ1 potassium channel locus Jeffrey H. Bock a, Mary E. Shuck b, Christopher W. Benjamin b, Mea Chee a, Michael J. Bienkowski b, Jerry L. Slightom a,* a Molecular Biology Unit 7242-267-510, Pharmacia and Upjohn Company, Kalamazoo, MI 49007, USA b Cell and Inflammation Biology Unit 7239, Pharmacia and Upjohn Company, Kalamazoo, MI 49007, USA Received 23 August 1996; revised 1 October 1996; accepted 2 October 1996; Received by A. Dugaiczyk

Abstract Detailed analyses of transcripts encoding various isoforms of the human potassium ( K+, inward rectifying) channel ROM-K (also referred to as K 1.1) revealed the existence of at least five distinct transcripts [Shuck et al., J. Biol. Chem. 269 (1994) ir 24261–24270]. These five hROM-K transcripts appear to be the result of alternative splicing of five exons. The nucleotide sequence of the genomic DNA including and spanning these exons (the KCNJ1 locus) was obtained directly from l and P1 clones (a total of 40 kb). The organization of the hKCNJ1 gene was determined by combining this sequence information with data obtained from primer extension and RT-PCR experiments. It appears that the hKCNJ1 gene utilizes multiple promoters, with promoterlike elements found 5∞ of exons 1, 4, or 5. The promoter 5∞ of exon 5 was unexpected; thus, it appears that the hKCNJ1 gene is capable of producing six distinct hROM-K transcripts via the use of three promoters and alternative splicing of five exons. Comparisons of the rat and human ROM-K cDNA sequences find human homologs (orthologs) for two of the three distinct rROM-K transcripts. A search of the complete human KCNJ1 sequence with the exon sequence that defines the other rROM-K transcript located a region of shared nucleotides, a putative sixth exon, in the hKCNJ1 gene. This finding suggests that the rKCNJ1 gene may contain an exon that is no longer or infrequently used in transcripts derived from the hKCNJ1 gene. Keywords: Alternative splicing; Chromosome 11; Direct sequencing; K 1.1; Multiple promoters; Multiple transcripts; Rat ortholog; ir Repetitive DNA; ROM-K

1. Introduction We previously reported (Shuck et al., 1994) the isolation of five distinct hROM-K transcripts, hROM-K1 to K5, by screening a human kidney specific cDNA library with the rROM-K1 cDNA clone described by Ho et al. (1993) or by using RT-PCR based cloning methods. Although five distinct hROM-K transcripts were identified, they apparently encode only three polypeptides that differ in their N-terminal extensions (hROM-K1, * Corresponding author. Tel. +1 616 8331304; Fax +1 616 8332599; e-mail: [email protected] Abbreviations: aa, amino acid(s); bp, base pair(s); h, human; KCNJ1, genetic locus symbol for the ROM-K gene; LINE, long interspersed element; MER, medium reiteration frequency repeat; nt, nucleotide(s); PCR, polymerase chain reaction; r, rat; ROM-K, kidney ‘ATP regulated’ potassium channel; ROM-K, gene (RNA or DNA) encoding ROM-K; RT, reverse transcriptase; SINE, short interspersed element; tsp transcription start point(s); UTR, untranslated region(s)

-K2 and -K3). Alternative splicing appeared to be the major mechanism responsible for generating these five hROM-K transcripts. To obtain a better description of the hKCNJ1 gene and understanding of the molecular mechanism(s) responsible for generating the five hROMK transcripts, we and others have isolated genomic clones containing the KCNJ1 locus. Two overlapping l clones that spanned exons 4 and 5 were isolated by Yano et al. (1994), and they used inverse-PCR to isolate and show linkage between exons 1 and 2. Yano et al. (1994) did not physically link exons 1 and 2 to exons 4 and 5 or isolate any genomic DNA region containing exon 3, but they did map the KCNJ1 locus to human chromosome 11q24. We isolated a series of overlapping genomic clones, one l and three P1, that appeared to link all five KCNJ1 exons. These l and P1 clones were directly sequenced using the cycle sequencing procedures. From this genomic DNA sequence information and subsequent 5∞/3∞ primer extension and RT-PCR experiments we deter-

