Supporting Online Material for
Parietal-Eye Phototransduction Components and Their Potential
Chih-Ying Su,* Dong-Gen Luo, Akihisa Terakita, Yoshinori Shichida, Hsi-Wen Liao,
Manija A. Kazmi, Thomas P. Sakmar, King-Wai Yau*
*To whom correspondence should be addressed. E-mail: firstname.lastname@example.org (C.-Y.S.);
Published 17 March, Science 311, 1617 (2006)
This PDF file includes:
Materials and Methods
Figs. S1 to S6
Tables S1 and S2
Supporting Online Material
Parietal-Eye Phototransduction Components and Their Potential Evolutionary
Chih-Ying Su1,*, Dong-Gen Luo1, Akihisa Terakita2, Yoshinori Shichida2, Hsi-Wen Liao1,
Manija A. Kazmi3, Thomas P. Sakmar3 and King-Wai Yau1*
1. Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore,
Maryland 21205, USA.
2. Department of Biophysics, Graduate School of Science, Kyoto University and Core Research
for Evolutional Science and Technology, Japan Science and Technology Agency, Kyoto 606-
3. Laboratory of Molecular Biology and Biochemistry, Rockefeller University, 1230 York
Avenue, New York, New York 10021, USA.
* To whom correspondence should be addressed:
Drs. Chih-Ying Su/ King-Wai Yau
Room 907 Preclinical Teaching Building
Johns Hopkins University School of Medicine
725 N. Wolfe St., Baltimore, MD 21205, USA
Phone: (410) 955-1260
Fax: (410) 614-3579
E-mail: email@example.com, firstname.lastname@example.org
Materials and Methods
Degenerate PCR and cDNA library screening. Degenerate PCR primers were designed
based on consensus sequences in vertebrate opsins. The successful primers were 5’-
CTGAACTACATC(C/T)TGGT(C/G)AACCT-3’ (forward, second transmembrane domain) and
5’-CGGAACTGTTTGTTCATGA(A/C/G)GAC-3’ (reverse, seventh transmembrane domain).
The cDNA libraries were constructed in Lambda ZAP II Vector (Stratagene) using poly(A)+
RNA from side-blotched lizard (Uta stansburiana) parietal and lateral eyes, respectively. From
degenerate PCR on the cDNA libraries, we found partial cDNA fragments of pinopsin in the
parietal eye, and RH2 in the lateral eyes. These degenerate-PCR products were used to
synthesize DNA probes for library screening, yielding full-length clones of pinopsin from the
parietal eye and of five rod and cone opsins (except for the long-wavelength-sensitive opsin,
which was 5’-truncated) from the lateral eyes. Parietopsin was isolated by library-screening with
zebrafish VAL opsin (coding region nts 779-1128 (S1)), which is divergent from other opsins.
For other phototransduction constituents, we used probes derived from DNA fragments
(~500 bp) for mouse rod transducin (Gt1α), Go, Gi, Golf, Gα12, Gα15, Gz, Gβ1, Gβ5, Gγ13,
PDEα’, PDEγ, human CNGA1, and rat RGS9, for low-stringency screenings of both cDNA
libraries. The GenBank accession numbers for all identified clones were summarized in table S2.
Sequence alignment and phylogenetic analysis. Sequence alignments were conducted
with the ClustalW Alignment program (MacVector 7.2, Accelrys). Phylogenetic analysis was
based on multiple protein sequence alignments by ClustalX (1.83) and performed with the
Molecular Evolutionary Genetics Analysis (MEGA), Version 3.0 (http://www.megasoftware.net)
using the Neighbor-Joining method on the Parsim-Info sites, with bacteriorhodopsin as an
outgroup. The tree was condensed at bootstrap probability ≤ 50%, based on 1000 replicates.
