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


     

(Investigative Ophthalmology and Visual Science. 2003;44:4622-4629.)
© 2003 by The Association for Research in Vision and Ophthalmology, Inc.
DOI:  10.1167/iovs.03-0420
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Google Scholar
Google Scholar
Right arrow Articles by Chen, F.
Right arrow Articles by Lee, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, F.
Right arrow Articles by Lee, R. H.

Cone Photoreceptor ß{gamma}-Transducin: Posttranslational Modification and Interaction with Phosducin

Fayu Chen,1 Pei-San Ng,2 Kym F. Faull,3 and Rehwa H. Lee1,2

1From the Jules Stein Eye Institute, UCLA School of Medicine, Los Angeles, California; the 2Molecular Neurology Laboratory, VA Greater Los Angeles Healthcare System at Sepulveda, Sepulveda, California; the 3Pasaraw Mass Spectrometry Laboratory, Department of Chemistry, Biochemistry, Psychiatry and Behavioral Science, and The Neuropsychiatry Institute, UCLA, Los Angeles, California.


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
PURPOSE. To characterize the structure of cone ß{gamma}-transducin (Tß3{gamma}8) and its interaction with phosducin (pdc).

METHODS. The T{gamma}8 subunit of Tß3{gamma}8 was isolated by column chromatography for peptide mapping with mass spectrometry. Tß3{gamma}8 was compared with rod ß{gamma}-transducin (Tß1{gamma}1) in terms of the electrophoretic mobility, pdc binding affinity, and the effects of phosphorylation and methylation, and then the correlation to the crystal structures and functional domains of Tß1{gamma}1 was determined.

RESULTS. The mature T{gamma}8 is a 65-amino-acid peptide encoded by the G{gamma}8 gene with an acetylated and a farnesylated–methylated N- and C-terminus, respectively. Purified Tß3{gamma}8 is similar to Tß1{gamma}1 in that (1) both are heterogeneous, containing methylated and demethylated T{gamma} subunits; (2) each demethylated dimer migrates faster than its methylated counterpart during native gel electrophoresis, and the methylation-associated mobility differential is masked by pdc binding; and (3) both dimers bind pdc with the same affinity, and the affinity is reduced threefold by PKA phosphorylation of pdc and twofold by demethylation at the C-terminus of T{gamma}. Tß3{gamma}8 differs from Tß1{gamma}1 in exhibiting lower intrinsic electrophoretic mobility, and the difference is unaffected by either pdc binding or the status of T{gamma} methylation.

CONCLUSIONS. Tß3{gamma}8 is identical with Tß1{gamma}1 in T{gamma} isoprenylation, the spatial organization, and the mode of pdc binding, indicating that its interaction with pdc does not play an important role in the specialization of cones. Changes in Tß{gamma} characteristics by T{gamma} methylation reveal conformational changes on a surface domain that is essential for Tß{gamma} functions and support a regulatory role for reversible methylation.

The family of large G proteins plays an essential role in transducing extracellular signals from cell-surface receptors to intracellular effectors.1 2 Members of this family are heterotrimers composed of G{alpha}, Gß, and G{gamma} subunits, with the Gß and G{gamma} subunits forming a tightly associated Gß{gamma} dimer and acting as a single functional unit. In rods and cones, the conversion of light signals into neuronal signals (phototransduction) is mediated by the cell-specific G protein, transducin (T{alpha}Tß{gamma}).3 4 On activation by light, T{alpha}Tß{gamma} dissociates and T{alpha} in turn activates a phosphodiesterase (PDE) that reduces the intracellular concentrations of cGMP and ultimately triggers membrane hyperpolarization. Light also triggers a plethora of other reactions that orchestrate the deactivation of phototransduction proteins and the resynthesis of cGMP to ensure timely termination of membrane hyperpolarization and the return of photoreceptors to the dark condition.5 6

The ß{gamma}-subunit of transducin (Tß{gamma}) regulates several reactions in the light responses, such as promoting the receptor–transducin interaction,3 4 regulating the transducin–PDE interaction,7 and regulating the activity of guanylate cyclase,8 an enzyme that synthesizes cGMP. The normal functions of Tß{gamma} require proper hydrophobic translational modifications at the C-terminus of the T{gamma} subunit. The nascent T{gamma}, like other members of the G{gamma} family, contains a C-terminal CAAX motif that signals three steps of reactions collectively called isoprenylation: (1) the addition of a polyisoprenyl group to the cysteine (C) residue; (2) proteolytic cleavage next to the modified cysteine to remove the AAX residues; and (3) methylation of the carboxyl group of the newly generated C-terminal cysteine.9 10 Of the three reactions, only the last methylation is reversible and is believed to be a potential step for regulating the activities of isoprenylated proteins.11 All G{gamma} isoforms are modified by the geranylgeranyl group, a 20-carbon isoprenoid,9 10 except that T{gamma} is modified by a farnesyl group, a 15-carbon isoprenoid. Because the latter isoprenoid is less hydrophobic, it has been suggested that methylation plays a more important regulatory role in Tß{gamma} functions.11

The light responses of rods and cones differ in their sensitivity and kinetics. Compared with rods, cones are less sensitive to light, respond and recover more quickly, and adapt to a wider range of illumination.3 Considerable evidence has accumulated to indicate that the deactivation of phototransduction plays a major role in the rod and cone differences.5 6 12 13 The primary proteins of the cone phototransduction cascade are isoforms that are distinct from those of rods, whereas many proteins involved in deactivation and recovery are shared.3 5 6 12 13 Transducins from rods and cones are composed of the T{alpha}1, Tß1, T{gamma}1, and the T{alpha}2, Tß3, T{gamma}8 subunits.14 15 16 Mutations in T{alpha}117 and T{alpha}218 have been found to be associated with visual defects in rods and cones, respectively, but it is not known how these isoforms contribute to the specialization of rod and cone physiology. Because of the scarcity of cones in most mammalian retinas, the biochemistry of cone transducin is not as well characterized as the rod counterpart. A cDNA encoding for cone T{gamma}8 has been characterized.16 The predicted amino acid sequence contains the CAAX motif, but the structure of the mature T{gamma}8 has not been determined. It has been reported that cone transducin19 and cone PDE20 are more soluble than their rod counterparts. In view of the role of isoprenylation in membrane targeting, it is of interest to determine whether the C-terminus of T{gamma}8 is indeed posttranslationally modified in the same manner as T{gamma}1.

In mammalian photoreceptors, Tß{gamma} also binds tightly to phosducin (pdc), an abundant phosphoprotein that is phosphorylated by PKA and by Ca2+/calmodulin protein kinase II (CAMKII).21 22 23 The affinity of pdc for Tß{gamma} is reduced modestly by PKA phosphorylation at one serine residue24 but drastically after multisite phosphorylation by CAMKII.23 In the photoreceptor, the levels of phosphorylated pdc are highest in the dark and lowest in the light.25 In vitro, pdc inhibits the light-activated PDE by binding to Tß{gamma} and inhibiting the receptor–transducin interaction. The inhibition is abolished by phosphorylation.23 26 27 Pdc also inhibits ubiquitylation of Tß{gamma} and the inhibition is abolished by phosphorylation.28 It has been reported that transgenic mice expressing pdc with a defective PKA phosphorylation site exhibit reduced sensitivity to light as well as slow and incomplete recovery from exposure to light (Hamasaki DI, et al. IOVS 1996;37:ARVO Abstract 3720). Moreover, these mice show slow retinal degeneration, indicating that normal and reversible phosphorylation of pdc is essential for the viability of photoreceptor cells (Hamasaki DI, et al. IOVS 1995;36:ARVO Abstract 2933). Besides the outer segment, both pdc and Tß{gamma} are found abundantly in the photoreceptor cell body, the synaptic terminal, and the nucleus.21 29 Clearly, the light- and phosphorylation-regulated interaction between pdc and Tß{gamma} represents a key regulatory step in coordinating light-regulated photoreceptor activities. Our laboratory has recently shown that pdc expressed by bovine cones is identical with rod pdc, and it also colocalizes with Tß{gamma} throughout the cone photoreceptor (Lee RH, et al., unpublished observations, 2003). A comparison between the binding of pdc to the rod and cone Tß{gamma} will provide information to indicate or exclude a role for pdc and Tß{gamma} interaction in the specialized rod and cone physiology.

We have developed a purification protocol that separates cone Tß{gamma} from the very similar and much more abundant rod Tß{gamma}. The present study was focused on the posttranslational modification of T{gamma}8, the interaction between pdc and cone Tß{gamma}, and how the interaction is affected by posttranslational modifications on both T{gamma}8 and pdc. By comparing the characteristics of rod and cone Tß{gamma} against the crystal structures of rod Tß{gamma} and the rod pdc/Tß{gamma} complex, we determined that cone Tß{gamma} has the same conformation and interacts with pdc in the same manner as rod Tß{gamma}. Most interesting, changes in Tß{gamma} characteristics by T{gamma} methylation reveal conformational changes on a surface domain that is essential for Tß{gamma} functions and support a regulatory role for reversible methylation. In the remainder of this report, Tß{gamma} is used to refer to the general characteristics of ß{gamma}-transducin. The rod and cone Tß{gamma} are referred to as Tß1{gamma}1 and Tß3{gamma}8, respectively, when the identity of the individual subunit is essential for clarity.


   Methods
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Preparation of Cone Tß3{gamma}8
Pdc/Tß1{gamma}1, pdc, Tß1{gamma}1, and a fraction enriched in pdc/Tß3{gamma}8 were purified from frozen bovine retinas as described.15 The pdc/Tß3{gamma}8-enriched fraction was further separated from pdc/Tß1{gamma}1 by a separation column (CHT10-I; Bio-Rad, Hercules, CA) pre-equilibrated in 10 mM potassium phosphate (pH 6.8) at 25°C and eluted with a 10- to 90-mM phosphate gradient at 1.5 mL/min over a period of 40 minutes, and a peak containing purified pdc/Tß3{gamma}8 was eluted by 60 mM phosphate. To obtain Tß3{gamma}8, pdc/Tß3{gamma}8 was applied to a Sepharose column (1.6 x 15 cm, Q-Sepharose; Amersham Bioscience, Piscataway, NJ) pre-equilibrated at room temperature with 0.3 M Tris-HCl (pH 8.0) and eluted at 3 mL/min with a 400 mL 0.3- to 0.8-M Tris-HCl linear gradient. Tß3{gamma}8 was eluted as a sharp peak at the beginning of the gradient.

Reversed-Phase HPLC
The Pdc/Tß3{gamma}8 or pdc/Tß1{gamma}1 complex, dissolved in 6 M guanidine HCl, was injected onto a silica based C4 column (BioRad RP-304, 4.6 x 250 mm). The solvent consisted of 0.1% aqueous trifluoroacetic acid (TFA, solvent A) and 95% acetonitrile containing 0.1% TFA (solvent B). The column was pre-equilibrated in 5% B, and the bound proteins were eluted at 1 mL/min by a 60-minute linear gradient from 5% to 100% B. The effluent was monitored by absorbance at 280 nm, and 0.5-mL fractions were collected. Individual fractions were subjected to SDS-PAGE followed by Coomassie brilliant blue (CBB) staining and by Western blot analysis with antisera against the protein of interest.

Enzymatic Digestion of T{gamma}8
T{gamma}8 was dissolved in 50 µL of 50 mM NH4HCO3 (pH 9.5) and incubated with Arg-C or Lys-C (CalBiochem, La Jolla, CA) at room temperature for 16 hours at the substrate-to-enzyme ratio of 100:1. The digestion was stopped by freezing and lyophilization.

Electrospray Ionization Mass Spectrometry
The masses of intact T{gamma}8 and its peptide fragments were measured by a triple quadrupole mass spectrometer (API III; Perkin-Elmer Sciex, Thornhill, Ontario, Canada) fitted with an ion spray source. Positive ion protein spectra were produced by injection of the intact proteins or proteolytic peptides dissolved in water/acetonitrile/TFA (95/5/0.1, vol/vol/vol). Data were recorded with the mass spectrometer scanning from m/z 300 to 2200 (step size 0.3 Da, dwell time 1 ms, 6.66 seconds/scan, orifice voltage 90 V for intact proteins or 65 V for tryptic peptides). The average of the spectra contributing to the peak in ion current was computed. Calculation of molecular weights from the series of multiply charged ions found in the spectra was performed on computer (MacSpec software, ver.3.3; Perkin-Elmer Sciex). The theoretical protein or peptide average (chemical) molecular weights were downloaded from http:\\prospector.ucsf.edu.

Antibodies and Other Biochemical and Analytical Procedures
Antisera against pdc (gertie), Tß1 (ß-636), Tß3 (ß-638), and T{gamma}8 have been described.15 16 24 Anti-T{gamma}115 was a kind gift from Bernard Fung (UCLA). The PKA catalytic subunit was purified from rabbit skeletal muscle as described.21 PKA phosphorylation, SDS-PAGE, and the bandshift binding assay for pdc and Tß{gamma} interaction were performed as described.24 Native-PAGE24 was performed at either pH 7.5 or 8.8 and at either 4°C or 22°C, as indicated in text. The separated protein bands were visualized by CBB staining and identified by comigration with standard proteins and by Western blot analysis with appropriate antibodies. The intensities of CBB-stained bands and immunoreactive bands were quantified by densitometric scanning (GS700 densitometer; Bio-Rad).


   Results
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Separation of Rod T{gamma}1 and Cone T{gamma}8 by RP-HPLC
Pdc/Tß1{gamma}1, in parallel with pdc/Tß3{gamma}8, was subjected to reverse-phase (RP)-HPLC. The elution of proteins was monitored by absorbance at 280 nm (Fig. 1) , CBB staining, and Western blot analysis of individual fractions, using antisera against T{gamma}1, T{gamma}8, Tß1, Tß3, and pdc (results not shown). The elution profiles from both protein complexes were similar and contained multiple absorbance peaks, indicating subunit dissociation. T{gamma}1 was detected in the fraction corresponding to absorbance peak Gr (33.7 minutes and 36% acetonitrile). T{gamma}8 was detected in both peak Gc1 and peak Gc2 (30.5–31.5 minutes and approximately 33%–35% acetonitrile). Thus, RP-HPLC is useful for the complete separation of cone T{gamma}8 from rod T{gamma}1. The other subunits from both complexes eluted similarly. Pdc was detected in fractions corresponding to the absorbance peak at 38 minutes and 40% acetonitrile. The Tß1 and Tß3 subunits, respectively, were detected in multiple fractions eluting between 35 and 37 minutes. The immunoreactive signals for both peptides were much weaker than those of T{gamma} and pdc, and the respective profiles did not match the sharp UV absorbance peaks detected in the same region. The poor recovery of the Tß1 and the Tß3 peptides from the hydrophobic column is consistent with previous observations30 and their hydrophobic nature. In fact, none of the sharp UV-absorbing peaks could be correlated with any CBB stained bands, suggesting that they are associated with small molecules not fixed on the SDS gels. The nature of the absorbing substance(s) was not investigated further.



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 1. RP-HPLC of rod and cone pdc/Tß{gamma} complex and the molecular weight reconstructs of T{gamma}. The pdc/Tß1{gamma}1 (A) or the pdc/Tß3{gamma}8 (B) was subjected to RP-HPLC, and the elution profile was monitored by absorbance at 280 nm. Western blot analyses of individual fractions showed that absorbance peak Gr contained T{gamma}1, peaks Gc1 and Gc2 contained T{gamma}8, and peak pdc contained pdc (results not shown). Peaks Gr, Gc1, and Gc2 were subjected to ESIMS. The resultant mass spectra were deconvoluted to reveal the masses and the relative abundance of molecules in each peak (A, B, insets).

 
Determination of the Mass of Mature Cone T{gamma}8
The open reading frame in the cDNA of bovine cone T{gamma}8 translates into a 69-amino-acid peptide,16 but the mass and the structure of the mature T{gamma}8 protein have not been determined. Electrospray ionization mass spectrometry (ESIMS) of T{gamma}1 (peak Gr) revealed two masses of 8330.1 and 8316.2 Da, (Fig. 1A , inset) which agree with the reported values for the methylated (8329.7 Da) and the demethylated (8315.7 Da) T{gamma}1.31 ESIMS of T{gamma}8 in Peak Gc1 and peak Gc2 also revealed two masses (7529.9 and 7544.1 Da), with the lower mass present predominantly in peak Gc1 and the higher mass in peak Gc2 (Fig. 1B , insets). Based on the total ion current, the lighter and the heavier masses were estimated to compose approximately 15% and 85%, respectively, of the total T{gamma}8. Together, these indicate that there are two forms of T{gamma}8 that are separable by RP-HPLC—the heavier and major form being more hydrophobic. The mass of the major T{gamma}8 is 185 Da lower than the theoretical mass (7728.0 Da) of the deduced T{gamma}8 sequence, but is 42 Da higher than the theoretical mass, which assumes that T{gamma}8 undergoes the same posttranslational modifications as T{gamma}1—namely, isoprenylation with a farnesyl group at the C-terminus and the removal of methionine from the N-terminus of the nascent T{gamma}1 peptide.31 Because T{gamma}8 was not amenable to protein sequencing,16 the extra 42 Da is interpreted to indicate an acetyl group (42 Da) that blocks the new N-terminus generated by the removal of methionine. The minor T{gamma}8, being 14 Da lighter and less hydrophobic, is likely the demethylated form of T{gamma}8 (14 Da for —CH2).

Mass Peptide Mapping of Cone T{gamma}8
To confirm the proposed posttranslational modifications, T{gamma}8 was digested with either Arg-C or Lys-C. The masses of the resultant fragments were mapped to the proposed structure of the mature T{gamma}8 (Fig. 2) . ESIMS of the Arg-C digest revealed four molecules with observed masses of 7529.9, 4522.4, 4508.2, and 3039.1 Da. The heaviest mass is the same as that of the minor T{gamma}8, indicating incomplete digestion by Arg-C. The lightest mass matched the theoretical mass (3038.7 Da) of the N-terminally acetylated 1-25 peptide. The remaining pair of masses matched the theoretical masses (4522.0 and 4508.0 Da) for the methylated and demethylated forms, respectively, of the C-terminally farnesylated 26-65 peptide. Likewise, ESIMS of the Lys-C digest identified five fragments that covered 100% of the T{gamma}8 sequence with the proposed posttranslational modifications. These results conclusively established that the mature T{gamma}8 is a 65-amino-acid peptide with an acetylated N-terminus and a farnesylated and methylated C-terminus.



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 2. Mass peptide mapping of T{gamma}8. T{gamma}8 was digested with either Arg-C or Lys-C and the masses of the resultant fragments (identified by R’s and L’s, respectively) were determined by ESIMS. Each fragment was identified by its mass and mapped to the proposed sequence for the mature T{gamma}8. The location and the observed and the theoretical masses (in parenthesis) of each fragment are indicated by the brackets. The R2' fragment is the demethylated form of R2. The posttranslationally added moieties are indicated by italic notations, including the acetyl (Ac), the farnesyl (C12H25), and the methyl (CH3) groups. The sequence in parentheses identifies the amino acid residues that are removed from the predicted nascent peptide by posttranslational modification.

 
Native-PAGE of Cone Tß3{gamma}8: The Effects of Methylation and pdc Binding
The migration patterns of pdc/Tß3{gamma}8 and Tß3{gamma}8 during native-PAGE at pH 7.5 and 4°C were compared, and the identity of each stained protein band was established by Western blot analysis with antisera against Tß3, T{gamma}8, and pdc (Fig. 3A) . The pdc/Tß3{gamma}3 complex migrated as a single band with an average mobility of 0.41. In contrast, Tß3{gamma}8, which was judged to be homogeneous by SDS-PAGE and CBB staining of the Tß3 band, separated into two bands with an average electrophoretic mobility of 0.25 and 0.30. The relative intensity between the slow- and fast-moving bands varied from 10:1 to 1:1, among different Tß3{gamma}8 samples, with the intensity of the fast-moving band notably higher in the older preparations. Because demethylation at the C-terminus of T{gamma}8 introduces one additional negative charge, the slow and fast bands may be associated with the methylated and the demethylated forms of Tß3{gamma}8, respectively. Indeed, comparison of results from four different Tß3{gamma}8 samples showed that the relative intensity of the fast and slow bands paralleled the relative intensity of peaks Gc1 and Gc2, respectively, in the corresponding RP-HPLC profile (Fig. 1B) . To further ascertain our band assignment, we incubated Tß3{gamma}8 with immobilized pig liver esterase (iPLE), an enzyme shown to cleave the C-terminus methyl ester bond selectively.30 As expected, this treatment resulted in the disappearance of the slow band with concomitant increase in the intensity of the fast band (Fig. 3B) .



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 3. Native-PAGE of Tß3{gamma}8: the effect of pdc binding and methylation. (A) Effect of pdc binding: Pdc/Tß3{gamma}8 (sample 1) and Tß3{gamma}8 (sample 2), containing equal amounts of Tß3 were subjected to native-PAGE at pH 7.5 and 4°C or to SDS-PAGE followed by CBB staining. The identity of individual bands was confirmed by Western blot analysis with antisera against Tß3, T{gamma}8, or pdc. (mb), mobility of protein bands during Native-PAGE. (B) Effect of methylation: Pdc/Tß3{gamma}8 (sample 1) and Tß3{gamma}8 (sample 2) were incubated with or without iPLE and were subjected to native-PAGE at pH 7.5 or 8.8, followed by CBB staining.

 
The mobility of the pdc/Tß3{gamma}8 complex during native-PAGE at pH 7.5 was higher than Tß3{gamma}8, apparently as the result of binding to the more acidic pdc. The migration of the complex as a single band also suggests that pdc binding masks the heterogeneity in T{gamma}8 methylation that was detected by both RP-HPLC and mass spectrometric analysis (Figs. 1 2) . To confirm this notion, pdc/Tß3{gamma}8 was subjected to native-PAGE at pH 8.8 and 4°C, which separated the complex into three stained bands (Fig. 3B) . Western blot analysis (results not shown) indicated that they are, in the order of reducing mobility, pdc and the fast and the slow Tß3{gamma}8 bands. The identity of the fast and slow Tß3{gamma}8 bands as demethylated and methylated dimers, respectively, was confirmed by the ability of iPLE treatment to eliminate the slower band completely. These results indicate that pdc binding masks the methylation-associated mobility differential intrinsic to Tß3{gamma}8, but it does not block the access of the T{gamma}8 C-terminus to cleavage by iPLE. In contrast, the methyl group in the T{alpha}Tß{gamma} trimeric complex was reported to be resistant to iPLE cleavage.30 The difference may reflect steric hindrance by hydrophobic interaction between the lipid moieties from the N-terminus of T{alpha} and the farnesyl group of T{gamma}1.

The results of Western blot analysis of pdc/Tß3{gamma}8 and Tß3{gamma}8 from native-PAGE deserve further comment. Figure 3A (left) shows that the intensities of both Tß3 and T{gamma}8 immunoreactive signals for pdc/Tß3{gamma}8 were notably weaker than the unbound Tß3{gamma}8, as observed previously for the rod pdc/Tß1{gamma}1 complex.24 Parallel analysis (right) by SDS-PAGE confirmed that both samples contained equal amounts of Tß3 (and presumably T{gamma}8). Another control study showed that pdc/Tß3{gamma}8 and pdc samples containing equal amounts of pdc also gave identical immunoreactive signals for anti-pdc (results not shown), indicating that the low Tß3 signal in the pdc/Tß3{gamma}8 was not caused by ineffective transblot of the trimeric complex. Although the precise reason for the difference in signal intensities is not understood, these differences do not affect our confidence in the assignment of protein bands in the native-PAGE.

Comparison of Rod and Cone Tß{gamma} during Native-PAGE
Rod Tß1{gamma}1 separated during native PAGE at pH 7.5 into two bands with an average mobility of 0.33 and 0.37, respectively (Fig. 4A) . Preincubation with iPLE removed the slow-moving band while increasing the intensity of the fast band (results not shown). The trimeric pdc/Tß1{gamma}1 complex migrated as a single band with an average mobility of 0.46. Thus, the relative electrophoretic mobility of pdc/Tß1{gamma}1 and pdc/Tß3{gamma}8 paralleled that of Tß1{gamma}1 and Tß3{gamma}8, respectively. Together, this indicates that pdc binding masks the methylation-associated mobility differential in either rod or cone Tß{gamma} without affecting the intrinsic mobility differential between the rod and cone Tß{gamma}.



View larger version (35K):
[in this window]
[in a new window]
  		FIGURE 4. Comparison of the pdc binding affinity of Tß1{gamma}1 and Tß3{gamma}8. (A) Two micrograms each of Tß1{gamma}1 (lanes 1 to 5) and Tß3{gamma}8 (lanes 6 to 10) were incubated with indicated amounts of pdc before being subjected to native-PAGE at pH 7.5 and at 4°C. The formation of pdc/Tß{gamma} complex and the concomitant removal of unbound Tß{gamma} dimers were visualized by CBB staining. (B) The binding of pdc to Tß1{gamma}1 and Tß3{gamma}8 was monitored by dose-dependent removal of the Tß{gamma} dimers. The amounts of individual unbound dimers were quantified by densitometric scanning and expressed as a percentage of corresponding signals detected in the absence of added pdc (control). ({circ}) Tß1{gamma}1-OCH3; ({square}) Tß1{gamma}1-OH; (•) Tß3{gamma}8-OCH3; ({blacksquare}) Tß3{gamma}8-OH.

 
Characterization of Tß3{gamma}8 and pdc Interaction: Comparison to Tß1{gamma}1
A native-PAGE–based bandshift binding assay was developed to examine the binding between Tß1{gamma}1 and pdc.24 To compare the interaction between Tß3{gamma}8 and pdc, equal amounts of Tß1{gamma}1 and Tß3{gamma}8 were incubated with limited but increasing amounts of pdc (Fig. 4A) . After native-PAGE at pH 7.5 and 4°C, the formation of the respective pdc/Tß{gamma} complexes was monitored by the dose-dependent removal of the unbound Tß{gamma} (Fig. 4B) . Tß1{gamma}1 and Tß3{gamma}8 showed similar dose–response curves, with 50% of methylated Tß{gamma} removed by a 0.7 molar ratio of added pdc (pdc:Tß), and 50% of the demethylated Tß{gamma} removed by 1.5 molar ratio of added pdc. Because the abundance of the methylated and demethylated forms of Tß{gamma} were essentially identical in both rod and cone samples, we interpret the results to indicate that the methylated Tß{gamma} showed twofold higher affinity for pdc than the demethylated Tß{gamma}. Thus, rod and cone Tß{gamma} not only bind pdc with essentially identical affinities, but are affected by T{gamma} methylation in an identical manner.

We have shown that the binding affinity between pdc and Tß1{gamma}1 is three times lower after PKA phosphorylation of pdc, but the phosphorylated pdc/Tß1{gamma}1 remained complexed during native-PAGE at pH 7.5 and at 4°C.24 Because increased temperature was found to favor the dissociation,21 we tested the effect of temperature on the migration pattern of PKA-phosphorylated pdc/Tß1{gamma}1 (Fig. 5) . Figure 5A shows that pdc/Tß1{gamma}1, at 4°C and with or without PKA phosphorylation, migrated similarly as a trimeric complex. Figure 5B shows that the migration patterns at 22°C are strikingly different. In the unphosphorylated pdc/Tß1{gamma}1, four bands were detected that corresponded to, in the order of increasing mobility, the Tß1{gamma}1-OCH3, Tß1{gamma}1-OH, pdc/Tß1{gamma}1, and pdc. The staining intensity of the pdc/Tß1{gamma}1 was considerably darker than those of the unbound Tß1{gamma}1 and pdc, indicating that most of the protein remained complexed. In the phosphorylated pdc/Tß1{gamma}1, four bands were also detected, but the staining intensity in the pdc/Tß1{gamma}1 region was greatly diminished, with concomitant increase in the staining of the Tß1{gamma}1 and pdc subunits, indicating dissociation of most of the phosphorylated pdc/Tß1{gamma}1. It was noteworthy that the unbound pdc in the phosphorylated sample was not only labeled by 32P (Fig. 5C) but also showed higher mobility than corresponding pdc in the unphosphorylated sample, consistent with an increase in the negative charges as the result of phosphorylation. Parallel analysis showed that pdc/Tß3{gamma}8 also underwent PKA phosphorylation-induced changes in migration pattern during native-PAGE at 22°C (Figs. 5D 5E) , suggesting that binding between phosphorylated pdc and Tß3{gamma}8 is also reduced by at least threefold.



View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 5. Effect of PKA phosphorylation on the binding of pdc to Tß3{gamma}8. Four micrograms each of pdc/Tß1{gamma}1 and pdc/Tß3{gamma}8 were incubated with PKA with or without [32P] adenosine triphosphate (ATP). At the end of incubation samples were subjected to native-PAGE at pH 7.5 and at either 4°C (A) or 22°C (B–E). After CBB staining, the gels were dried and autoradiographed to locate the 32P-labeled proteins. The designation of various Tß{gamma} and pdc/Tß{gamma} bands is as in the previous figures. Pdc and 32P-pdc are unphosphorylated and 32P-labeled pdc, respectively.

 

   Discussion
Top
 Abstract
 Methods
 Results
 Discussion
References
 
There is intense interest in understanding the molecular mechanism of cone physiology, because cones are responsible for visual acuity and cones survive better than rods during trauma and diseases. To gain better understanding of how the cell-specific transducin contributes to the cone’s unique physiology, we determined the structure of T{gamma}8. In addition to confirming the entire amino acid sequence as predicted by the G{gamma}8 cDNA,16 we showed that the mature T{gamma}8 is acetylated at the N-terminus and farnesylated and methylated at the C-terminus. This indicates that the lower membrane affinity of cone transducin is unrelated to isoprenylation of the Tß3{gamma}8 subunit. The modification observed at the N-terminus is also consistent with the empiric rule that peptides with an N-terminal alanine, such as in T{gamma}8, are usually acetylated, whereas those with an N-terminal proline, such as in T{gamma}1, are not.32 The purified cone Tß3{gamma}8 dimer, like rod Tß1{gamma}1, is heterogeneous with respect to the state of methylation at the C-terminus of T{gamma}8. The amount of methylated T{gamma}8 is highest in the freshly prepared pdc/Tß3{gamma}8 complex and lowest in the dissociated Tß3{gamma}8 that has been stored for an extended period, raising the intriguing possibility that pdc may exert protection against demethylation. Loew et al.33 showed that pdc induces minor conformational changes in Tß1{gamma}1, resulting in the formation of a shallow cavity in Tß1 that partially buries the farnesyl group. The effect of pdc on the conformation at the methylation site is less clear, because the precise conformation in the crystallized Tß1{gamma}1 is not known.34 Results from our iPLE experiments indicate that pdc does not block demethylation by this enzyme, but they do not rule out the possibility that pdc may slow down demethylation.

The presence of the same pdc in rods and cones of bovine retinas has been established (Lee RH, et al., unpublished observation, 2003). In this study, we compared the binding of the rod and cone Tß{gamma} to pdc and used the crystal structures of Tß1{gamma}1 and pdc/Tß1{gamma}1 as models to gain insight into the Tß3{gamma}8 and pdc interaction. It was found that both Tß{gamma} bind pdc with the same affinity and are affected in the same manners by PKA phosphorylation of pdc and by changes in the status of methylation in the respective T{gamma}. The binding assay for pdc and Tß{gamma} was developed based on the ability of the native-PAGE to separate the pdc/Tß{gamma} complex from the unbound pdc and Tß{gamma}. The electrophoretic mobility of Tß{gamma} is highly sensitive to variations and other subtler changes on the surface and offers an independent parameter for evaluating the conformation of Tß{gamma} and the topology of pdc and Tß{gamma} interaction. Rod Tß{gamma} shows higher mobility than cone Tß{gamma}, consistent with their respective isoelectric points (pIs) of 3.9 and 4.1.15 For each type of Tß{gamma}, the demethylated form shows higher mobility than the methylated form. Pdc binding masks the methylation-associated mobility differential, but not the intrinsic rod/cone difference. Collectively, these observations indicate that (1) the pdc binding surface on cone Tß{gamma} is the same as rod Tß{gamma}; (2) the methylation site is situated on a conserved surface on rod and cone Tß{gamma}, and this surface is either close to or overlaps with the pdc binding domain; and (3) away from the pdc binding domains, there exist a unique surface whose electrostatic characteristics underlies the intrinsic mobility differential between rod and cone Tß{gamma}. These notions are entirely consistent with the three-dimensional structure of Tß1{gamma}1.33 34 35 For example, the most variable segments of Tß1 and T{gamma}1, located at the far N-termini of both peptides, intertwine into helixes or loops that form an exposed surface away from the pdc binding surface. In summary, our results showed that rod and cone Tß{gamma} are identically modified at the C-terminus of the respective T{gamma} subunits, assume the same conformation, and interact with pdc in the same manner, indicating that pdc and Tß{gamma} interaction does not play a major role in the specialization of rod and cone physiology.

Hydrophobic binding through the isoprenyl group is the major mechanism by which Tß{gamma} and, other Gß{gamma} isoforms target cell membranes and their effector proteins.9 10 Reversible methylation at the neighboring carboxyl group is thought to play a role in regulating isoprenylated proteins.11 In rod outer segments, Tß1{gamma}1, along with the {alpha}, ß-subunits of PDE and the small G protein, undergoes reversible methylation.36 The enzymes that specifically catalyze the methylation and demethylation of the farnesylated cysteine have also been identified.37 38 Several in vitro studies have shown that methylation increases the Tß{gamma}-mediated activities30 31 39 ; most discussions credited the effects to enhanced hydrophobicity and membrane binding. However, several observations from this study indicate that reversible methylation also induces conformational changes that modify the electrostatic surface of Tß{gamma}. First, during native-PAGE the demethylated Tß{gamma} migrated much faster than the methylated dimer, indicating large electrostatic changes that could not be explained by the difference of one negative charge on the intrinsically acidic rod or cone Tß{gamma}.15 Second, pdc masked the mobility differential between methylated and demethylated Tß{gamma}, suggesting an overlap between the pdc binding domain and the surface that is electrostatically regulated by T{gamma} methylation. Third, the demethylated Tß{gamma} shows twofold lower affinity for pdc than the methylated dimer, even though the crystal structure shows that the T{gamma} methylation site is outside of the pdc binding surface on Tß{gamma}.33 35 The crystal structure of Tß1{gamma}1 suggests a plausible mechanism for such changes. It is known that the farnesyl group and the methylation site of T{gamma} is surrounded by a prominent patch of positive charges formed by 11 basic residues that scatter throughout the Tß peptide but come together in the three dimensional space.34 35 40 This means that reversible methylation, by regulating the presence or absence of a negative charge juxtaposed at the center of domain, is poised to induce far-reaching changes in the spatial arrangement of these basic residues. Of note, this basic surface on Tß{gamma} is not only a part of membrane binding domain but also overlaps with a highly conserved sequence motif, HIKE,42 which encompasses amino acids residues that are essential for as well as regulate Tß{gamma} interaction with pdc and most of the Gß{gamma} effectors identified to date (Fig. 6) . A recent report has shown that electrostatic interactions between this positively changed domain and the membranous acidic phospholipids increase the membrane partitioning of Tß{gamma} by one order of magnitude.40 This means that electrostatic changes induced by reversible methylation may also play a role in regulating Tß{gamma} binding to membranes. It is noteworthy that Tß{gamma} binding to pdc and the Tß{gamma}-stimulated membrane-dependent GTP-S exchange by T{alpha}30 are similarly affected by reversible methylation, suggesting a common underlying regulatory mechanism.



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 6. Alignment of the HIKE motif in Tß1 with amino acid residues that contact pdc and form the positive-charged domain. The amino acid sequence between residues 27 and 101 of bovine Tß1 is depicted by the one-letter code. The core of the HIKE sequence motif resides between residues 27 and 77.41 (+), amino acids that are completely conserved among all HIKE-containing molecules and across species; ({diamond}) residues that directly contact pdc33 35 ; () residues that are part of the 11 basic residues forming the positive-charged domain that surrounds the C-terminus of T{gamma} and regulates Tß{gamma} binding to membranes.40 Blades 1 and 7 identify residues in this segment of Tß1 that fold into blade 1 and a part of blade 7, respectively, in the ß propeller structure of the Tß{gamma} dimer (nomenclature according to Gaudet et al.35 ).

 
In summary, this study has presented the first evidence of methylation-induced conformational changes on a multifunctional surface domain on Tß{gamma} that is essential for its functions. We suggest that these changes are essential mechanisms by which reversible methylation regulates Tß{gamma} activities.


   Footnotes
 
Supported by VA Medical Research Service (RHL), National Eye Institute Grant EY09936 (RHL), and the W. M. Keck Foundation (KFF).

Submitted for publication April 29, 2003; revised August 4, 2003; accepted August 5, 2003.

Disclosure: F. Chen, None; P.-S. Ng, None; K.F. Faull, None; R.H. Lee, None

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Corresponding author: Rehwa H. Lee, Molecular Neurology Laboratory, mail code 151B9, VA Greater LA Healthcare System at Sepulveda, 16111 Plummer St., Sepulveda, CA 91343; rlee{at}ucla.edu.


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Hamm, HE. (1998) The many faces of G protein signaling J Biol Chem 273,669-672[Free Full Text]
  2. Gutkind, JS. (1998) The pathways connecting G protein-coupled receptors to the nucleus through divergent mitogen-activated protein kinase cascades J Biol Chem 273,1839-1842[Free Full Text]
  3. Pugh, EN, Lamb, TD. (2000) Phototransduction in vertebrate rods and cones: molecular mechanisms of amplification, recovery and light adaptation Stavenga, DG de Grip, WJ Pugh, EN, Jr eds. Molecular Mechanisms in Visual Transduction ,183-255 Elsevier Amsterdam.
  4. Arshavsky, VY, Lamb, TD, Pugh, EN, Jr (2002) G proteins and phototransduction Annu Rev Physiol 64,153-187[CrossRef][Medline][Order article via Infotrieve]
  5. Burns, ME, Baylor, DA. (2001) Activation, deactivation, and adaptation in vertebrate photoreceptor cells Annu Rev Neurosci 24,779-805[CrossRef][Medline][Order article via Infotrieve]
  6. Hurley, JB. (2002) Shedding light on adaptation J Gen Physiol 119,125-128[Free Full Text]
  7. Yamazaki, A, Hayashi, F, Tatsumi, M, Bitensky, MW, George, JS. (1990) Interaction between the subunits of transducin and cGMP phosphodiesterase in Rana catesbiana rod photoreceptor J Biol Chem 265,11539-11548[Abstract/Free Full Text]
  8. Wolbring, G, Baehr, W, Palczewski, K, Schnetkamp, PPM. (1999) Light inhibition of bovine retinal rod guanylyl cyclase mediated by ß{gamma}-transducin Biochemistry 38,2611-2616[CrossRef][Medline][Order article via Infotrieve]
  9. Marshall, CJ. (1993) Protein prenylation: a mediator of protein-protein interaction Science 259,1865-1866[Free Full Text]
  10. Wedgaetner, PL, Wilson, PT, Bourne, HR. (1995) Lipid modifications of trimeric G proteins J Biol Chem 270,503-506[Free Full Text]
  11. Rando, RR. (1996) Chemical biology of isoprenylation/methylation Biochem Soc Trans 24,682-687[Medline][Order article via Infotrieve]
  12. Cowan, CW, Fariss, RN, Sokal, I, Palczewski, K, Wensel, TG. (1998) High expression levels in cones of RGS9, the predominant GTPase accelerating protein of rods Proc Natl Acad Sci USA 95,5351-5356[Abstract/Free Full Text]
  13. Lyubarsky, AL, Chen, C, Simon, MI, Pugh, EN, Jr (2000) Mice lacking G-protein receptor kinase 1 have profoundly slowed recovery of cone-driven retinal responses J Neurosci 20,2209-2217[Abstract/Free Full Text]
  14. Simon, MI, Strathmann, MP, Gautam, N. (1991) Diversity of G proteins in signal transduction Science 252,802-808[Abstract/Free Full Text]
  15. Fung, BKK, Lieberman, BS, Lee, RH. (1992) A third form of the G protein ß subunit: purification and biochemical properties J Biol Chem 267,24782-24788[Abstract/Free Full Text]
  16. Ong, OC, Yamane, HK, Phan, KB, et al (1996) G protein {gamma} subunit of cone photoreceptor J Biol Chem 270,8496-8500
  17. Dryja, TP, Hahn, LB, Reboul, T, Arnaud, B. (1996) Missense mutation in the gene encoding the alpha subunit of rod transducin in the Nougaret form of congenital stationary night blindness Nat Genet 13,358-360[CrossRef][Medline][Order article via Infotrieve]
  18. Aligianis, IA, Forshew, T, Johnson, S, et al (2002) Mapping of a novel locus for achromatopsia (ACHM4) to 1p and identification of a germline mutation in the alpha subunit of cone transducin (GNAT2) J Med Genet 39,656-660[Abstract/Free Full Text]
  19. Lerea, CL, Somers, DE, Hurley, JB, Klock, IB, Bunt-Milam, AH. (1986) Identification of specific transducin alpha subunits in retinal rod and cone photoreceptors Science 234,77-80[Abstract/Free Full Text]
  20. Gillespie, PG, Beavo, JA. (1988) Characterization of a bovine cone photoreceptor phosphodiesterase purified by cyclic GMP-sepharose chromatography J Biol Chem 263,8133-8141[Abstract/Free Full Text]
  21. Lee, RH, Lieberman, B, Lolley, RN. (1987) A novel complex from bovine visual cells of a 33,000-dalton phosphoprotein with ß and {gamma}-transducin: purification and subunit structure Biochemistry 26,3983-3990[CrossRef][Medline][Order article via Infotrieve]
  22. Lee, RH, Brown, BM, Lolley, RN. (1990) Protein kinase A phosphorylates retinal Pdc on serine 73 in situ J Biol Chem 265,15860-15866[Abstract/Free Full Text]
  23. Thulin, CD, Savage, JR, Mclaughlin, JN, et al (2001) Modulation of G protein regulator phosducin by Ca/Calmodulin protein kinase II phosphorylation and 14–3-3 protein binding J Biol Chem 276,23805-23815[Abstract/Free Full Text]
  24. Chen, F, Lee, RH. (1997) Phosducin and betagamma-transducin interaction I: effects of posttranslational modifications Biochem Biophys Res Comm 233,370-374[CrossRef][Medline][Order article via Infotrieve]
  25. Lee, RH, Brown, BM, Lolley, RN. (1984) Light-induced dephosphorylation of a 33K protein in rod outer segments of rat retina Biochemistry 23,972-977
  26. Lee, RH, Ting, TD, Lieberman, BS, Tobias, DT, Lolley, RN, Ho, Y-K. (1992) Phosducin and transduction interactions: down regulation by pdc or retinal cGMP cascade in reconstituted rod outer segments J Biol Chem 267,25104-25112[Abstract/Free Full Text]
  27. Yoshida, T, Willardson, BM, Wilkins, JF, Jensen, GJ, Thornton, BD, Bitensky, MW. (1994) The phosphorylation state of phosducin determines its ability to block transducin subunit interactions and inhibit transducin binding to activated rhodopsin J Biol Chem 269,24050-24057[Abstract/Free Full Text]
  28. Obin, M, Lee, B, Thulin, C, et al (2002) Ubiquitylation of transducin (T) ß{gamma}: regulation by phosducin J Biol Chem 277,44566-44575[Abstract/Free Full Text]
  29. Marguli, A, Dang, L, Pulukuri, S, Lee, R, Sitaramayya, A. (2002) Presence of phosducin in the nuclei of bovine retinal cells Mol Vis 8,477-482[Medline][Order article via Infotrieve]
  30. Parish, CA, Rando, RR. (1994) Functional significance of G protein carboxymethylation Biochemistry 33,9986-9991[CrossRef][Medline][Order article via Infotrieve]
  31. Fukada, Y, Takao, T, Ohguro, H, Yoshizawa, T, Akino, T, Shimonishi, Y. (1990) Farnesylated {gamma}-subunit of photoreceptor G protein indispensable for GTP-binding Nature 346,658-660[CrossRef][Medline][Order article via Infotrieve]
  32. Yan, SC, Grinnell, BW, Wold, F. (1989) Posttranslational modifications of proteins: some problems left to solve Trends Biochem Sci 14,264-268[CrossRef][Medline][Order article via Infotrieve]
  33. Loew, A, Ho, YK, Blundell, T, Bax, B. (1998) Phosducin induces a structural change in transducin beta gamma Structure 6,1007-1019[Medline][Order article via Infotrieve]
  34. Sondek, J, Bohm, A, Lambright, DG, Hamm, HE, Sigler, PB. (1996) Crystal structure of a GA protein ß{gamma} dimer at 2.1 a resolution Nature 379,369-374[CrossRef][Medline][Order article via Infotrieve]
  35. Gaudet, R, Bohm, A, Sigler, PB. (1996) Crystal structure at 2.4 angstroms resolution of the complex of transducin beta, gamma and its regulator, phosducin Cell 87,577-588[CrossRef][Medline][Order article via Infotrieve]
  36. Perez-Sala, D, Tan, EW, Canada, FJ, Rando, RR. (1991) Methylation and demethylation reactions of guanine nucleotide-binding proteins of retinal rod outer segments Proc Natl Acad Sci USA 88,3043-3046[Abstract/Free Full Text]
  37. Gilbert, BA, Tan, EW, Perez-Sala, D, Rando, RR. (1992) Structure-activity studies on the retinal rod outer segment isoprenylated protein methyltransferase J Am Chem Soc 19,3966-3973[CrossRef]
  38. Tan, EW, Rando, RR. (1992) Identification of an isoprenylated cysteine methyl ester hydrolase activity in bovine rod outer segment membranes Biochemistry 31,5572-5578[CrossRef][Medline][Order article via Infotrieve]
  39. Parish, CA, Rando, RR. (2000) Isoprenylation/methylation and transducin function Methods Enzymol 316,451-464[Medline][Order article via Infotrieve]
  40. Murray, D, McLaughlin, S, Honig, B. (2001) The role of electrostatic interactions in the regulation of the membrane association of G protein beta gamma heterodimers J Biol Chem 276,45153-45159[Abstract/Free Full Text]
  41. Alberti, S. (1999) HIKE, a candidate protein binding site for PH domains, is a major regulatory region of Gb protein Proteins: Structure, Function and Genetics 35,360-363[CrossRef]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Google Scholar
Google Scholar
Right arrow Articles by Chen, F.
Right arrow Articles by Lee, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, F.
Right arrow Articles by Lee, R. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS