Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2006 Nov 30;579(Pt 2):303–312. doi: 10.1113/jphysiol.2006.121772

Removal of phosphorylation sites of γ subunit of phosphodiesterase 6 alters rod light response

S H Tsang 1, M L Woodruff 2, Kerstin M Janisch 1, M C Cilluffo 3, D B Farber 4, G L Fain 2,,4
PMCID: PMC2075409  PMID: 17138607

Abstract

The phosphodiesterase 6 γ (PDE6γ) inhibitory subunit of the rod PDE6 effector enzyme plays a central role in the turning on and off of the visual transduction cascade, since binding of PDE6γ to the transducin α subunit (Tα) initiates the hydrolysis of the second messenger cGMP, and PDE6γ in association with RGS9-1 and the other GAP complex proteins (Gβ5, R9AP) accelerates the conversion of TαGTP to TαGDP, the rate-limiting step in the decay of the rod light response. Several studies have shown that PDE6γ can be phosphorylated at two threonines, T22 and T35, and have proposed that phosphorylation plays some role in the physiology of the rod. We have examined this possibility by constructing mice in which T22 and/or T35 were replaced with alanines. Our results show that T35A rod responses rise and decay more slowly and are less sensitive to light than wild-type (WT). T22A responses show no significant difference in initial time course with WT but decay more rapidly, especially at dimmer intensities. When the T22A mutation is added to T35A, double mutant rods no longer showed the prolonged decay of T35A rods but remained slower than WT in initial time course. Our experiments suggest that the polycationic domain of PDE6γ containing these two phosphorylation sites can influence the rate of PDE6 activation and deactivation and raise the possibility that phosphorylation or dephosphorylation of PDE6γ could modify the time course of transduction, thereby influencing the wave form of the light response.

Photoreceptor cGMP phosphodiesterase (PDE6) is the effector enzyme of sensory transduction; its light-dependent activation reduces the concentration of cGMP, decreasing the probability of opening of the cyclic nucleotide-gated channels in the outer segment plasma membrane. The kinetics of PDE6 activation and re-inhibition are important determinants of the wave form of the photoreceptor light response. Binding of the transducin α subunit Tα-GTP to the inhibitory PDE6γ subunit activates PDE6, and the time course of this activation determines the kinetics of onset of the photocurrent (see Pugh & Lamb, 1993). The unbinding of Tα-GDP from PDE6γ is ultimately responsible for current recovery, since turning off of PDE6 activity has been shown to be the rate-limiting step for the decay of the rod response (Sagoo & Lagnado, 1997; Krispel et al. 2006; see also Tsang et al. 2006 a)

Because PDE6γ plays such a critical role in both activation and deactivation of phototransduction, the modulation of the PDE6γ could have significant effects on the rate of change of the cGMP concentration in the outer segment. Several groups have reported phosphorylation of PDE6γ at two sites, threonine 22 (Tsuboi et al. 1994a,b; Hayashi et al. 2000; Matsuura et al. 2000; Paglia et al. 2002) at a consensus site for a proline kinase such as cyclin-dependent kinase 5 or MAP kinase,and threonine 35 (Udovichenko et al. 1994; Xu et al. 1998; Paglia et al. 2002) at a consensus site for a serine/threonine kinase such as PKA or PKC. Since the part of the PDE6γ sequence containing these two sites is known to be important for the association of the PDE6γ with Tα (Morrison et al. 1989; Artemyev et al. 1992; Takemoto et al. 1992) and with PDE6αβ (Artemyev & Hamm, 1992; Mou & Cote, 2001; Guo et al. 2005), the placing of a phosphate group on one of these threonines might be expected to produce some alteration in PDE6 activation or life-time. Several studies have examined this possibility, but the results remain controversial. One limitation of this previous work has been the insensitivity of in vitro experiments to the detection of what may be rather subtle effects of PDE6γ phosphorylation on the physiology of the rod response. In frog rods, for example, only about 4% of the total PDE6γ is phosphorylated in 10 s when 0.3% of the rhodopsin is bleached (Hayashi, 1994).

We therefore re-investigated this question with a different approach. We generated mice expressing a mutation of either one or both of the T22 and T35 sites of the PDE6γ protein, replacing the threonines with alanines. Alanine substitution would eliminate phosphorylation but is likely to be sufficiently conservative to have little effect itself on the secondary structure of the PDE6γ molecule. We then made suction-electrode recordings from single mouse rods from these animals. Our results show that the T35A and T22A mutations have dramatic effects on the rise and decay of the light response. These observations provide the first physiological evidence that phosphorylation of PDE6γ may have an important modulatory effect on rod light responses. Some of these results have been previously communicated at the annual meeting of the Association for Research in Vision and Ophthalmology (Tsang et al. 2006b).

