Abnormal glycosylation and altered Golgi structure in colorectal cancer: dependence on intra-Golgi pH

Abnormal glycosylation of cellular glycoconjugates is a common phenotypic change in many human tumors. Here, we explore the possibility that an altered Golgi pH may also be responsible for these cancer-associated glycosylation abnormalities. We show that a mere dissipation of the acidic Golgi pH results both in increased expression of some cancer-associated carbohydrate antigens and in structural disorganization of the Golgi apparatus in otherwise normally glycosylating cells. pH dependence of these alterations was confirmed by showing that an acidification-defective breast cancer cell line (MCF-7) also displayed a fragmented Golgi apparatus, whereas the Golgi apparatus was structurally normal in its acidification-competent subline (MCF-7/AdrR). Acidification competence was also found to rescue normal glycosylation potential in MCF-7/AdrR cells. Finally, we show that abnormal glycosylation is also accompanied by similar structural disorganization and fragmentation of the Golgi apparatus in colorectal cancer cells in vitro and in vivo. These results suggest that an inappropriate Golgi pH may indeed be responsible for the abnormal Golgi structure and lowered glycosylation potential of the Golgi apparatus in malignant cells.     

Regulation of the pH sensitivity of human P2X receptors by N-linked glycosylation

The human (h) P2X(3) receptor and its mutants deficient in one out of four N-glycosylation sites were expressed in HEK293 cells. Concentration-response curves were generated by whole-cell recordings of alpha,beta-methylene ATP (alpha,beta-meATP)-induced currents. A gradual change of external pH from the alkaline 8.0 to the acidic 5.0 successively decreased the maximum current amplitude (E(max)) without affecting the EC(50) value. The replacement of Asn-139 and -170 by Asp (N139D, N170D) abolished the pH sensitivity of the wild-type (WT) hP2X(3) receptor. In the case of N194D, the E(max) was again the highest at the alkaline pH value with no change from 7.4 to 6.5, whereas in the case of N290D, there was an inverse pH sensitivity, with an increase of E(max) in the acidic range. However, this effect appeared to be due to enhanced protonation by the insertion of Asp into the receptor, because replacement of Asn by the neutral Thr resulted in a comparable potency of alpha,beta-meATP at any of the pH values investigated. In accordance with the reported finding that His-206 is involved in the modulation of WT P2X(3) receptors by protons, we showed that the normal change of E(max) by an acidic, but not alkaline pH was abolished after substitution of this His by Ala. However, the double mutant H206A + N290D did not react to acidification or alkalinization with any change in E(max). In conclusion, only fully N-glycosylated P2X(3) receptors recognize external pH with a modified sensitivity towards alpha,beta-meATP.     

Photohemolysis sensitized by psoralen: dependence on pH

The effect of pH on the hemolysis of erythrocytes photosensitized (366 nm, 23 Wt/m2) by psoralen has been studied. The dependence of the photohemolysis rate (V) on irradiation dose (D) was described by the equation V = Vo + kD, where Vo is the rate of hemolysis without irradiation (dark), and k is the constant. The index of the power at dose x was approximately equal to 2, and its value did not change as the pH of the erythrocyte suspension was changed. It was found that changes in pH led to a sharp change in the value of coefficient k and correspondingly V. The lowest rate of photohemolysis was observed in the pH range from 8.0 to 8.4. As pH was changed from 3.4 to 9.0 or from 8.0 to 7.4, the V value increased approximately twofold. At pH below 7.4, an abrupt increase (approximately fourfold) in V was observed, with the pK value being equal to 7.3. The psoralen molecule lacks titratable acidic and basic groups; therefore, the effects of pH can hardly be assigned to changes in the photophysical properties of the sensitizer. The increase in V in the alkaline region is prohably related to the acceleration of photooxidation of reduced glutathione, whereas the jump of V at pH of about 7.3 may be due to the titration of the product of psoralen photooxidation. The latter assumption is confirmed by the data of hign performance liquid chromatography. In these experiments, psoralen was oxidized in ethanol and mixed with the phosphate buffer at different pH values followed by a qualitative and quantitative analysis by high performance liquid chromatography of photoproducts. Several photoproducts of psoralen have been identified whose content depended on pH. The curve of titration of one photoproduct was similar in shape to the pH dependence of psoralen-photosensitized hemolysis.     

