cDNA cloning and sequencing

A cDNA library was constructed from adult chicken breast muscle poly(A)⁺ RNA. A human skeletal muscle connectin cDNA clone (clone 5) was kindly donated by Dr. H. Sorimachi and Professor K. Suzuki of the University of Tokyo (cf. Fig. 1A of Sorimachi et al., 1995). The clone 5 was digested with EcoRI to produce 2.7 kb cDNA encoding the very C terminal region of human skeletal muscle connectin. The probe for library screening was obtained by labeling this 2.7 kb cDNA with DIG-dUTP using DIG DNA Labeling Kit (Boehringer Mannheim, Germany). The screening of the library and the analyses of the clone were performed according to the manufacturer's instructions. Sequencing was carried out by dideoxy nucleotide chain-termination method using BcaBEST Sequencing Kit (TAKARA, Japan).

Fig. 1.

Nucleotide and predicted amino acid sequences of the 3∣ terminal region of chicken skeletal muscle cDNA (CC4).

Fig. 1.

Continued.

Northern blotting

Northern blotting was carried out as described before (Endo and Nadal-Ginard, 1987). RNAs were prepared from various tissues of 12 day-old embryos and adults of chicken as described previously (Endo and Nadal-Ginard, 1987). The probes used were the whole length of Cn1 cDNA (3.33 kb) (Maruyama et al., 1994), the 5∣ Hincll fragment (1.87 kb), and the 3∣ Hincll fragment (0.82 kb).

The predicted amino acid sequence of C terminal region of connectin

A cDNA clone of 3660 bp, CC4, obtained using the human clone (Sorimachi et al., 1995) as probe, was predicted to have a coding region of 3065 bp, and to encode 1021 amino acids ending with the stop codon (Fig. 1). There are six motif II (immunoglobulin C2) domains with five interdomains (Fig. 2). They have 70–75% homology with the C terminal region of human cardiac connectin (Gautel et al., 1993). In the interdomain between the third and fourth motif lls there are 13 amino acids in chicken connectin, whereas there are 181 amino acids including one motif II in human connectin (Fig. 2). Thus 168 amino acids are missing in chicken connectin.

Fig. 2.

Comparison of the domains of chicken skeletal muscle and human cardiac connectins at the C terminal region. Chicken skeletal muscle (this paper); human cardiac muscle (Labeit and Kolmerer, 1995). KSP, KSP repeat region.

In the interdomain between the second and third motif II, there are four KSP regions that are very similar to those of human connectin (Fig. 3a). Only one out of 22 amino acids is different between the two species (proline in chicken and alanine in human). In the interdomain between the fifth and sixth motif lls there is serine rich region: 30 serine residues are found in 95 amino acids (Fig. 3b). There are two repeats of 4 serine residues, five repeats of 2 serine residues and twelve serine residues. This serine rich region is also similar to that of human cardiac connectin. There are only two serines more in human connectin.

Fig. 3.

Comparison of the amino acid sequences of chicken skeletal muscle and human cardiac connectins at the KSP and serine rich regions. Chicken skeletal muscle (this paper); human cardiac muscle (Labeit and Kolmerer, 1995). a, KSP region; 72.4% homology for 58 amino acids, b, serine rich region; 77.6% homology for 98 amino acids.

Cn1 is not the 3∣ terminal region of connectin cDNA

In the previous paper (Maruyama et al., 1994), we described that Cn1, a clone screened with an antibody to connectin (Pc 1200), hybridized to an mRNA of chicken muscle with a mobility beyond the upper limit of resolution (∼ 25 kb). However, it also cross-hybridized to 2.0 and 1.5 kb mRNA, suggesting the possibility that Cn1 contained another gene. The amino acid sequence at the C terminal region, 775 to 805, had a homology to that of porin, voltage dependent channel protein (Kayser et al, 1989).

Reexamination of the sequence of Cn1 showed that N terminal 774 amino acids are derived from connectin at the N terminal region not at the C terminal region (Yajima et al, unpublished data). The C terminal 37 amino acids come from porin. In the present study, the 175 amino acid sequence of porin starting the 108th amino acid up to its C terminus was determined as shown in Figure 4. This new sequence had 74% homology with that of human porin.

Evidence by Northern blotting

When Northern blotting was carried out by using a probe of the whole length of Cn1 cDNA, two distinctive sizes of mRNAs were detected in embryonic and adult tissues (Fig. 5A). One of the mRNAs had a mobility beyond the upper limit of resolution (∼25 kb) and was specific for skeletal and cardiac muscles as reported previously (Maruyama et al., 1994). According to its extremely large size and tissue specificity, the hybridization band represents connectin mRNA.

In contrast, another mRNA was ∼1.5 kb and present in all the tissues examined. To address whether or not the band was generated by the hybridization to the porin-like sequence, Northern blotting was carried out by use of the 5∣ Hincll fragment (1.87 kb) and the 3∣ Hincll fragment (0.82 kb) of the cDNA. The former fragment did not contain the porin-like sequence but the latter contained exclusively the sequence. The 5∣ fragment hybridized only to the mRNA with an apparent size of ∼25 kb in striated muscles (Fig. 5B), whereas the 3∼ fragment did exclusively to the ubiquitous ∼1.5 kb mRNA. These results strongly suggest that Cn1 cDNA is a ligated form of connectin and porin-like sequences. These two cDNAs were likely to be ligated into one λgt11 vector during the course of cDNA library construction (Fig. 4).

Fig. 4.

The structure of Cn1, a connectin-porin ligated cDNA. Upper: Maruyama et al., 1994 ; Lower: this paper. VDAC, voltage dependent channel protein. 5∣ Terminal region of connectin cDNA was ligated with a portion of porin cDNA.

Fig. 5.

Northern blotting of RNAs from various tissues probed with Cn1 cDNA. (A) Northern blotting probed with the whole length of Cn1 cDNA. (B) Blotting probed with the 5∣ Hincll fragment of the cDNA. (C) Blotting probed with the 3∣ Hincll fragment. (D) Ethidium bromide staining pattern showing 28S and 18S rRNAs. Ten micrograms of the following total RNAs were loaded to each lane: lane 1, embryonic skeletal muscle; 2, embryonic cardiac muscle; 3, embryonic gizzard smooth muscle; 4, adult skeletal muscle; 5, adult cardiac muscle; 6, adult gizzard smooth muscle; 7, adult brain; 8, adult liver. The positions of connectin and porin-like mRNAs are indicated by Cn and Pr, respectively. The positions of 28S and 18S rRNAs are denoted by 28 and 18, respectively.