PH regulation of connexin43: molecular analysis of the gating particle.

Gap junction channels allow for the passage of ions and small molecules between neighboring cells. These channels are formed by multimers of an integral membrane protein named connexin. In the heart and other tissues, the most abundant connexin is a 43-kDa, 382-amino acid protein termed connexin43 (Cx43). A characteristic property of connexin channels is that they close upon acidification of the intracellular space. Previous studies have shown that truncation of the carboxyl terminal of Cx43 impairs pH sensitivity. In the present study, we have used a combination of optical, electrophysiological, and molecular biological techniques and the oocyte expression system to further localize the regions of the carboxyl terminal that are involved in pH regulation of Cx43 channels. Our results show that regions 261-300 and 374-382 are essential components of a pH-dependent "gating particle," which is responsible for acidification-induced uncoupling of Cx43-expressing cells. Regions 261-300 and 374-382 seem to be interdependent. The function of region 261-300 may be related to the presence of a poly-proline repeat between amino acids 274 and 285. Furthermore, site-directed mutagenesis studies show that the function of region 374-382 is not directly related to its net balance of charges, although mutation of only one amino acid (aspartate 379) for asparagine impairs pH sensitivity to the same extent as truncation of the carboxyl terminal domain (from amino acid 257). The mutation in which serine 364 is substituted for proline, which has been associated with some cases of cardiac congenital malformations in humans, also disrupts the pH gating of Cx43, although deletion of amino acids 364-373 has no effect on acidification-induced uncoupling. These results provide new insight into the molecular mechanisms responsible for acidification-induced uncoupling of gap junction channels in the heart and in other Cx43-expressing structures.     

Intramolecular interactions mediate pH regulation of connexin43 channels.

We have previously proposed that acidification-induced regulation of the cardiac gap junction protein connexin43 (Cx43) may be modeled as a particle-receptor interaction between two separate domains of Cx43: the carboxyl terminal (acting as a particle), and a region including histidine 95 (acting as a receptor). Accordingly, intracellular acidification would lead to particle-receptor binding, thus closing the channel. A premise of the model is that the particle can bind its receptor, even if the particle is not covalently bound to the rest of the protein. The latter hypothesis was tested in antisense-injected Xenopus oocyte pairs coexpressing mRNA for a pH-insensitive Cx43 mutant truncated at amino acid 257 (i.e., M257) and mRNA coding for the carboxyl terminal region (residues 259-382). Intracellular pH (pHo) was recorded using the dextran form of the proton-sensitive dye seminaphthorhodafluor (SNARF). Junctional conductance (Gj) was measured with the dual voltage clamp technique. Wild-type Cx43 channels showed their characteristic pH sensitivity. M257 channels were not pH sensitive (pHo tested: 7.2 to 6.4). However, pH sensitivity was restored when the pH-insensitive channel (M257) was coexpressed with mRNA coding for the carboxyl terminal. Furthermore, coexpression of the carboxyl terminal of Cx43 enhanced the pH sensitivity of an otherwise less pH-sensitive connexin (Cx32). These data are consistent with a model of intramolecular interactions in which the carboxyl terminal acts as an independent domain that, under the appropriate conditions, binds to a separate region of the protein and closes the channel. These interactions may be direct (as in the ball-and-chain mechanism of voltage-dependent gating of potassium channels) or mediated through an intermediary molecule. The data further suggest that the region of Cx43 that acts as a receptor for the particle is conserved among connexins. A similar molecular mechanism may mediate chemical regulation of other channel proteins.     

Evidence for the Presence of a Free C-Terminal Fragment of Cx43 in Cultured Cells

Migration of the gap junction protein connexin 43 (Cx43) in SDS-PAGE yields 2 to 4 distinct bands, detectable in the 40-47 kDa range. Here, we show that antibodies against the carboxy-terminal domain of Cx43 recognized an additional 20-kDa product. This protein was detected in some culture cell lysates. The presence of the 20-kDa band was not prevented by the use of protease inhibitors (Complete® and phenylmethylsulfonyl fluoride (PMSF), 1-5 mM). The band was absent from cells treated with Cx43-specific RNAi, and from those derived from Cx43-deficient mice, indicating that this Cx43-immunoreactive protein is a product of the Cx43 gene. Treatment of CHO cells with cyclosporin A caused a reduction in the amount of full-length Cx43 and a concomitant increase in the amount of the 20-kDa band. Overall, our data show that a fraction of the Cx43-immunoreactive protein pool within a given cell may correspond to a C-terminal fragment of the protein.     

Retinal inhibits TPA activated, calcium-dependent, phospholipid-dependent protein kinase ("C" kinase)

RAW264 macrophage-like cells contain a kinase which is dependent on Ca++ and phosphatidylserine for activity (C kinase) and is stimulable by the tumor promoter 12-0-tetradecanoyl-phorbol-13-acetate. Retinal inhibits the activity of the tumor promoter-activated kinase in a concentration-dependent manner. The apparent Ki for inhibition is 1 X 10(-5)M. Retinal is not a general inhibitor of phosphotransferase reactions as it did not inhibit the activity of purified cyclic AMP-dependent protein kinase. It is possible, therefore, that the action of retinoids to antagonize tumor promoter effects on cell function may be mediated at the level of regulation of C-kinase.     