0378-1119/97/$17.00 Copyright © 1997 Elsevier Science B.V. All rights reserved PII S 03 7 8 -1 1 1 9 ( 9 6 ) 0 0 7 59 - 7

10

J.H. Bock et al. / Gene 188 (1997) 9–16

mined the organization of the hKCNJ1 gene. These analyses of the hKCNJ1 gene and its transcripts support the existence of three hKCNJ1 promoters with the third promoter producing a sixth distinct transcript (hROMK6) that was not identified previously (Shuck et al., 1994). We compared the three distinct rROM-K transcript sequences (Ho et al., 1993; Boim et al., 1995) with these six hROM-K transcripts and found that the exons that define two of the three rROM-K transcript have human orthologs. The exon that defines the third rROM-K transcript appears to have a human ortholog; however, it may no longer be included in any hROM-K transcripts. In addition, the 40 kb KCNJ1 sequence was subjected to computer-aided searches to locate repetitive DNA elements, the results of which show that this region of chromosome 11 contains relatively few repetitive elements.

2. Experimental and discussion 2.1. Cloning and primer-directed sequencing of hKCNJ1 Our investigation of the hKCNJ1 locus was initiated by screening a human genomic DNA-l phage library, followed by the screening of a human P1 library by GenomeSystems (St. Louis, MO, USA) with PCR primer pairs specific for exon 1 and 5. A total of four clones were obtained, lHGK-2 and three P1 clones, nos. 395, 851 and 901. The relative locations of these overlapping clones are shown in Fig. 1A. Our DNA sequencing strategy was to sequence directly the l and P1 clones because of our previous experience in obtaining sequence information from large DNA molecules, such as 50 kb cosmids (Siemieniak et al., 1991; Slightom et al., 1994) and 130 kb baculovirus genome (Slightom and Sieu, 1992). However, we decided to use the Taq DNA polymerase cycle sequencing procedure instead of a T7 polymerase-based method because it allowed us to use much less template DNA. The primer-directed sequencing strategy was considered feasible because the five exons and the overlapping clone junctions ( Fig. 1A) provided a total of 14 primer-walking initiation points. The total amount of sequence information obtained was 40 kb which is available from GenBank (accession no. U65406). The primer-walking sequencing strategy proceeded smoothly because this region of chromosome 11 contained few repetitive DNA elements (SINEs, LINEs, or MERs, see Section 2.3). Only one DNA region presented a problem to the primer-walking sequencing strategy, an A+T-rich (90%) region (>559 bp) located at the extreme 5∞-end of the sequence. Our 5∞ sequencing effort was terminated at this point because we lacked any evidence supporting the existence of hKCNJ1 exons

extending 5∞ of this A+T-rich region, and it appears that the exon 1 5∞ flanking sequence obtained (1 kb) contains the proximal elements associated with the exon 1 promoter (Fig. 2 and Section 2.2). 2.2. The hKCNJ1 gene: multiple promoters but a single poly(A) addition region Comparison between the hKCNJ1 sequence and the hROM-K cDNA sequences reported by Shuck et al. (1994) defines much of the organization of the hKCNJ1 gene as outlined in Figs. 1A,B and 2. The sizes of these exons and introns are listed in Table 1. The spacing of exon/intron sequences varis considerably, with exons 1 and 2 being separated by 517 bp (intron 1) while exons 2 and 3 are separated by more than 15.6 kb (intron 2). However, because of alternative splicing, intron sizes are dependent upon which exons are used; e.g., the transcript referred to as hROM-K2 (Shuck et al., 1994) splices exon 1 onto exon 5 which spans 26.9 kb. Shuck et al. (1994) suggested that the five hROM-K transcripts (GenBank accession No. U12541 to U12546) initiate from either the 5∞ end of exon 1 (tsp at nt 972) or exon 4 (tsp at nt 25874) ( Fig. 2). This expectation is supported by finding sequences that closely match the consensus RNA polymerase II promoter elements, CCAAT and TATAAA ( Efstratiadis et al., 1980) in the 5∞ flanking DNAs of exons 1 and 4. These CCAAT and TATAAA elements are located 118 and 32 bp, respectively, 5∞ of the exon 1 tsp and 89 and 30 bp, respectively, of exon 4 tsp ( Fig. 2). The possibility that the 5∞ flanking regions of exons 2, 3 and 5 also contain promoter elements was investigated by Shuck et al. (1994), and the evidence suggested that exons 2 and 3 are only involved in alternative splicing of transcripts derived from the exon 1 promoter. This suggestion is supported by our analysis of their 5∞ flanking sequences which finds no close sequence matches to these consensus RNA polymerase II elements. Shuck et al. (1994) did provide evidence derived from RT-PCR experiments for the expression of exon 5 in the absence of exons 1–4 in certain tissues. To complete this analysis we subjected exon 5 to a 5∞ primer extension analysis to determine the location of any exon 5 tsp. The 32P-labeled products from this 5∞ extension reaction (Fig. 3) revealed a transcript that extends exon 5 by 12 bp, indicating that a third hKCNJ1 tsp is located at nucleotide position 28 066 ( Fig. 2). This result is supported by the finding of consensus RNA polymerase II promoter element sequences, CCAAT and TATAAA, located 111 and 33 bp, respectively, 5∞ of this tsp. This is an interesting finding because it suggests the existence of a sixth distinct hROM-K transcript, hROM-K6. Assuming that translation of hROM-K6 is initiated at the same ATG codon as hROM-K2, -K4 and -K5 the deduced polypeptide would be identical to hROM-K2 (Fig. 2; and Shuck