Methods

Generation of mutant mouse lines

The mice studied in these experiments were kept in cyclic light (12 h on/12 h off) and used in accordance with the Policy on the Use of Animals in Neuroscience Research of the Society for Neuroscience. DNA constructs for the expression of PDE6γ contained 4.4 kb of the mouse opsin promoter, the complete open reading frame of the PDE6γ cDNA (Tuteja & Farber, 1988), and the polyadenylation signal of the mouse protamine gene (Lem et al. 1991). The T22A single, T35A single and T22A-T35A double point mutations were introduced by a standard PCR-based site-specific mutagenesis strategy (Tsang et al. 1998). The entire PDE6γ cDNA coding region in the transgenic construct was sequenced to confirm the introduction of the point mutation and the absence of any other changes created inadvertently. Kpn I and Xba I were used to excise vector sequences from the constructs. Oocytes were obtained from superovulated F1(DBA × C57BL/6) females mated with homozygous Pde6gtm1/Pde6gtm1 males, which lacked the gene for PDE6γ (Tsang et al. 1996). The construct was injected into the male pronuclei of oocytes under a depression slide chamber. These microinjected oocytes were cultured overnight in M16 and transferred into the oviducts of 0.5 day post-coitum pseudopregnant F1 females. The resulting transgenic mice were then backcrossed to Pde6gtm1/Pde6gtm1 mice to place the transgene into the knockout background so that the mice only expressed the mutant PDE6γ. Thereafter, mutant mice were maintained in the MF1 genetic background. Immunoblots using retinal homogenate of each transgenic line demonstrated that the levels of mutant PDE were not detectably different from those in control C57BL/6 mice (see Fig. 1). Measurements of trypsin-activated PDE activity (Tsang et al. 1998) gave similar values for the wild-type (WT) and mutant mouse lines, also indicating that PDE6γ expression was not significantly altered. C57BL/6 mice and a transgenic line expressing the wild-type level of PDE6γ were used as controls for the physiological measurements. For each of the mutant alleles, two independent transgenic lines expressing wild-type levels of mutant PDE6γ were analysed. The mice were also tested for the absence of the rd1 mutation (Pittler & Baehr, 1991).

Figure 1. Immunoblot analysis of the expression of PDE and other rod transduction proteins.

Figure 1

Open in a new tab

B6 signifies wild-type C57BL/6 control. Protein normalized to 50 μg unless otherwise stated. A, polyclonal antibody recognizing the N-terminal of the PDE6γ subunit [gift of the Fung Laboratory (UCLA, Los Angeles, CA, USA)]. Protein normalized to equivalent of 150 pmol rhodopsin. Lane 1, PDE6γ knockout mice (KO), Pde6gtm1/Pde6gtm1; lanes 2 and 3, different lines of T22A/T35A; lane 4, T22A; lane 5, WT. B, same as A, but protein normalized to 50 μg. Lane 1, homozygote Pde6gtm1/Pde6gtm1; lane 2, WT; lane 3, T35A. C, polyclonal antibody recognizing rod PDE6α and PDE6β as well as cone PDE6α′ (a generous gift of the Fung Laboratory). Protein normalized to 150 pmol rhodopsin. Lane 1, WT; lane 2, T35A. In 6.5% acrylamide gel, the PDE6α and PDE6β subunits are seen as a doublet. D, same as C but with 16% acrylamide gel and protein normalized to 50 μg. Lane 1, T22A; lane 2, WT; lane 3, T22A/T35A. E, 1D4 monoclonal antibody to opsin [a generous gift of the Molday Laboratory (University of British Columbia, Vancouver, Canada)]. Lane 1, WT; lane 2 T22A/T35A; lane 3, T35A; lane 4, T22A. F, polyclonal antibody sc-391 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) to GRK1 (rhodopsin kinase). Lane 1, T22A/T35A; lane 2, T35A; lane 3, T22A; lane 4, WT. G, polyclonal antibody, sc-391 (Santa Cruz Biotechnology) to GNAT1 (rod transducin α subunit). Lane 1, T22A/T35A; lane 2, T35A; lane 3, T22A; lane 4, WT. H, polyclonal antibody to the GTPase accelerating protein RGS9 [a gift of the Wensel Laboratory (Baylor College of Medicine, Houston, TX, USA)]. Lane 1, T22A/T35A; lane 2, T35A; lane 3, T22A; lane 4, WT. I, polyclonal antibody to guanylyl cyclase 2D [GUCY2D, a generous gift of the Yamazaki Laboratory (Wayne State University, Detroit, MI, USA)]. Lane 1, T22A/T35A; lane 2, T35A; lane 3, T22A; lane 4, WT.

Identification of transgenic mice

DNA was isolated from tail tips or liver samples by homogenizing the tissue, digesting extensively with proteinase K and extracting with phenol. DNAs were analysed by PCR. The DNAs were also digested by Sac I and analysed by Southern blot hybridization with a Pde6g cDNA probe. Additional restriction digests were performed to analyse the structure of the integrated sequences, and to ensure that the DNA flanking the transgene was intact. The standard nomenclature of the T35A transgenic mouse should be Tg(Pde6gT35A); Pde6gtm1/Pde6gtm1; hereafter referred to as T35A. The standard nomenclature of the T22A transgenic mouse should be Tg(Pde6gT22A); Pde6gtm1/Pde6gtm1; hereafter referred to as T22A. The standard nomenclature of the T22A-T35A transgenic mouse should be Tg(Pde6gT22A, T35A); Pde6gtm1/Pde6gtm1; hereafter referred to as T22A/T35A.

Isolation of rod outer segments (ROS)

Under dim red light conditions, ROS from dark-adapted mice were isolated in Hepes–phosphate balanced salt solution (4.09 mm NaH2PO4,148.4 mm NaCl, 4.91 mm KCl, 2.45 mm CaCl2, 1.23 mm MgSO4, and 4.7 mm Hepes at pH 7.2). Rhodopsin content was determined by the difference in absorbance at 500 nm before and after bleaching under a non-saturating halogen light source (Zimmerman & Godchaux, 1982).