The calcium uptake of the rat heart sarcoplasmic reticulum is altered by dietary lipid

Summary Small amounts of dietary n-3 fatty acids can have dramatic physiological effects, including the reduction of plasma triglycerides and an elevation of cellular eicosapentanoic (EPA) and docosahexanoic acids (DHA) at the expense of arachidonic acid (AA). We investigated the effects of alterations in the fatty acid compositions of cardiac sarcoplasmic reticulum (CSR) produced by dietary manipulation on the calcium pump protein that is required for energy dependent calcium transport. CSR was isolated from rats fed menhaden oil, which is rich in n-3 fatty acids, and from control animals that were given corn oil. Relative to control membranes, those isolated from rats fed menhaden oil, had a lower content of saturated phospholipids, an increased DHA/AA ratio, and an increased ratio of n-3 to n-6 fatty acids. These changes were associated with a 30% decrease in oxalate-facilitated, ATP-dependent calcium uptake and concomitant decreased Ca-ATPase activity in the membranes from the animals fed menhaden oil. In contrast, there was no alteration in active pump sites as measured by phosphoenzyme formation. Thus, the CSR Ca-ATPase function can be altered by dietary interventions that change the composition, and possibly structure, of the phospholipid membranes thereby affecting enzyme turnover.     

Connexin diversity and gap junction regulation by pHi

The molecular mechanisms controlling pH-sensitivity of gap junctions formed of two different connexins are yet to be determined. We used a proton-sensitive fluorophore and electrophysiological techniques to correlate changes in intracellular pH (pHi) with electrical coupling between connexin-expressing Xenopus oocytes. The pH sensitivities of alpha 3 (connexin46), alpha 2 (connexin38), and alpha 1 (connexin43) were studied when these proteins were expressed as: 1) nonjunctional hemichannels (for alpha 3 and alpha 2), 2) homotypic gap junctions, and 3) heterotypic gap junctions. We found that alpha 3 hemichannels are sensitive to changes in pHi within a physiological range (pKa = 7.13 +/- 0.03; Hill coefficient = 3.25 +/- 1.73; n = 8; mean +/- SEM); an even more alkaline pKa was obtained for alpha 2 hemichannels (pKa = 7.50 +/- 0.03; Hill coefficient = 3.22 +/- 0.66; n = 13). The pH sensitivity curves of alpha 2 and alpha 3 homotypic junctions were indistinguishable from those recorded from hemichannels of the same connexin. Based on a comparison of pKa values, both alpha 3 and alpha 2 gap junctions were more pHi-dependent than alpha 1. The pH sensitivity of alpha 2-containing heterotypic junctions could not be predicted from the behavior of the two connexons in the pair. When alpha 2 was paired with alpha 3, the pH sensitivity curve was similar to that obtained from alpha 2 homotypic pairs. Yet, pairing alpha 2 with alpha 1 shifted the curve similar to homotypic alpha 1 channels. Pairing alpha 2 with a less pH sensitive mutant of alpha 1 (M257) yielded the same curve as when alpha 1 was used. However, the pH sensitivity curve of alpha 3/alpha 1 channels was similar to alpha 3/alpha 3, while alpha 3/M257 was indistinguishable from alpha 3/alpha 1. Our results could not be consistently predicted by a probabilistic model of two independent gates in series. The data show that dissimilarities in the pH regulation of gap junctions are due to differences in the primary sequence of connexins. Moreover, we found that pH regulation is an intrinsic property of the hemichannels, but pH sensitivity is modified by the interactions between connexons. These interactions should provide a higher level of functional diversity to gap junctions that are formed by more than one connexin.     

Measurement of aortic input impedance in mice: effects of age on aortic stiffness

Mice are used with increasing frequency as models of human cardiovascular diseases, but significant gaps exist in our knowledge of vascular function in the aging mouse. We determined aortic input impedance spectra, pulse wave velocity, and augmentation index in adult (8-mo-old) and old (29-mo-old) mice to determine whether arterial stiffening occurred with age in mice as it does in humans. Pressure and blood velocity signals measured simultaneously from the same location in the ascending aorta were used to determine input impedance spectra (0-10 harmonics). The first minimum of the impedance modulus occurred at the second harmonic in adult mice but shifted to the fourth harmonic in old mice. Characteristic impedance (average of 2nd-10th harmonic) was 57% higher in old mice: 471 +/- 62 vs. 299 +/- 10 (SE) dyn.s.cm-3 (P < 0.05). Pulse pressure and augmentation index, determined from the aortic pressure signals, were also higher in old mice: 42 +/- 2.2 vs. 29 +/- 4.9 mmHg (P < 0.05) and 37 +/- 5 vs. 14 +/- 2% (P < 0.005). Aortic pulse wave velocity measured from the timing of upstrokes of the Doppler velocity signals was 45% higher in old mice: 416 +/- 22 vs. 286 +/- 14 cm/s (n = 3, P < 0.01). These results reproduce age-related findings reported in humans and confirm that mice may be used as models of age-related vascular stiffening.