J.H. Bock et al. / Gene 188 (1997) 9–16

11

Fig. 1. Organization of the hKCNJ1 gene and schematic maps of human and rat ROM-K transcripts. (A) The bold line represents the region of human chromosome 11 that contains the hKCNJ1 locus, and the vertical lines with numbers above show the location of the five known hKCNJ1 exons (Shuck et al., 1994; Yano et al., 1994). The location and identity of the overlapping l and P1 clones containing inserts from this chromosomal region are shown below the gene map. (B) Schematic map showing the alternative spliced exons used to derive each of the five hROM-K transcripts described by Shuck et al. (1994). Solid line indicates exons that correspond to the exon numbers shown above in (A) and dashed lines indicate introns. (C ) Schematic map showing the alternative spliced exons used to derive each of the three rROM-K transcripts described by Boim et al. (1995). Rat ROM-K exons are indicated below the corresponding hROM-K exons, except for the second exon shown in the rROM-K3 transcripts (see text). Parallel lines dissecting the dashed lines of rROM-K2 and -K3 transcripts indicate that the size of these rat introns remains to be determined. Methods: The human genomic library lFixII (5×105 clones, Stratagene No. 946203) was screened with a 32P-labeled probe prepared from the hROM-K1 cDNA (Shuck et al., 1994) using procedures described by Sambrook et al. (1989). A single l clone referred to as lHGK-2 was completely sequenced. P1 clones were obtained by GenomeSystems using exon 1 and 5 specific primer pairs (exon 1 primers 5∞-GGTTGCATACAGATGAGTTGGCAG and 5∞-CATGGATTGCTGGAGTAAATGTAGAA; exon 5 specific primer pair, 5∞-TATGGCAGTCACATTTATGGAAAGC and 5∞-TCATCTGTTTCATTGACTTCTGACAA). lHGK-2 was grown as phage lysate using E. coli bacterial host DP50 supF. lHGK-2 DNA was isolated using the procedure described for the Qiagen (Chatsworth, CA, USA) l Midi kit. P1 DNAs were purified using the Qiagen Plasmid Midi kit procedure. This procedure was modified in its final step by heating the DNA elution buffer (QF buffer) to 50°C prior to loading onto the column. This was carried out at Qiagen’s recommendation (see Qiagen News no. 3, 1994) to elute the large P1 plasmids from the Qiagen resin. P1 DNA purifications generally yielded between 5 and 10 mg per 100 ml of growth, and generally the purity was adequate to allow direct sequencing using either the 32P- or fluorescence-labeled cycle sequencing procedures.

et al., 1994). It should be noted that hROM-K5 has the potential to initiate translation at the same point as that found for hROM-K3, within exon 2; however, for hROM-K5 this putative translation is terminated by the addition of exon 3 (see Figs. 1B and 2 and Shuck et al., 1994). Thus, it appears that for some unknown reason the hKCNJ1 gene potentially uses two promoters and alternative splicing to produce four distinct transcripts

that are capable of encoding the identical hROM-K2 polypeptide. The longest hROM-K cDNA clones described by Shuck et al. (1994) contained about 1800 bp, but only included a 3∞ UTR of 380 bp. The length of the 3∞ UTR was expected to be considerably longer because the size of most hROM-K transcripts was estimated by Northern blot analysis to be about 2800 nt (Shuck et al., 1994).