The accession numbers of the opsins in the analysis are as follows. Lizard (Uta
stansburiana) RH1, DQ100323; Human (Homo sapiens) rhodopsin, AAC31763; Lizard RH2,
DQ100324; Chicken (Gallus gallus) green opsin, AAA48786; Lizard SWS2, DQ100326; Human
blue opsin, P03999; Lizard SWS1, DQ100325; Mouse (Mus musculus) blue opsin, P51491;
Lizard pinopsin, DQ100321; Chicken pinopsin, 2022322A; Lizard LWS, DQ129866; Mouse
green opsin, 035599; Human red opsin, P04000; Human green opsin, P04001; Salmon (Salmo
salar) VA opsin, AAC60124; Zebrafish (Danio rerio) VA opsin, BAA94289; Catfish (Ictalurus
punctatus) parapinopsin, AAB84050; Lamprey (Lethenteron japonicum) parapinopsin,
BAD13381; Xenopus (Xenopus tropicalis) parapinopsin, BAD17960; Lizard parietopsin,
DQ100320; Xenopus parietopsin, DQ284780; Zebrafish parietopsin, XM_681591; Human
encephalopsin, AAH36773; Mouse RGR, AAC69836; Human OPN5, Q6U736; Mouse peropsin,
AAC53344; Mouse melanopsin, AAF24979; Bacteriorhodopsin, AAA72504.
Immunohistochemistry. We generated specific antibodies against lizard pinopsin,
parietopsin, cone transducin-α, and gustducin-α. Peptides corresponding to the last 20 residues of
lizard pinopsin and parietopsin, and residues 97-116 of cone transducin-α and gustducin-α (see
fig. S1 and fig. S4) were synthesized (Princeton Biomolecules). Thyroglobulin-conjugated
peptides were used to immunize rabbits (for pinopsin, cone transducin-α, and gustducin-α) or
rats (for parietopsin) (Covance Research Products). Rabbit antisera were purified by peptide-
affinity columns and diluted 1:10 (pinopsin and gustducin-α) or 1:100 (cone transducin-α) before
use for immunostaining. Rat antiserum against parietopsin or the preimmune serum was diluted
1:100 before use. The specificity of the antibodies against lizard pinopsin and parietopsin was
further verified by immunostaining on transfected HEK293 cells (fig. S3 and legend).
The isolated parietal and lateral eyes were fixed in 4% paraformaldehyde at 4°C for 2
hours, and cryoprotected in 30% sucrose before being embedded and frozen in OCT (Tissue-
Tek). Cryostat sections (5 µm for parietal eye and 10 µm for lateral eye) were blocked (1X PBS/
5% normal goat serum/ 0.3% Triton X-100) for 1 hour at room temperature, then incubated with
the primary antibody overnight at 4°C. Besides our own primary antibodies, we used others,
including: rhodopsin (B630 and 1D4), 1:500 dilution; chicken green opsin (CERN874, which
recognizes lizard RH2 (S2)), 1:1000; transducin (TF15, Cytosignal), 1:500; Go (K-20, Santa Cruz
Biotechnology), 1:500; human Gβ1 (β-636), Gβ3 (β-638) (S3) and bovine cone Gγ (γ-5893-AS)
(S4), 1:300; bovine Gγ1 (PAB-00901-G, Cytosignal), 1:500; rod PDEγ (#9710 from R. Cote),
1:300; mouse Gγ13 (#4897 from R.R. Reed), 1:200. Labeled sections were washed three times in
1X PBS/0.3% Triton X-100, then incubated with the secondary antibodies for 1 hour at room
temperature. After three more washes in 1X PBS, the sections were mounted with Vectashield
medium containing DAPI nucleus stain (Vector Laboratories) and examined by confocal
Absorption spectrum. The coding region of lizard pinopsin or parietopsin with a C-
terminal 1D4 tag (last 9 amino acids of bovine rhodopsin) was introduced into a eukaryotic
vector. 400-µg plasmid DNA was used to transfect 40 dishes (100-mm diameter) of HEK293s
cells with the calcium-phosphate method. Two days after transfection, the cells were collected
by centrifugation and incubated in darkness with excess 11-cis-retinal overnight to reconstitute
the pigment. The pigment was extracted with 1% (w/v) dodecyl β-D-maltoside (DM) in buffer A
(50 mM HEPES, pH 6.5; 140 mM NaCl), bound to 1D4-agarose, washed with buffer B (0.02%
DM in buffer A), and eluted with 650-µl buffer B containing the C-terminus nona-peptide of
bovine rhodopsin. Absorption spectrum of purified parietopsin was recorded at 0ºC with a
Shimazu UV2400 spectrophotometer, as described (S5).
Whole-cell recordings. Parietal eyes were dissected from the pithed lizard under infrared
light and subjected to a series of enzymatic treatments, including collagenase (CLS1, 2.5 mg/ml;
Worthington) and pronase (2 mg/ml; Boehringer) in CO -independent medium (CIM; Life
Technologies), followed by
trypsin (2 mg/ml; Sigma) in Ca -free medium, as described (S6).
The tissue was triturated gently with glass pipettes with progressively smaller diameters (500−50
µm). The dissociated cell suspension was plated on a glass coverslip pretreated with
concanavalin A (Sigma). Owing to the technical difficulties from the small cell sizes (4-6 µm in
soma diameter), gigaseals were made under dim microscope light (S6). After 10 min in darkness,
membrane breakthrough was made to achieve whole-cell recording. Membrane potential was
clamped at −45 mV. The recording chamber was continuously perfused with CIM at 0.7−0.8
ml/min at room temperature. Mastoparan or Mas-17 (Calbiochem) was dissolved in pseudo-
intracellular solution in the recording pipette (in mM): 10 KCl, 120 K-gluconate, 5 MgCl , 1 Na-
EGTA, 10 K-HEPES, 3 Na ATP, 1 Na GTP, pH 7.4)
. The light stimuli were unattenuated white
(uncalibrated) or monochromatic light at 480 nm (4.1 × 10 photons µm ) or 520 nm (9.6 × 10
photons µm ), 1 sec in duration in all cases.
Parietal- and lateral-eye-specific opsins
PCR analysis of the lizard parietal-eye cDNA library with specific primers indicated that
only pinopsin and parietopsin, but not the lateral-eye opsins, were expressed in the parietal eye
(fig. S2, left panel). Likewise, the lateral eyes expressed only the rod and cone opsins (fig. S2,
right panel). Interestingly, in the course of our cDNA library screening, we also found one partial
RH2 clone from the parietal-eye cDNA library, which seemingly contradicted with the above
PCR result. We reasoned that the expression level of RH2 might be too low in the parietal eye to
be detected in our PCR condition (40 cycles of amplification). To investigate whether RH2 was
translated in the parietal eye, we stained the parietal-eye sections with a polyclonal antibody that
labels RH2 of lizard origin (S2). Although we observed positive signals in the lateral-eye
photoreceptors, we did not see any labeling in the parietal eye (not shown). Because our PCR
aimed to demonstrate the dominant opsin species, we did not exclude that some lateral-eye opsin
genes might be transcribed at a very low level in the parietal eye. This could also explain the
discrepancy between our results and an earlier study on American chameleon opsins (S7). The
authors performed Southern blotting of RT-PCR products with a mixture of opsin probes, with
which they detected the expression of pinopsin and three other lateral-eye opsins (SWS1, SWS2
and LWS) mRNAs in the parietal eye. Nonetheless, it was not reported whether these PCR
products had the appropriate sequences and whether full-length mRNAs were present in the
λmax with A2 chromophore
It has been reported that some lizards may use A2 chromophore (11-cis-3,4-
didehydroretinal) or a mixture of A1 and A2 chromophores for their pigments (S8, S9). For a
given opsin, the A2 pigment tends to have a λmax at somewhat longer wavelength than the A1
pigment (S10). If the side-blotched lizard in our study does use A2-retinal in the parietal eye, the
λmax of parietopsin in the native photoreceptor will shift from 522 nm to 550 nm (data not
shown). Correspondingly, the λmax of pinopsin may shift from 482 nm to 499-503 nm (S11).
Inhibition of cGMP-phosphodiesterase by Go
How Go acts in the parietal-eye photoreceptor is still unclear. Because activated Goα also
binds to PDEγ but does not activate PDE (S12), Goα may act as a competitive inhibitor of
gustducin-α. The affinity between gustducin-α and PDEγ is unknown, but that between its close
relative, transducin-α, and PDEγ is high (dissociation constant Kd < 0.1 nM), much higher than
between Goα and PDEγ (Kd ~ 1 µM), at least in vitro (S12). Moreover, there appears to be a
relative abundance of gustducin-α over Goα, at least at the cDNA level (not shown). Thus, a
competitive inhibition by Goα is perhaps unlikely. An alternative mechanism would be for Go to
act via another effector. PDEγ has been shown to be a substrate for phosphorylation, with the
phosphorylated form showing a much weaker affinity for activated transducin (S13-S17). Thus,
potentially, Go can antagonize gustducin by inducing phosphorylation of PDEγ.
Gustducin but not transducin in the parietal-eye photoreceptor
Why does the parietal-eye photoreceptor use gustducin instead of transducin for
phototransduction? In the parietal-eye photoreceptor, which relies on the inhibition of a
phosphodiesterase instead of the activation of a guanylate cyclase to elevate cGMP and produce
a depolarizing response (S6), the phosphodiesterase has to be active in darkness. In principle, the
basal phosphodiesterase activity may originate from spontaneous activity at any stage in the
phototransduction cascade: phosphodiesterase, G protein or the pigment. In the parietal-eye
photoreceptor, a basal phosphodiesterase activity driven by an active G protein in darkness has
been found (S6), thus implicating the pigment or the G protein as the source of constitutive
activity. Because spontaneous thermal activation appears to be predominant for long-
wavelength-sensitive pigments (S18) and pinopsin is a short-wavelength-sensitive pigment, the
thermal isomerization of pinopsin is perhaps too week for driving the basal phosphodiesterase
activity. This leaves the G protein as the source. As it turns out, gustducin appears to have higher
basal GTP-binding activity than transducin (S19), making it appropriate for the task.
Figure S1. Parietal-eye opsins. Sequence alignment of lizard pinopsin (POP) and parietopsin
(PtOP). Identical residues are shaded in gray. Predicted transmembrane domains (TM1-7) are
overlined. Conserved residues for chromophore linkage (*) and disulfide-bond formation (^) are
indicated. The position corresponding to the counterion, E113, in bovine rhodopsin is shaded in
pink. An uncharged glutamine is found in parietopsin at this site. The yellow-shaded residues at
the C-termini indicate the peptide immunogens used for generating specific antibodies.
Figure S2. PCR analysis of lizard parietal-eye and lateral-eye opsins. Expected PCR fragment
sizes are: parietopsin, 450 bp; pinopsin, 346 bp; RH1, 708 bp; RH2, 526 bp; SWS1, 474 bp;
SWS2, 539 bp; LWS, 451 bp. The PCR products smaller than the expected sizes are non-
specific. Parietal and lateral eyes expressed non-overlapping sets of opsin molecules.
Figure S3. Immunostaining on HEK293 cells transfected with either N-terminal His-tagged
pinopsin (His-Pinopsin) or parietopsin (His-Parietopsin) to verify respective antibody specificity.
A GFP construct was co-transfected as a positive control for transfection. Positive Cy3
fluorescence signals were in red, and the nuclei were counter-stained with DAPI in blue. Anti-
His-tag antibody (α-His tag) labeled cells transfected with both N-His-tagged pinopsin and
parietopsin (left panels), whereas anti-pinopsin antibody or anti-parietopsin antiserum only
recognized the respective transfected protein (middle and right panels; transfected cells from the
same dish as those shown in the left panels). Scale bar = 10 µm.
Figure S4. Sequence alignment of lizard gustducin-α (Ggustα) found in the parietal eye, and
lizard cone (Gt2α) and rod (Gt1α) transducin-α subunits found in the lateral eyes. Identical and
similar residues are shaded in dark and light gray, respectively. The N-terminal sequence of Gt1α
was incomplete and thus indicated by the dashed line. The most divergent region between Ggustα
and Gt2α (yellow-shaded) was used as peptide-immunogen to generate specific antisera for each
Figure S5. Double-immunostaining of lateral-eye sections with antibodies against
transducin-α (TF15) and Goα (K20), the same antibodies used in Fig. 2C. TF15 labeled
transducin-α throughout the photoreceptors (OS/IS, ONL and OPL), whereas Goα signals were
found mainly in the expected bipolar cell bodies (INL) and its processes (OPL and IPL). This
segregation of signals indicates that each antibody specifically recognizes its respective
immunogen of lizard origin. Abbreviations: OS/IS, outer segment and inner segment; ONL,
outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform
layer; GCL, ganglion cell layer; PRL, photoreceptor layer. Scale bars=10 µm.
Figure S6. Scheme for parietal-eye phototransduction. The hyperpolarizing light response is
mediated by the blue-sensitive pinopsin, which activates a cGMP-phosphodiesterase via
gustducin to lower cGMP concentration and close CNG channels. The depolarizing light
response is mediated by the green-sensitive parietopsin, which inhibits the phosphodiesterase via
Go to elevate cGMP and open CNG channels.
Table S1. Percentage amino-acid identities from pair-wise alignments of lizard (Uta)
parietopsin, pinopsin and rod/cone opsins (except for LWS, because it is 5’-truncated) with
parapinopsin and VA from other species. The LWS opsin from another lizard (Anolis) was
included. Abbreviations: PtOP, parietopsin; PpOP, parapinopsin; VA, vertebrate ancient opsin;
LWS, long-wavelength-sensitive opsin; SWS, short-wavelength-sensitive opsin; RH, rhodopsin.
GenBank accession numbers are the same as those shown in Fig. 1A (see Materials and
Table S2. Molecular components for lizard parietal- and lateral-eye phototransduction. The
molecules were identified by cDNA library screening (with GenBank accession numbers shown
in parentheses) and/or immunocytochemistry (§). Note that some molecules, although originally
found in the cDNA library, are not expressed in the photoreceptors (*). Abbreviations: RH,
rhodopsin; SWS, short-wavelength-sensitive opsin; LWS, long-wavelength-sensitive opsin;
Ggustα, gustducin-α; Gt1α, rod transducin-α; Gt2α, cone transducin-α; PDE, phosphodiesterase;
CNG: cyclic-nucleotide-gated channel; RGS, regulator of G-protein signaling.
Parietal eye Lateral eye
Gγ13 (DQ100315) #
# Using mouse Gγ13 cDNA as probe, we also identified this subunit in both parietal- and
lateral-eye cDNA libraries, and it showed 86% identity in amino acid-sequence to mouse
Gγ13. Gγ13 has previously been found to co-localize with Gβ3 in gustducin-positive taste
receptor cells (S21) and in retinal ON bipolar cells (S22), which express Go (S22, S23). An
antibody against mouse Gγ13 did not label the parietal-eye photoreceptor, but this result was
inconclusive because the same antibody also did not label the bipolar cells of the lizard
lateral eye (data not shown).
Supplementary References Download full-text
S1. D. Kojima, H. Mano, Y. Fukada, J. Neurosci. 20, 2845 (2000).
S2. M. Pasqualetti et al., Eur. J. Neurosci. 18, 364 (2003).
S3. R. H. Lee, B. S. Lieberman, H. K. Yamane, D. Bok, B. K. Fung, J. Biol. Chem. 267,
S4. O. C. Ong et al., J. Biol. Chem. 270, 8495 (1995).
S5. A. Terakita et al., Nat. Struct. Mol. Biol. 11, 284 (2004).
S6. W.-H. Xiong, E. C. Solessio, K.-W. Yau, Nat. Neurosci. 1, 359 (1998).
S7. S. Kawamura, S. Yokoyama, Vision Res 37, 1867 (1997).
S8. I. Provencio, E. R. Loew, R. G. Foster, Vision Res. 32, 2201 (1992).
S9. I. Provencio, R. G. Foster, Neurosci. Lett. 155, 223 (1993).
S10. F. I. Harosi, Vision Res. 34, 1359 (1994).
S11. S. Kawamura, S. Yokoyama, Vision Res. 38, 37 (1998).
S12. A. Otto-Bruc, T. M. Vuong, B. Antonny, FEBS Lett. 343, 183 (1994).
S13. S. Tsuboi et al., J. Biol. Chem. 269, 15024 (1994).
S14. I. P. Udovichenko, J. Cunnick, K. Gonzalez, A. Yakhnin, D. J. Takemoto, Biochem. J.
317 ( Pt 1), 291 (1996).
S15. L. X. Xu et al., Biochemistry 37, 6205 (1998).
S16. I. Matsuura et al., J. Biol. Chem. 275, 32950 (2000).
S17. M. J. Paglia, H. Mou, R. H. Cote, J. Biol. Chem. 277, 5017 (2002).
S18. F. Rieke, D. A. Baylor, Neuron 26, 181 (2000).
S19. M. A. Hoon, J. K. Northup, R. F. Margolskee, N. J. Ryba, Biochem. J. 309 (Pt 2), 629
S20. L. Ruiz-Avila et al., Nature 376, 80 (1995).
S21. L. Huang et al., Nat. Neurosci. 2, 1055 (1999).
S22. L. Huang et al., J. Comp. Neurol. 455, 1 (2003).
S23. N. Vardi, J. Comp. Neurol. 395, 43 (1998).