Regulation of Notch signaling by glycosylation

Notch receptors are approximately 300 kDa cell surface glycoproteins whose activation by Notch ligands regulates cell fate decisions in the metazoa. The extracellular domain of Notch receptors has many epidermal growth factor like repeats that are glycosylated with O-fucose and O-glucose glycans as well as N-glycans. Disruption of O-fucose glycan synthesis leads to severe Notch signaling defects in Drosophila and mammals. Removal or addition of O-fucose glycan consensus sites on Notch receptors also leads to Notch signaling defects. Ligand binding and ligand-induced Notch signaling assays have provided insights into how changes in the O-fucose glycans of Notch receptors alter Notch signaling.     

Regulation of inward rectifier K+ channels by shift of intracellular pH dependence

The mechanistic link between mitochondrial metabolism and inward rectifier K+ channel activity was investigated by studying the effects of a mitochondrial inhibitor, carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) on inward rectifiers of the Kir2 subfamily expressed in Xenopus oocytes, using two-electrode voltage-clamp, patch-clamp, and intracellular pH recording. FCCP inhibited Kir2.2 and Kir2.3 currents and decreased intracellular pH, but the pH change was too small to account for the inhibitory effect by itself. However, pre-incubation of oocytes with imidazole prevented both the pH decrease and the inhibition of Kir2.2 and Kir2.3 currents by FCCP. The pH dependence of Kir2.2 was shifted to higher pH in membrane patches from FCCP-treated oocytes compared to control oocytes. Therefore, the inhibition of Kir2.2 by FCCP may involve a combination of intracellular acidification and a shift in the intracellular pH dependence of these channels. To investigate the sensitivity of heteromeric channels to FCCP, we studied its effect on currents expressed by heteromeric tandem dimer constructs. While Kir2.1 homomeric channels were insensitive to FCCP, both Kir2.1-Kir2.2 and Kir2.1-Kir2.3 heterotetrameric channels were inhibited. These data support the notion that mitochondrial dysfunction causes inhibition of heteromeric inward rectifier K+ channels. The reduction of inward rectifier K+ channel activity observed in heart failure and ischemia may result from the mitochondrial dysfunction that occurs in these conditions.     

Regulation of TRP channels by N-linked glycosylation

A subset of TRP channel proteins undergoes regulatory N-linked glycosylation. A glycosylation site in the first extracellular loop of TRPV5 is enzymatically cleaved by a secreted glucuronidase, indirectly regulating channel function. Members of the TRPC family share a similar site, although details about a regulatory role are lacking. A second conserved TRP channel glycosylation site is found immediately adjacent to the channel pore-forming loop; both TRPV1 and TRPV4--and perhaps other TRPV family members--are influenced by glycosylation at this site. N-linked glycosylation, and the dynamic regulation of this process, substantially impacts function and targeting of TRP channels.     

Regulation of (Na++K+)-ATPase by inorganic phosphate: pH dependence and physiological implications

Inhibition of Na++K+-dependent ATPase activity by Pi was maximal in the pH range of 6.1-7, but decreased with increasing pH in the range of 7-8.5. Ki of Pi was 2.8 mM at pH 7.1, and 12 mM at pH 7.8. K+-dependent phosphorylation of the enzyme by Pi, which is thought to be responsible for inhibition of ATPase activity, also decreased with increasing pH. The data suggest that (a) previously observed requirement of high Pi concentrations for inhibition of ATPase activity and associated pump fluxes may have been due to high pH of the assays; (b) at normal values of intracellular pH the pump may be partially inhibited by intracellular Pi; and (c) this effect of Pi may be amplified or dampened with alterations in intracellular pH and ATP/Pi ratio.     