12

J.H. Bock et al. / Gene 188 (1997) 9–16

J.H. Bock et al. / Gene 188 (1997) 9–16

To determine the 3∞ end of these transcripts we performed RT-PCR experiments using human kidney mRNA, an exon 5 sense primer ( located 125 bp 3∞ of the translation termination codon) and a dT/anchor primer designed to anneal with the poly(A) region of the mRNA. This PCR amplification yielded DNA fragments of about 850 bp in length (data not shown) which were cloned and sequenced (three independent clones). The consensus sequence showed that the hKCNJ1 3∞ UTR does not contain any additional exon/intron boundaries and the presence of a poly(A) track in these cDNAs (after genomic nt positions 30 322) suggests that all transcripts utilize a single, non-consensus, polyadenylation signal (AATATA, Fig. 2). Thus, for most of the hROM-K transcripts exon 5 is 2245 bp in length, except for hROM-K6 which contains an additional 12 bp. The 3∞ UTR of the hKCNJ1 gene appears to be much simpler than that found for the rKCNJ1 gene which appears to use multiple polyadenylation signals and alternative splicing (Boim et al., 1995). 2.3. Comparison of rat and human ROM-K transcripts; identification of orthologous exons ROM-K transcript sequences have been obtained from only one other species, rat, and thus far, four transcripts, rROM-K1, -K2a, K2b and -K3, have been identified (Ho et al., 1993; Zhou et al., 1994; Boim et al., 1995). Three of these transcripts appear to be distinct, rROM-K1, K2 and -K3 (Fig. 1C ), but their orthologous relationships with the hKCNJ1 exons have not been established (Boim et al., 1995). For this comparative analysis we will (when appropriate) refer to the orthologous human and rat KCNJ1 exons using the exon numbering system established for the hKCNJ1 gene (Figs. 1A and 2); the organization of the rKCNJ1 gene has not yet been

13

determined. Similar to the hROM-K transcripts, all of the rROM-K transcripts share a common 3∞ core region (exon 5 for hROM-K transcripts) that encodes the major portion of the rROM-K polypeptide. These human and rat exon 5 coding regions encode 372 aa which share 89 and 87% identities at the amino acid and nucleotide levels, respectively. The assignment of orthologous exon pairs for the remaining rat and human KCNJ1 exons is expected to be more difficult because these exons contain very little or no coding DNAs. The degree of nucleotide sequence identity shared between known orthologous noncoding regions of rodent and human genes has been found to be in the range of 53% or higher ( Koop and Hood, 1994). That is, it is doubtful that regions which share less than 53% nucleotide sequence identity are indeed orthologous. The first rat and human ROM-K transcripts to be identified, ROM-K1 ( Fig. 1B,C ), were given the same designation (Ho et al., 1993; Shuck et al., 1994) and they are indeed orthologs as they include exons 4 and 5 (GenBank accession Nos. X72341 and U12541, respectively). Both rat and human exon 4 regions encode 12 aa (sharing 7/12 aa or 58% identity) which are spliced in frame onto exon 5. However, a comparison of these putative orthologous exon 4 regions does reveal some differences; rat exon 4 is considerably longer than the human exon 4, 186 bp versus 87 bp, respectively. The 52 bp of noncoding human exon 4 is orthologous with the corresponding nucleotides of the rat exon 4 as they share 74% identity. The question which remains is, what does the remaining 99 bp of the rROM-K1 5∞ UTR correspond to in the human KCNJ1 gene? The possibility exists that this additional 99 bp of the rROM-K1 transcript is contained within an additional 5∞ exon. A search of the complete hKCNJ1 sequence with these rROMK1 99 bp finds the best match to be located immediately