Immunoblot analyses

Proteins from murine ROS (185 pmol rhodopsin per lane) or retinal extract were separated by electrophoresis on either a 6.5–9.5% acrylamide–1.5% crosslinker (for PDE6 α and β subunits and other phototransduction enzymes) or a 16% acrylamide–1.5% crosslinker (for the PDE6γ subunit) inverted gradient polyacrylamide gel previously described by Tsang and colleagues (Tsang et al. 1996). Proteins were then transferred to 0.2 μm PVDF membranes (Bio-Rad Laboratories, Hercules, CA, USA) overnight at 4 V cm−1 (Towbin et al. 1979). Membranes were blocked in 3% bovine serum albumin in 500 mm NaCl, 20 mm Tris, pH 7.6 and 0.1% Tween 20. Different phototransduction enzymes were detected with a polyclonal antiserum or monoclonal antibodies (see legend to Fig. 1). Western blots were visualized with the DuoLux Chemiluminescence substrate kit (Vector Laboratories, Inc., Burlingame, CA, USA) using a goat anti-rabbit IgG–alkaline phosphatase conjugate. Blots were exposed to Hyperfilm-MP (Amersham Pharmacia Biotech, Piscataway, NJ, USA) and pre-flashed to increase sensitivity and linearity according to the Sensitize protocol (Amersham Pharmacia Biotech, Piscataway, NJ, USA).

Suction-electrode recordings

Methods for recording responses of mouse rods have been given previously (Woodruff et al. 2002, 2003). Rods were perfused with 37°C physiological solution containing amino acids and nutrients. Data were filtered at 30 Hz (8-pole, Bessel) and sampled at 100 Hz. Flashes of 500 nm light 20 ms in duration were attenuated to different light levels by absorptive neutral density filters. At dim intensities, 10–20 individual responses presented at 5 s intervals were averaged to obtain the mean flash responses. At medium intensities 5–10 responses were averaged, and the interflash interval was increased to 10 s. At bright intensities above saturation for the rods, only 3–5 responses were averaged, and the interflash interval was increased to 15–20 s. Recordings always proceeded from dim intensities to brighter intensities, and the complete response–intensity data for an individual rod took about 20 min and bleached less than 0.5% of the visual pigment. The single-photon response was calculated from the squared mean and variance as in previous studies (see for example Chen et al. 2000; Tsang et al. 2006a). The method of data acquisition for the single photon responses is explained in the figure legends. Errors are given as standard errors of the mean (s.e.m.). The time course of PDE activity for the data of Fig. 2A–C was calculated from eqn (24) of Pugh & Lamb (1993); the rate of change of activity was then computed by fitting a straight line to the initial rising phase as in Tsang et al. (1998).

Figure 2. Recordings from WT, T35A and T22A rods.

Figure 2

Open in a new tab

A, suction electrode-recorded current responses of a single WT rod to 20 ms flashes of 500 nm light at intensities 17, 43, 160, 450 and 1120 photons μm−2. Each trace was averaged from 3 to 20 individual flashes. B, current responses of a single T35A rod to same flashes, with traces averaged from 5 to 39 flashes. C, current responses of a single T22A rod to same flashes, with traces averaged from 3 to 15 flashes. D, averaged responses (±s.e.m.) of WT rods (•, n= 32), T35A rods (▪, n= 37) and T22A rods (○, n= 32) to flashes at eight intensities from 1.3 to 4230 photons μm−2. Current records were obtained with 30 Hz 8-pole Bessel filtering at a sampling frequency of 100 Hz.

Histology

Mice were killed with an intraperitoneal injection of Nembutal. Each eye was rapidly removed, punctured at the 12 o' clock position along the limbus, and placed in a separate solution of 3% glutaraldehyde in phosphate-buffered saline. After fixation for 1–2 days, the eyes were washed with saline and the 12 o' clock limbal puncture was used to orientate the right and left eyes, which were kept in separate buffer, so that the posterior segment containing the retina could be sectioned along the vertical meridian. A rectangular piece containing the entire retina from superior to inferior ora serrata, including the optic nerve, was prepared for post-fixing in osmic acid, dehydration and epon embedding. A corner was cut out at the superior ora to allow identification of the upper retinal half of the segment. Sectioning proceeded along the long axis of the segment so that each section contained upper and lower retina as well as posterior pole. Semi-serial sections were stained with either haemotoxylin-eosin or toluidine blue, mounted and examined by light microscopy. For electron microscopy, thin sections were cut on a Reichert Ultracut ultramicrotome, stained with uranyl acetate, and examined with a JEOL 100CX transmission electron microscope (Tokyo, Japan). Negatives were scanned and dimensions measured in Photoshop (Adobe, San Jose, CA, USA).

Results

To test the effects of phosphorylation on PDE6γ function in living rod photoreceptors, transgenic lines expressing different phosphorylation mutant alleles were generated and backcrossed with Pde6gtm1/Pde6gtm1, which do not have any PDE6γ in their rods (Tsang et al. 1996). This gave animals expressing only the mutant PDE6γ. Immunoblots of rod outer segments (ROS) of each transgenic line, normalized by the amount of rhodopsin or total protein present, revealed that the amounts of PDE6γ (Fig. 1A and B) as well as of the catalytic PDE6 α and β subunits (Fig. 1C and D) were all within normal limits. This was also true of the phototransduction proteins we assayed, including opsin (Fig. 1E), rhodopsin kinase (Fig. 1F), transducin α (GNAT1, Fig. 1G), RGS9 (Fig. 1H), and guanylyl cyclase 2D (Fig. 1I). Assays of trypsin-activated PDE activity (Tsang et al. 1998) gave comparable values of between 0.8 and 1.5 pmol cGMP min−1 (ng protein)−1 for the various mouse lines. Our evidence therefore indicates that the expression of PDE as well as other transduction proteins was not affected by our substitution of the mutant for normal PDE6γ. Retinal morphology was also normal when either the T35A or T22A mutant PDE6γ was substituted for normal PDE6γ, as well as when both were substituted in the double mutant T22A/T35A animals, with no evidence of degeneration of the photoreceptors (see Fig. 1 in online Supplemental material). The lengths of rod outer segments were similar in WT and T22A/T35A animals but somewhat smaller in T22A and T35A (see below). Finally, immunolocalization experiments indicate that the PDE6γ is confined to the outer segments in WT and the various mutant mouse lines (K. M. Janisch & S. H. Tsang, unpublished observations).