Regulation of dolichol-linked glycosylation

In the majority of congenital disorders of glycosylation, the assembly of the glycan precursor GlcNAc₂Man₉Glc₃ on the polyprenol carrier dolichyl-pyrophosphate is compromised. Because N-linked glycosylation is essential to life, most types of congenital disorders of glycosylation represent partial losses of enzymatic activity. Consequently, increased availability of substrates along the glycosylation pathway can be beneficial to increase product formation by the compromised enzymes. Recently, we showed that increased dolichol availability and improved N-linked glycosylation can be achieved by inhibition of squalene biosynthesis. This review summarizes the current knowledge on the biosynthesis of dolichol-linked glycans with respect to deficiencies in N-linked glycosylation. Additionally, perspectives on therapeutic treatments targeting dolichol and dolichol-linked glycan biosynthesis are examined.     

Regulation of signal transduction pathways in development by glycosylation

Recent studies from several laboratories have provided evidence that cell surface complex carbohydrates play key roles in the regulation of developmentally relevant signal transduction events. The demonstration that Fringe, a known modifier of Notch function, is a fucose-specific N-acetylglucosaminyltransferase provided strong evidence that the Notch signaling pathway could be regulated by alterations of O-fucose structures. More recently, the demonstration that O-fucose modification of Cripto is essential for Nodal-dependent signaling provides further evidence of a role for glycosylation in signal transduction. These and other examples provide a new paradigm for the regulation of signal transduction events by glycosylation.     

Paramagnetism in melanins: pH dependence

Changes in the paramagnetic properties of aqueous suspensions of melanin polymers have been monitored over a pH range from 1 to 12. Distinct changes in spin concentration and electron spin resonance spectral parameters (effective g value and line shape) are shown to occur. These data are interpreted in terms of pH- and temperaturedependent equilibria between diamagnetic and paramagnetic units on the melanin polymer, which can be partly or completely quenched if the melanin is precipitated by lowering the pH or by increasing the salt concentration. The heterogeneity of these units and possible chemical structures are discussed.     

Regulation of mTORC1 signaling by pH

Background

Acidification of the cytoplasm and the extracellular environment is associated with many physiological and pathological conditions, such as intense exercise, hypoxia and tumourigenesis. Acidification affects important cellular functions including protein synthesis, growth, and proliferation. Many of these vital functions are controlled by mTORC1, a master regulator protein kinase that is activated by various growth-stimulating signals and inactivated by starvation conditions. Whether mTORC1 can also respond to changes in extracellular or cytoplasmic pH and play a role in limiting anabolic processes in acidic conditions is not known.

Methodology/Findings

We examined the effects of acidifying the extracellular medium from pH 7.4 to 6.4 on human breast carcinoma MCF-7 cells and immortalized mouse embryo fibroblasts. Decreasing the extracellular pH caused intracellular acidification and rapid, graded and reversible inhibition of mTORC1, assessed by measuring the phosphorylation of the mTORC1 substrate S6K. Fibroblasts deleted of the tuberous sclerosis complex TSC2 gene, a major negative regulator of mTORC1, were unable to inhibit mTORC1 in acidic extracellular conditions, showing that the TSC1–TSC2 complex is required for this response. Examination of the major upstream pathways converging on the TSC1–TSC2 complex showed that Akt signaling was unaffected by pH but that the Raf/MEK/ERK pathway was inhibited. Inhibition of MEK with drugs caused only modest mTORC1 inhibition, implying that other unidentified pathways also play major roles.

Conclusions

This study reveals a novel role for the TSC1/TSC2 complex and mTORC1 in sensing variations in ambient pH. As a common feature of low tissue perfusion, low glucose availability and high energy expenditure, acidic pH may serve as a signal for mTORC1 to downregulate energy-consuming anabolic processes such as protein synthesis as an adaptive response to metabolically stressful conditions.     

The pH dependence of peptide hydrolysis by nitrocarboxypeptidase A

Hydrolysis of benzyloxycarbonyl-GlyGlyPhe by nitro(Tyr 248)carboxypeptidase A over the pH range 4.88–8.04 has been examined. The nitroenzyme retains appreciable activity near pH 6.5, and the limiting value of K_m is scarcely affected. The peptidase activity has a pH dependence characterized by the following parameters: pK_E1 of 6.37 ± 0.19 and pK_E2 of 6.60 ± 0.17 in $\text{k}{}_{\mathrm{<mtext>cat</mtext>}}\text{K}{}_{\mathrm{m}}$, and apparent pK of 5.59 ± 0.06 in K_cat. A spectroscopic pK of 6.75 ± 0.01, attributable to the nitro-Tyr 248 residue, has been determined. This correlates with the base-limb pK_E2 in the $\text{k}{}_{\mathrm{<mtext>cat</mtext>}}\text{K}{}_{\mathrm{m}}$ profile, which appears to be shifted from a higher value, pK_E2 of 9.0, for the native enzyme. The single (acid-limb) pK which characterizes the k_cat profile of the native enzyme is also found to be perturbed to a lesser extent by nitration. A kinetically competent reverse protonation mechanism, based on chemical modification and crystallographic evidence for the enzyme, is described.     

pH dependence of proton translocation by Escherichia coli.

Proton translocation in spheroplasts from Escherichia coli has been studied in two mutants, one of which expresses cytochrome o and the other cytochrome d as the terminal oxidase. Using the O2 pulse method, the H+/e- ratio of proton translocation associated with cytochrome o was confirmed to be near 2 at neutral pH, but was found to decrease considerably when the medium pH was raised above 8. At high pH there was an increase in H+/OH- permeability of the cell membrane, but this was not sufficient to explain the decline in proton ejection. The pH effect was confined to cytochrome o-linked activity. It was not present when cytochrome d generated the electrochemical proton gradient. This makes it improbable that the Na+/H+ antiporter is responsible. The most likely explanation for our finding is that there is a "slip" in the proton-pumping mechanism of cytochrome o at high pH.     

Galactose mutarotase: pH dependence of enzymatic mutarotation

Here we report pH dependence of kinetic parameters for the mutarotation of alpha-D-glucose catalyzed by galactose mutarotase (GalM) from Escherichia coli. The values of k(cat) and k(cat)/K(m) for the mutarotation of alpha-D-galactose were found to be 1.84 x 10(4) s(-1) and 4.6 x 10(6) M(-1) s(-1), respectively, at pH 7.0 and 27 degrees C. The corresponding values for alpha-D-glucose were 1.9 x 10(4) s(-1) and 5.0 x 10(5) M(-1) s(-1). Inasmuch as the value of k(cat)/K(m) for the reaction of alpha-D-galactose is 10 times that for alpha-D-glucose, and the diffusional rate constants should be essentially the same for the two sugars, the mutarotation of alpha-D-glucose should not be diffusion controlled. Therefore, pH-rate profiles should not be distorted by diffusion. The k(cat) for the mutarotation of alpha-D-glucose is independent of pH. Therefore, either the enzyme-substrate complexes do not undergo ionization of catalytic groups, or the rate-limiting step is neither mutarotation nor diffusion. The profile of log k(cat)/K(m) versus pH is a distorted bell-shaped curve, with slopes of +1 on the acid side and -2 on the alkaline side. The values of pK(a) are 6.0 and 7.5, and mutarotation depends on the ionization states of three functional groups in the free enzyme, one unprotonated and two protonated. On the acid side, ring opening of alpha-D-glucose limits the rate, and on the alkaline side, ring closure of the open-chain sugar limits the rate. A mutarotation mechanism is presented in which one of the catalytic groups shuttles a proton to and from the endocyclic oxygen and the other two shuttle protons to the anomeric oxygen atoms. In this mechanism, three catalytic groups overcome the problem of nonstereospecificity in mutarotation. The groups are postulated to be His 104, His 175, and Glu 309. Mutations of these residues grossly impair catalytic activity. Variants H104Q- and E309Q-GalM display sufficient activity to allow profiles of log k(cat)/K(m) versus pH to be constructed. Both profiles show breaks on the acid side corresponding to pK(a) values of 5.8 for H104Q and 6.3 for E309Q. Apparently, ring opening of alpha-D-glucose limits the rate at low pHs, but ring closure does not become rate limiting at pHs up to 8.5 in reactions of these variants.