Fig. 2. The nucleotide sequence of hKCNJ1 exon and promoter regions. The nucleotide sequence of the hKCNJ1 gene is presented in a noncontinuous form (as indicated by the vertical double-headed arrows), the remainder of the sequence can be obtained from GenBank (accession No. U65406). The locations of the exons that are parts of the various hROM-K transcripts identified by Shuck et al. (1994) are labeled above the sequence line. The deduced amino acids are presented below the counting line in single-letter codes that are placed at the first nucleotide of the corresponding codon. The location of introns that interrupt this gene are indicted by arrows (,). Putative proximal promoter elements (CCAAT and TATAAA, see 5∞ flanking DNA regions for exons 1, 4 and 5) and poly(A) addition elements (AATAAA) are indicated above the nucleotide sequence line. Methods: Both l and P1 template DNAs were sequenced using the cycle sequencing procedure described by Perkin-Elmer/Applied Biosystems Division (PE/ABD, Foster City, CA, USA) with Taq DNA polymerase. Cycle sequencing reactions (32P-labeled primers) involved using about 1 mg of template DNA (l or P1), an initial denaturation at 95°C for 1 min, followed by 30 cycles (denaturation 95°C for 30 s, primer annealing at 55°C for 30 s, and extension at 70°C for 1 min). Cycle-sequenced samples were electrophoresed on 1 m thermostated 5% Long Ranger@ (AT Biochem, Malvern, PA, USA) sequencing gels (Slightom et al., 1994), and sequence readings in the range of 600–700 bp were obtained for both l and P1 template DNAs. Fluorescence-based sequencing (PE/ABD, PRISM ready dye-deoxy terminators) was performed using about 1 mg of DNA and the following cycle sequencing reaction conditions: initial denaturation at 98°C for 1 min, followed by 50 cycles of 96°C for 30 s, annealing at 50°C for 1 min, and extension at 60°C for 4 min. Extension products were purified using Centriflex gel filtration cartridges (Advanced Genetic Technologies, Gaithersburg, MD, USA). Column-purified samples were dried under vacuum for about 40 min and then dissolved in 5 ml of a DNA loading solution (83% deionized formamide, 8.3 mM EDTA and 1.6 mg/ml blue dextran). The samples were then heated to 90°C for 3 min and loaded into the gel sample wells for sequence acquisition by the ABI373A sequencer (stretch modification). Sequence analysis was carried out by importing ABI373A files into the Sequencer program, obtained from Gene Codes (Ann Arbor, MI, USA), and generally sequence readings of 600–700 bp were obtained for both l and P1 template DNAs. For both 32P-labeled and fluorescence-based sequencing, errors were minimized by obtaining sequence information from both DNA strands, and the 32P-labeled sequences were proofread from the films back into the assembled sequence to ensure accuracy.

14

J.H. Bock et al. / Gene 188 (1997) 9–16

Table 1 Exon and intron repetitive DNA elements within the hKCNJ1 locus Exon and intron locations

1 2 3 4 5 6 7 9 10

Gene region

5∞-end–3∞-end

Length (bp)

Exon 1 Intron 1 Exon 2 Intron 2 Exon 3 Intron 3 Exon 4 Intron 4 Exon 5

971–1 169 1 170–1 686 1 687–1 819 1 820–17 451 17 452–17 620 17 621–25 873 25 874–25 960 25 961–28 077 28 078–30 322

199 517 133 15 632 169 8 253 87 2 117 2 245

Noncoding/coding 199/0 – 103/30 – 169/0 – 51/36 – 1 105/1 140

Repetitive DNA element locations Type

Position

Length

Orientation

Identity (%)

Class

Alu 1 2 3 4 5 6 7 8

3 961/4 245 8 795/9 077 19 850/20 031 20 045/20 334 20 335/20 462 37 950/38 240 38 950/39 079 39 093/39 379

285 283 182 290 128 291 130 287

[ [ [ Z [ Z [ [

85.3 79.2 78.6 91.7 81.2 90.4 81.5 79.4

Sx J J Sb J Sx J J

LINE 1 2 3 4

13 837/13 883 14 968/15 134 16 233/16 273 28 029/28 089

47 167 41 61

[ Z Z Z

72.2 64.7 80.5 68.9

MER 1 2 3 4 5 6 7 8 9

18/73 84/253 103/168 172/311 406/546 421/677 14 741/14 910 14 911/15 057 32 584/32 674

56 170 66 131 141 244 174 151 91

[ Z Z [ Z Z [ [ [

75.0 58.6 79.8 65.7 65.2 73.8 65.9 63.5 69.8

5∞ of human exon 4. Surprisingly this match (65%) overlaps the putative CCAAT and TATAAA RNA polymerase II promoter elements of the human exon 4 specific promoter ( Fig. 2). The reason for this unexpected result (divergence of the rat and human exon 4 promoter regions) can only be determined after obtaining the corresponding rat sequence. However, the present finding does provide circumstantial evidence suggesting that the rROM-K1 transcript may not utilize an additional upstream exon(s). The 199 bp exon 1 defines hROM-K2 (GenBank accession No. U12542) and it is entirely 5∞ UTR ( Table 1 and Fig. 2). The rROM-K2b transcript (GenBank accession Nos. S78153 and S78154) contains the orthologous exon 1 (Fig. 1C ), which in rat consists of at least 170 bp (the

1 29 38 29 38 1 12 38 24

5∞ end of the rat exon 1 has not been determined; Boim et al., 1995). These human and rat exon 1 sequences share 129/140 positions (76% identity), including a 13 bp 100% match immediately adjacent to the 3∞ exon/intron splice junction. This match does require the placement of several gaps (including one 17 bp gap) in the rat sequence to maximize the alignment and thus resulting in only a 140 bp comparison (gaps are counted as one mismatch regardless of length). The rROM-K3 transcript (GenBank accession No. S78155) apparently uses the same promoter as the rROM-K2 and -K2b transcripts, but it contains an alternatively spliced exon of 121 bp that potentially contains 65 bp of noncoding and 56 bp of coding DNA (Boim et al., 1995). Comparison of this rROM-K3

J.H. Bock et al. / Gene 188 (1997) 9–16

15

specific exon sequence against the hROM-K transcript sequences reveals no human ortholog. A search of the complete hKCNJ1 sequence with this 121 bp rat sequence reveals a match of 63% located between positions 15 994 and 16 117. This putative 121 bp hKCNJ1 exon ortholog is located about 1.3 kb 5∞ of the hKCNJ1 exon 3 (as indicated in Fig. 1C ). This region of the human gene could potentially be an exon as its flanking sequences conform to the consensus intron/exon splice acceptor sequence (5∞-CACCTCCTTTCTGAG/AAG) and donor sequence (GGA/GT-3∞) (Mount, 1982). However, if this putative hKCNJ1 exon is used it is most likely not utilized in the same manner as the 121 bp rat exon because its protein coding region does not include a coding frame that shares >10% identity with the deduced rat polypeptide. At present we have no evidence (Shuck et al., 1994) that this putative human 121 bp exon is alternatively spliced into any hROM-K transcripts and for this reason we suggest that it may be infrequently used or may be a pseudo-exon. 2.4. Searches for the location of repetitive DNA elements

Fig. 3. 5∞ extension analysis of hKCNJ1 exon 5. A primer extension reaction (Sambrook et al., 1989) was run using human kidney RNA and an exon 5 specific primer. DNA sequence size ladders ( lines 1–4, A, C, G and T reactions, respectively) were obtained by sequencing the 5HT gene ( Veldman and Bienkowski, 1992) using Sequenase 1Db ( US Biochemical, Cleveland, OH, USA). The primer extension reaction was run in lane 5. The arrow indicates the relative nucleotide position of the exon 5 primer extension product which corresponds to nucleotide position 28 066 of the hKCNJ1 sequence ( Fig. 2). The band located near the top of lane 5 is most likely the 5∞ primer extension product derived from the hROM-K1 transcript. The hROM-K1 5∞ extension product, containing exon 4 (Fig. 2), should be 87 bp longer than the hROM-K6 derived product which is consistent with the location of this second band. Methods: Total RNA was prepared from human kidney using Tri-Reagent (MRC, Cincinnati, OH, USA) as recommended by the manufacturer with one modification. Prior to use in the extension reaction, 75 mg of the RNA was re-precipitated with 2.5 M ammonium acetate and 3 volumes ethanol. The pellet was dried under vacuum. An exon 5 specific 21 bp antisense primer (5∞-GCACCTTCCATCTTTGGAGAC ) that would anneal at a location 110 bp downstream from the intron/exon splice junction was endlabeled using 5-fold molar excess [c-32P]ATP and polynucleotide kinase at 37°C for 30 min. The labeled primer was precipitated by the addition of 20 mg glycogen, sodium acetate to 0.3 M and 3 volumes ethanol. The pellet was washed extensively in 70% ethanol, dried, and resuspended in H O. The primer extension reactions contained 50 mg 2 of RNA, 4×106 dpm of the 32P-labeled primer in 50 ml of 1×reaction buffer (100 mM Tris (pH 8.3), 294 mM KCl, 1 mM EDTA, 2 mM DTT, 14 mM MgCl and 1 mM each dNTP). This mixture was heated 2 to 70°C for 3 min and then allowed to slowly cool to room temperature. The primer extension reaction was initiated by the addition of 5 ml

The Pythia program (Jurka et al., 1992) was used to search the hKCNJ1 sequence for repetitive DNA elements (SINEs, LINEs and MERs) and the results of this search are listed in Table 1. The most prominent SINE is the Alu element, and the Pythia analysis reveals the location of five complete (290 bp) and three partial Alu elements. Five of these Alu elements belong to the ‘older’ Alu-J subfamily and the remaining three belong to the ‘younger’ Alu-Sb or -Sx subfamilies ( Table 1) (Jurka and Milosavljevic, 1991). It is interesting to note that the two partial Alu-J elements located at nucleotide positions 19 850–20 031 and 20 335–20 462 were most likely formed as a result of the insertion of a younger Alu-Sb. On average, the distribution of these eight Alu elements across 40 kb (one every 5 kb) is below the expected density of one element every 4 kb or 0.25 element/kb (Hwu et al., 1986). This density is even lower (one every 5.7 kb) if we count the three Alus located between nucleotide positions 19 850 and 20 462 as two full-length elements. Thus, the Alu density of this region of chromosome 11 is nearly 1/2 the expected average. AMV-RT (Seikagaku America, Rockville, MD, USA) and incubating the sample at 42°C for 70 min. The RNA was degraded by the addition of 12.5 ml of 0.5 M NaOH and heating to 90°C for 3 min. The reaction was stopped by adding 237 ml of a stop buffer (25 mM HCl, 50 mM Tris (pH 7.4)) and 600 ml RNA-buffer saturated phenol/chloroform. After extraction the cDNA was precipitated using tRNA as carrier, washed with 70% ethanol and dried briefly under vacuum. The pellet was resuspended in 3 ml H O and 3 ml sequencing gel loading solution, 2 heated to 95°C for 2 min, and loaded on an 8% acrylamide 7 M urea sequence gel. After electrophoreses the gel was dried and exposed to BioMax XR film at −70°C for 1 h ( lanes 1–4) or 48 h ( lane 5).

16

J.H. Bock et al. / Gene 188 (1997) 9–16

The hKCNJ1 locus does not contain any full-length LINEs; in fact, the four LINE-related elements identified are very short, <200 bp in length (Table 1). The hKCNJ1 locus contains several DNA regions that share identities to a total of nine MER-type elements. However, it should be noted that six of these MER elements are clustered between nucleotide positions 1 and 677 ( Table 1). There are two possible explanations for this finding; firstly, this region may be a hotspot for the integration of MER-type elements, or secondly, the mapping of these MER elements to this region could be a coincidence of a shared high A+T composition (90% A+T for this region of the hKCNJ1 locus and >65% for MER 29 and 38). 2.5. Conclusions The complete nucleotide sequence of the hKCNJ1 gene has been determined and deposited in GenBank (accession No. U65406), which when combined with our additional 5∞ and 3∞ analyses of exon 5 should completely describe the organization of the hKCNJ1 gene. Our analysis of this genomic sequence has allowed us to determine the exact size of each exon and intron and to identify three potential hKCNJ1 promoter elements which are located in the 5∞ flanking DNAs of exons 1, 4 and 5 (Fig. 2). It is very possible that each of these promoters is subjected to tissue-specific expression signals, because in rat the distribution of specific rROM-K transcripts appears to be tissue specific (Boim et al., 1995). The nucleotide sequence of the hKCNJ1 locus provides detailed genetic information and material needed to pursue an even more detailed analysis of the mechanisms (multiple promoters and alternative splicing) and the tissue-specific expression of these hROMK transcripts.

References Boim, M.A., Ho., K., Shuck, M.E., Bienkowski, M.J., Bock, J.H., Slightom, J.L., Yang, Y., Tomasonni, S., Brenner, B.M. and Hebert,

S.C. (1995) ROMK inwardly rectifying ATP-sensitive K+ channel II. Cloning and distribution of alternatively forms. Am. J. Physiol. (Renal Fluid Electrolyte Physiol. 37) 268, F1132–F1140. Efstratiadis, A., Posakony, J.W., Maniatis, T., Lawn, R.M., O’Connell, C., Spritz, R.A., DeRiel, J.K., Forget, B., Weissman, S.M., Slightom, J.L., Blechl, A.E., Smithies, O., Barralle, F.E., Shoulders, C.C. and Proudfoot, J.J. (1980) The structure and evolution of the human b-globin gene family. Cell 21, 653–668. Ho, K., Nichols, C.G., Lederer, W.J., Lytton, J, Vassilev, P.M., Kanazirska, M.V. and Hebert, S.C. (1993) Cloning and expression of an inwardly rectifying ATP-regulated potassium channel. Nature 362, 31–38 Hwu, H.R., Roberts, J.W., Davidson, E.H. and Britten, R.J. (1986) Insertion and/or deletion of many repeated DNA sequences in human and higher ape evolution. Proc. Natl. Acad. Sci. USA 83, 3875–3879. Jurka, J. and Milosavljevic, A. (1991) Reconstruction and analysis of human Alu Genes. J. Mol. Evol. 32, 105–121. Jurka, J., Walichiewicz, J. and Milosavljevic, A. (1992) Prototypic sequences for human repetitive DNA. J. Mol. Evol. 35, 286–291. Koop, B.F. and Hood, L. (1994) Striking sequence similarity over almost 100 kilobases of human and mouse T-cell receptor DNA. Nature Genet. 7, 48–53. Mount, S.M. (1982) A catalogue of splice junction sequences. Nucleic Acids Res. 10, 459–472. Sambrook, J. Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Shuck, M.E., Bock, J.H., Benjamin, C.W., Tsai, T.D., Lee, K.S., Slightom, J.L. and Bienkowski, M.J. (1994) Cloning and characterization of multiple forms of the human kidney ROM-K potassium channel. J. Biol. Chem. 269, 24261–24270. Siemieniak, D.R., Sieu,L.C. and Slightom, J.L. (1991) Strategy and methods for directly sequencing cosmid clones. Anal. Biochem. 192, 441–448. Slightom, J.L., Siemieniak, D.R., Sieu, L.C., Koop, B.F. and Hood, L. (1994) Nucleotide sequence analysis of 77.7 kb of the human V T-cell receptor gene locus, Genomics 20, 149–168. b Slightom, J.L. and Sieu, L.C. (1992) Direct sequencing of baculovirus genomic DNA: sequence determination of the engineered Respiratory Syncytial Virus chimeric FG gene. BioTechniques 13, 94–105. Veldman, S.A. and Bienkowski, M.J. (1992) Cloning and pharmacological characterization of a novel human 5-hydroxytryptamine 1D receptor subtype. Mol. Pharmacol. 42, 439–444. Yano, H., Philipson, L.H., Kugler, J.L., Tokuyama, Y., Davis, E.M., Le Beau, M.M., Nelson, D.J., Bell, G.I. and Takeda, J. (1994) Alternative splicing of human inwardly rectifying K+ channel ROMK1 mRNA. Mol. Pharmacol. 45, 854–860. Zhou, H., Tate, S.S. and Palmer, L.G. (1994) Primary structure and functional properties of an epithelial K channel. Am. J. Physiol. 266 (Cell Physiol. 35), C809–C824.