Suction-electrode recording from mutant rods

In Fig. 2 we compare responses to a series of light flashes of three representative rods from WT, T35A and T22A animals. The T35A responses (Fig. 2B) were smaller in peak response amplitude and decayed much more slowly. The mean dark current was 11.8 ± 0.5 pA (n= 50) for WT and 7.5 ± 0.6 pA (n= 39) for T35A in the sample of rods from which we recorded; mean integration times were 169 ± 10 ms (WT) and 296 ± 20 (T35A). Both were significantly different at P < 0.01 (Student's t test). Removal of the T22 phosphorylation site had quite a different effect (see Fig. 2C). Responses were not significantly different in peak amplitude but decayed more rapidly than WT responses, particularly at dimmer light intensities.

In Fig. 2D we compare the averaged response–intensity curves of WT and the two mutant rods. The T35A rods were considerably less sensitive; half-saturation constants (I0) from a fit of the response–intensity curves of each cell to an exponential saturation function (Lamb et al. 1981) gave mean values of 33.5 ± 2.2 (WT) and 94.2 ± 10.6 (T35A) photons μm−2, which were significantly different (P < 0.01). The half-saturation constant for the T22A photoreceptors (mean I0= 35.3 ± 2.5 photons μm−2) was not statistically significantly different from WT. Small-amplitude flash sensitivities were 0.24 ± 0.01 pA photon−1μm−2 for WT (n= 49), 0.19 ± 0.01 pA photon−1μm−2 for T22A (n= 31) and 0.08 ± 0.01 pA photon−1μm−2 for T35A (n= 42) and were again significantly different for WT and T35A.

Comparison of wave form of WT and mutant rod responses

In Fig. 3 we show averaged recordings from many rods to flashes of the same three intensities, namely 17, 159 and 1120 photons μm−2. Responses for each rod were normalized to the peak amplitude of that rod; this is equivalent to calculating for each cell the percentage of channels that have closed as a function of time in response to the light. Figure 3A–C compares the mean normalized photocurrent of T35A responses (blue) with those of WT responses (black). Insets give the mean and s.e.m. at 10 ms intervals.

Figure 3. Averaged wave form of responses of WT and mutant rods.

Figure 3

Open in a new tab

A–C, light-induced currents of rods from T35A (blue traces) and WT (black traces) mice at light intensities of 17 (A), 160 (B) and 1120 (C) photons μm−2. The insets show responses for the first 120 ms after the flash (with standard error bars) on an expanded time scale. WT current traces were averaged from 35 (A), 35 (B) and 31 (C) rods; T35A traces from 23 (A), 24 (B) and 22 (C) rods. D–F, light-induced currents for T22A rods (red traces), T35A/T22A double mutant rods (green) and WT rods (black). Flash intensities in D–F were the same as those in A-C, respectively. WT currents are those used in A–C and have been repeated for ease of comparison. The T22A currents were averaged from 27 (D), 28 (E) and 25 (F) rods; T35A/T22A currents from 22 (D), 22 (E) and 22 (F) rods. Insets are as in A–C.

T35A responses were smaller, reflecting their decreased sensitivity (see Fig. 2D). Responses also decayed more slowly; this is most apparent at the two brighter intensities (Fig. 3B and C). Finally, the insets show that the T35A responses rose uniformly along a slower time course than the WT responses, indicating that removal of the threonine at this position had a large effect on the gain of transduction. We estimated the rate of change of light-activated PDE activity from the slope of the initial time course of the response as in Pugh & Lamb (1993) and Tsang et al. (1998). For the responses in Fig. 3A, the rate of change PDE activity was a factor of 7–8 smaller in T35A rods than in WT (9.3 versus 68 s−2), and similar though smaller changes were observed at brighter intensities (Fig. 3B, 99 versus 445 s−2; Fig. 3C, 443 versus 810 s−2).

It is conceivable that this difference could have been caused by a decrease in the expression of the PDE in the T35A mice, but the evidence from Fig. 1A–D and measurements of trypsin-activated PDE activity indicated that expression was similar in the two animals. A difference in the collecting area of the two rods could also in theory explain this difference, since if the T35A rod outer segments were smaller, they would absorb fewer photons, and this would give the appearance of a difference in PDE activation rate. To examine this possibility, we measured rod outer segment lengths and diameters of the different animals in the electron microscope (see Methods). WT and T22A/T35A rods were nearly the same in mean length (25.7 ± 0.5 μm and 25.2 ± 0.2 μm) and were both longer than either of the single phosphorylation mutants (T35A, 22.3 ± 0.4 μm; T22A, 18.4 ± 0.3 μm). The diameters of all the rod outer segments were similar and in the range 1.3–1.4 μm. The volume and collecting area of the T35A rod outer segments was thus somewhat smaller than of WT, but the difference was not large enough to account for the much slower rise time and PDE activation rate of the T35A receptors. Thus, the removal of the phosphorylation site in the T35A rod decreased the rate of activation of the enzyme, for example by altering the rate of binding of PDE6γ with TαGTP.

In Fig. 3D–F, we give a similar comparison for T22A rods (red), and for T22A/T35A double mutant rods (green). In contrast to T35A, the T22A responses decayed more rapidly than WT, though this difference is greatest at the dimmest intensity (Fig. 3D); at the brightest (Fig. 3F), only a small difference in wave form of decay was observed. The insets show that there is little or no difference in the initial wave form of the T22A and WT rods, indicating that the rate of activation of the PDE was unaffected. The addition of the T22A mutation on top of the T35A background (green traces) largely nulled the pronounced slowing of decay of T35A alone, but it only partially rescued the effect of the T35A mutation on the initial rate of PDE activation (compare Fig. 3D–F with A–C).

Single-photon responses and recovery from background illumination

The conclusions of Fig. 3 are re-enforced by the data in Fig. 4A. Here we show the mean single-photon responses for rods from the different animals, calculated from the squared mean and variance as in previous experiments (see for example Chen et al. 2000; Woodruff et al. 2003; Tsang et al. 2006a). Once again, the T35A response is smaller than WT and rises more slowly. T22A rods decay more rapidly and show an undershoot, clearly visible also in Fig. 3D.

Figure 4. Single-photon responses and recovery from background illumination.

Figure 4

Open in a new tab

A, averaged single-photon responses from T35A (blue, n= 24), T22A (red, n= 18) and T35A/T22A (green, n= 25) rods superimposed on averaged single-photon response from WT rods (black, n= 48). Single-photon responses were calculated for each rod individually from 15–60 dim flashes with the squared mean-variance method (Chen et al. 2000; Woodruff et al. 2003; Tsang et al. 2006a). Inset gives first 120 ms after flash on an expanded scale with s.e.m.B, rods were exposed to 2830 photons μm−2 s−1 for 4 min. Traces show responses beginning at time of turning off of illumination at t= 0 and have been averaged from 3 WT rods (black), 5 T35A rods (blue), 8 T22A rods (red) and 5 T22A/T35A rods (green). The mean peak amplitude (saturating) response of the rods used for the figure were (with s.e.m.) 13.1 ± 2.1 pA (WT),11.5 ± 1.1 pA (T35A), 13.0 ± 1.0 pA (T22A) and 11.8 ± 1.5 pA (T22A/T35A). Responses to the 2830 photons μm−2 s−1 illumination were initially 93.3 ± 2.7% (WT), 91.5 ± 4.5% (T35A), 95.8 ± 1.5% (T22A) and 93.1 ± 1.6% (T22A/T35A) of saturation, respectively. The responses of WT and of all the mutant rods declined somewhat during the 4 min light exposure, and at t= 0 just as the light was turned off were 53.2 ± 8.2% (WT), 83.8 ± 5.9% (T35A), 70.3 ± 8.6% (T22A) and 73.7 ± 4.8% (T22A/T35A) of saturation. After the adapting light was extinguished, flashes were presented every 5 s to test response kinetics. The responses of WT, T35A, T22A and T22A/T35A rods initially showed more rapid kinetics compared to the responses to the same flashes prior to the adapting light (Tsat reduced by 65–75%), which then recovered to > 90% of dark-adapted Tsat within 90 s (Krispel et al. 2003).

The results in Fig. 4B show that the T35A mutation also affects the recovery of current after exposure to background illumination. Here we exposed rods from the different animals to 4 min of a steady intensity of 2830 photons μm−2 s−1. The averaged responses of the cells are shown after the light was turned off at t= 0, normalized for each cell to the amplitude of the current just before the light was extinguished. After 4 min of illumination, the circulating current of the rods had reached steady state, at a value below saturation for all of the animals (see legend). When the light was turned off, recovery of T35A to baseline was slower and T22A more rapid than WT, in agreement with Fig. 3.

In a separate series of experiments (data not shown), we measured the change in response wave form to a bright flash after the 4 min exposures shown in Fig. 4B, to see if the slow recovery of time in saturation (Tsat) previously reported by Krispel and colleagues (Krispel et al. 2003) would be abolished by removal of one or both of the phosphorylation sites of the PDE6γ. In WT rods the light exposure produced a shortening of Tsat, which slowly recovered with a time constant within experimental error of the one previously reported by Krispel et al. All of the mutant rods also showed a shortening of Tsat and a similar slow time course of recovery, though we did not do a sufficient number of experiments to exclude the possibility that the recovery time constant of one or another of the mutations was subtly altered.

T35A slows rate-limiting time constant for rod decay

The data in Figs 3 and 4 suggest that the T35A mutation, in addition to affecting the kinetics of activation of PDE6, also alters the time course of deactivation. The data in Fig. 5 confirm this notion. In Fig. 5A and B, we show the recovery of the light response of a single WT and T35A rod to a series of brief flashes of increasing intensity. As others have previously shown (Pepperberg et al. 1992), rod responses decay with a similar time course but with a delay that increases with increasing illumination. This is equally true of the WT and T35A rods, but the time course of decay is markedly slower for the T35A rods, and the delay before the decay commenced was longer and grew more rapidly with flash intensity than for WT rods.

Figure 5. Increase in dominant time constant for response recovery of T35A rods.

Figure 5

Open in a new tab

A, representative WT rod showing current responses at intensities from just saturating to 10-fold greater than saturation. B, T35A rod as in A. C, average time necessary to recover 25% of circulating current after flash as a function of the natural log of the flash intensity for 6 WT rods and 10 T35A rods. Error bars give s.e.m. values. Straight lines are fits to the data from 645 to 4230 photons μm−2, and slopes give dominant time constant Td. See text.

In Fig. 5C, we have plotted Tsat, the time for decay of the response to a criterion level of 25% of the dark current (or 75% of the maximum response), as a function of the natural log of the light intensity. The means have been fitted with straight lines; their slopes give values for Td, the rate-limiting time constant of decay, of 179 ms (WT) and 461 ms (T35A). Tsat was plotted individually for each of the rods as functions of light intensity and fitted to obtain values for Td; these were then averaged to give 172 ± 7 ms (n= 48) for WT and 471 ± 34 ms (n= 33) for T35A, which are statistically different (P < 0.01). Td values for T22A and double mutant rods were also determined and were not significantly different from WT.

Discussion

Our results show that the substitution of alanine for threonine at the T35 position of the PDE6γ molecule has no observable effect on the expression of the PDE or other transduction proteins but a pronounced effect on the wave form of the rod light response. It decreases the rate of rise of the response by slowing the rate of activation of the PDE and slows the rate of response decay. It also decreases sensitivity, probably as a consequence of the decrease in the gain of transduction produced by the slowing in the rate of PDE activation; in addition it decreases the mean value of the dark current, for reasons that are unclear and were not further investigated. Substitution of threonine 22 with alanine has no significant effect on rise time, sensitivity, or dark current but accelerates response decay at least at dim flash intensities. Responses may undershoot the baseline, perhaps because the PDE activity in some cases can decline so rapidly that the change in cyclase velocity cannot track the change in PDE. The undershoot is unlikely to be caused by a change in cyclase expression, since this appears normal (see Fig. 1I). Finally, if both T22A and T35A substitutions are made in the same mouse rod, the rate of activation of the PDE remains somewhat slowed, but response decay is much as in WT animals.

One interpretation of our experiments is that the changes we observed are caused by alanine substitution itself and are unrelated to the putative phosphorylation of the threonines. This seems to us unlikely, since preliminary investigations with three different protein structure-predicting algorithms–PROF (Rost & Sander, 1993), NORSp (Liu & Rost, 2003) and Phyre (http://www.sbg.bio.ic.ac.uk/phyre/) – indicate that changes in structure produced by introducing alanines at either or both positions 22 and 35 are likely to be small. Since both of these threonines have been shown to be phosphorylated (Tsuboi et al. 1994a,b; Udovichenko et al. 1994; Xu et al. 1998; Hayashi et al. 2000; Matsuura et al. 2000; Paglia et al. 2002), it is more likely that the changes in wave form are due to removal of either a constitutive or light-dependent phosphorylation of PDE6γ. The addition of a phosphate group to this region of the PDE6γ molecule, which (between residues 20 and 40) contains seven basic residues (lysines and arginines), would be expected to have a large effect on the association of PDE6γ with other proteins in the transduction cascade. It is known that this region of the PDE6γ molecule contributes to the association of the PDE6γ both with Tα (Morrison et al. 1989; Artemyev et al. 1992; Takemoto et al. 1992) and with PDEαβ (Artemyev & Hamm, 1992; Mou & Cote, 2001; Guo et al. 2005). It is therefore possible that T35A alters the rate of activation by changing the rate of binding of PDE6γ with Tα or the rate of PDE6γ displacement from the catalytic sites of PDEαβ.

The effects of T35A on response decay appear to result from an influence of the mutation that is different from the one that alters activation, since in the double mutant animals the addition of T22A restores the decay of the T35A response nearly to that of WT rods, but the rise time of the double mutants remains significantly slower than WT (see Fig. 3). The results of Fig. 5 show that the T35A mutation produces a change in Td, the rate-limiting time constant of decay of the receptor response, which the results of Krispel et al. (2006) indicate to be the binding of RGS9-1 and/or hydrolysis of TαGTP to TαGDP by the associated GAP complex. PDE6γ is known to enhance the activity of the GAP complex, probably by increasing the affinity of TαGTP for RGS9/Gβ5 (Skiba et al. 2000). The T35A mutation may disrupt this enhancement, much as has been previously postulated for the PDE6γ W70A mutation (Tsang et al. 1998). The T22A mutation appears to accelerate the decay of the rod light response, but this effect is more difficult to interpret, since the acceleration was greater at dimmer than at brighter light intensities, and no statistically significant difference could be observed between the rate-limiting constants Td for the T22A and WT rods.

If the effects we have observed are the result of the removal of phosphorylation of PDE6γ, they argue for some role of phosphorylation in the production of the normal light response. Our findings should stimulate future investigations in defining the nature of PDE6γ phosphorylation and its function in the physiology of the rod.

Acknowledgments

We are grateful to Clyde K. Yamashita and Wen-Ho Lee for their assistance in the keeping of the PDE6γ mutant mice. This work was supported by the Foundation Fighting Blindness (D.B.F. and S.H.T.), Burroughs Wellcome Fund (S.H.T.), Hirschl Trust (S.H.T.), Dennis W. Jahnigen Award of the American Geriatrics Society (S.H.T.), Joel Hoffmann Foundation (S.H.T.), Schneeweis Stem Cell Fund (S.H.T.), Eye Surgery Fund of the Columbia College of Physicians and Surgeons (S.H.T.), Association of University Professors in Ophthalmology-Research to Prevent Blindness (S.H.T.), the Whiteley Center of the Friday Harbor Marine Laboratories (G.L.F.), and NIH grants K08 EY004081 (S.H.T.), EY02651 (D.B.F.), and EY01844 (G.L.F.).

Supplementary material

The online version of this paper can be accessed at:

DOI: 10.1113/jphysiol.2006.121772

http://jp.physoc.org/cgi/content/full/jphysiol.2006.121772/DC1 and contains supplemental material consisting of a figure and legend entitled:

Supplemental Figure 1

Retinal photomicrographs of WT and mutant mice

This material can also be found as part of the full-text HTML version available from http://www.blackwell-synergy.com

Supplemental data
jphysiol_2006.121772_index.html (729B, html)

References

  1. Artemyev NO, Hamm HE. Two-site high-affinity interaction between inhibitory and catalytic subunits of rod cyclic GMP phosphodiesterase. Biochem J. 1992;283:273–279. doi: 10.1042/bj2830273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Artemyev NO, Rarick HM, Mills JS, Skiba NP, Hamm HE. Sites of interaction between rod G-protein α-subunit and cGMP- phosphodiesterase γ-subunit. Implications for the phosphodiesterase activation mechanism. J Biol Chem. 1992;267:25067–25072. [PubMed] [Google Scholar]
  3. Chen CK, Burns ME, He W, Wensel TG, Baylor DA, Simon MI. Slowed recovery of rod photoresponse in mice lacking the GTPase accelerating protein RGS9-1. Nature. 2000;403:557–560. doi: 10.1038/35000601. [DOI] [PubMed] [Google Scholar]
  4. Guo LW, Grant JE, Hajipour AR, Muradov H, Arbabian M, Artemyev NO, Ruoho AE. Asymmetric interaction between rod cyclic GMP phosphodiesterase γ subunits and αβ subunits. J Biol Chem. 2005;280:12585–12592. doi: 10.1074/jbc.M410380200. [DOI] [PubMed] [Google Scholar]
  5. Hayashi F. Light-dependent in vivo phosphorylation of an inhibitory subunit of cGMP-phosphodiesterase in frog rod photoreceptor outer segments. FEBS Lett. 1994;338:203–206. doi: 10.1016/0014-5793(94)80365-x. [DOI] [PubMed] [Google Scholar]
  6. Hayashi F, Matsuura I, Kachi S, Maeda T, Yamamoto M, Fujii Y, Liu H, Yamazaki M, Usukura J, Yamazaki A. Phosphorylation by cyclin-dependent protein kinase 5 of the regulatory subunit of retinal cGMP phosphodiesterase. II. Its role in the turnoff of phosphodiesterase in vivo. J Biol Chem. 2000;275:32958–32965. doi: 10.1074/jbc.M000703200. [DOI] [PubMed] [Google Scholar]
  7. Krispel CM, Chen D, Melling N, Chen YJ, Martemyanov KA, Quillinan N, Arshavsky VY, Wensel TG, Chen CK, Burns ME. RGS expression rate-limits recovery of rod photoresponses. Neuron. 2006;51:409–416. doi: 10.1016/j.neuron.2006.07.010. [DOI] [PubMed] [Google Scholar]
  8. Krispel CM, Chen CK, Simon MI, Burns ME. Novel form of adaptation in mouse retinal rods speeds recovery of phototransduction. J Gen Physiol. 2003;122:703–712. doi: 10.1085/jgp.200308938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Lamb TD, McNaughton PA, Yau KW. Spatial spread of activation and background desensitization in toad rod outer segments. J Physiol. 1981;319:463–496. doi: 10.1113/jphysiol.1981.sp013921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Lem J, Applebury ML, Falk JD, Flannery JG, Simon MI. Tissue-specific and developmental regulation of rod opsin chimeric genes in transgenic mice. Neuron. 1991;6:201–210. doi: 10.1016/0896-6273(91)90356-5. [DOI] [PubMed] [Google Scholar]
  11. Liu J, Rost B. NORSp: Predictions of long regions without regular secondary structure. Nucl Acids Res. 2003;31:3833–3835. doi: 10.1093/nar/gkg515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Matsuura I, Bondarenko VA, Maeda T, Kachi S, Yamazaki M, Usukura J, Hayashi F, Yamazaki A. Phosphorylation by cyclin-dependent protein kinase 5 of the regulatory subunit of retinal cGMP phosphodiesterase. I. Identification of the kinase and its role in the turnoff of phosphodiesterase in vitro. J Biol Chem. 2000;275:32950–32957. doi: 10.1074/jbc.M000702200. [DOI] [PubMed] [Google Scholar]
  13. Morrison DF, Cunnick JM, Oppert B, Takemoto DJ. Interaction of the γ-subunit of retinal rod outer segment phosphodiesterase with transducin. Use of synthetic peptides as functional probes. J Biol Chem. 1989;264:11671–11681. [PubMed] [Google Scholar]
  14. Mou H, Cote RH. The catalytic and GAF domains of the rod cGMP phosphodiesterase (PDE6) heterodimer are regulated by distinct regions of its inhibitory γ subunit. J Biol Chem. 2001;276:27527–27534. doi: 10.1074/jbc.M103316200. [DOI] [PubMed] [Google Scholar]
  15. Paglia MJ, Mou H, Cote RH. Regulation of photoreceptor phosphodiesterase (PDE6) by phosphorylation of its inhibitory γ subunit re-evaluated. J Biol Chem. 2002;277:5017–5023. doi: 10.1074/jbc.M106328200. [DOI] [PubMed] [Google Scholar]
  16. Pepperberg DR, Cornwall MC, Kahlert M, Hofmann KP, Jin J, Jones GJ, Ripps H. Light-dependent delay in the falling phase of the retinal rod photoresponse. Vis Neurosci. 1992;8:9–18. doi: 10.1017/s0952523800006441. [DOI] [PubMed] [Google Scholar]
  17. Pittler SJ, Baehr W. Identification of a nonsense mutation in the rod photoreceptor cGMP phosphodiesterase β-subunit gene of the rd mouse. Proc Natl Acad Sci U S A. 1991;88:8322–8326. doi: 10.1073/pnas.88.19.8322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Pugh EN, Jr, Lamb TD. Amplification and kinetics of the activation steps in phototransduction. Biochim Biophys Acta. 1993;1141:111–149. doi: 10.1016/0005-2728(93)90038-h. [DOI] [PubMed] [Google Scholar]
  19. Rost B, Sander C. Prediction of protein secondary structure at better than 70% accuracy. J Mol Biol. 1993;232:584–599. doi: 10.1006/jmbi.1993.1413. [DOI] [PubMed] [Google Scholar]
  20. Sagoo MS, Lagnado L. G-protein deactivation is rate-limiting for shut-off of the phototransduction cascade. Nature. 1997;389:392–395. doi: 10.1038/38750. [DOI] [PubMed] [Google Scholar]
  21. Skiba NP, Hopp JA, Arshavsky VY. The effector enzyme regulates the duration of G protein signaling in vertebrate photoreceptors by increasing the affinity between transducin and RGS protein. J Biol Chem. 2000;275:32716–32720. doi: 10.1074/jbc.C000413200. [DOI] [PubMed] [Google Scholar]
  22. Takemoto DJ, Hurt D, Oppert B, Cunnick J. Domain mapping of the retinal cyclic GMP phosphodiesterase γ-subunit. Function of the domains encoded by the three exons of the γ-subunit gene. Biochem J. 1992;281:637–643. doi: 10.1042/bj2810637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A. 1979;76:4350–4354. doi: 10.1073/pnas.76.9.4350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Tsang SH, Burns ME, Calvert PD, Gouras P, Baylor DA, Goff SP, Arshavsky VY. Role of the target enzyme in deactivation of photoreceptor G protein in vivo. Science. 1998;282:117–121. doi: 10.1126/science.282.5386.117. [DOI] [PubMed] [Google Scholar]
  25. Tsang SH, Gouras P, Yamashita CK, Kjeldbye H, Fisher J, Farber DB, Goff SP. Retinal degeneration in mice lacking the gamma subunit of the rod cGMP phosphodiesterase. Science. 1996;272:1026–1029. doi: 10.1126/science.272.5264.1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Tsang SH, Woodruff ML, Chen CK, Yamashita CY, Cilluffo MC, Rao AL, Farber DB, Fain GL. GAP-independent termination of photoreceptor light response by excess γ subunit of the c-GMP-phosphodiesterase. J Neurosci. 2006 a;26:4472–4480. doi: 10.1523/JNEUROSCI.4775-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Tsang SH, Woodruff ML, Fain GL. A possible phosphorylation switch for PDE6γ. Invest Ophthalmol Vis Sci. 2006b;47 ARVO E abstract 3746. [Google Scholar]
  28. Tsuboi S, Matsumoto H, Jackson KW, Tsujimoto K, Williams T, Yamazaki A. Phosphorylation of an inhibitory subunit of cGMP phosphodiesterase in Rana catesbeiana rod photoreceptors. I. Characterization of the phosphorylation. J Biol Chem. 1994a;269:15024–15029. [PubMed] [Google Scholar]
  29. Tsuboi S, Matsumoto H, Yamazaki A. Phosphorylation of an inhibitory subunit of cGMP phosphodiesterase in Rana catesbeiana rod photoreceptors. II. A possible mechanism for the turnoff of cGMP phosphodiesterase without GTP hydrolysis. J Biol Chem. 1994 b;269:15016–15023. [PubMed] [Google Scholar]
  30. Tuteja N, Farber DB. γ-Subunit of mouse retinal cyclic-GMP phosphodiesterase: cDNA and corresponding amino acid sequence. FEBS Lett. 1988;232:182–186. doi: 10.1016/0014-5793(88)80413-7. [DOI] [PubMed] [Google Scholar]
  31. Udovichenko IP, Cunnick J, Gonzalez K, Takemoto DJ. Functional effect of phosphorylation of the photoreceptor phosphodiesterase inhibitory subunit by protein kinase C. J Biol Chem. 1994;269:9850–9856. [PubMed] [Google Scholar]
  32. Woodruff ML, Sampath AP, Matthews HR, Krasnoperova NV, Lem J, Fain GL. Measurement of cytoplasmic calcium concentration in the rods of wild-type and transducin knock-out mice. J Physiol. 2002;542:843–854. doi: 10.1113/jphysiol.2001.013987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Woodruff ML, Wang Z, Chung HY, Redmond TM, Fain GL, Lem J. Spontaneous activity of opsin apoprotein is a cause of Leber congenital amaurosis. Nature Genet. 2003;35:158–164. doi: 10.1038/ng1246. [DOI] [PubMed] [Google Scholar]
  34. Xu LX, Tanaka Y, Bonderenko VA, Matsuura I, Matsumoto H, Yamazaki A, Hayashi F. Phosphorylation of the γ subunit of the retinal photoreceptor cGMP phosphodiesterase by the cAMP-dependent protein kinase and its effect on the γ subunit interaction with other proteins. Biochemistry. 1998;37:6205–6213. doi: 10.1021/bi973087i. [DOI] [PubMed] [Google Scholar]
  35. Zimmerman WF, Godchaux W., 3rd Preparation and characterization of sealed bovine rod cell outer segments. Meth Enzymol. 1982;81:52–57. doi: 10.1016/s0076-6879(82)81011-2. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Figure 1

Retinal photomicrographs of WT and mutant mice

jphysiol_2006.121772_1.pdf (259.6KB, pdf)
Supplemental data
jphysiol_2006.121772_index.html (729B, html)
jphysiol_2006.121772_1.pdf (259.6KB, pdf)

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES