Cnga3

Cyclic nucleotide-gated channel alpha-3 (CNG channel alpha-3) (CNG-3) (CNG3) (Cone photoreceptor cGMP-gated channel subunit alpha-3) (Gustatory cyclic nucleotide-gated cation channel) (CNGgust)

Spatial tier: RESTRICTED · Class: ION_CHANNEL · Top region: MEPO (Median preoptic nucleus) · Positive regions: 21/554

46.6Discovery Score
40.1Targetability Score
CEvidence grade C
What this means: Cnga3: fine CCF profile is restricted toward MEPO (top regions: MEPO, SFO, AVPV, OV, IPR; 21 positive regions); direct spatial validation is still required.

Function in MEPO: evidence, prediction, innovation

Gene function
CNGA3 is the principal alpha subunit of the cone photoreceptor cyclic-nucleotide-gated channel, a non-selective cation channel opened directly by cGMP binding rather than by voltage. In cones it carries the dark current, and human CNGA3 mutations cause achromatopsia. The same subunit also forms cyclic-nucleotide-gated channels in olfactory and gustatory tissue.
Region function
The median preoptic nucleus is the integrative node of the lamina terminalis thirst and thermoregulation circuit, combining osmotic and angiotensin signals from the subfornical organ and OVLT with skin and core temperature information to drive drinking, sodium appetite, heat loss and sleep.
Published in this region
direct Direct at the level of the preoptic area. Feketa et al. 2020 compared mice with hibernating thirteen-lined ground squirrels and found that mice have a larger proportion of cold-sensitive neurons in the preoptic area of the hypothalamus, that these mouse cold-sensitive POA neurons express CNGA3, and that mouse but not squirrel CNGA3 is potentiated by cooling, identifying CNGA3 as a hypothalamic cold sensor and a molecular marker for the thermoregulatory circuit. The study sampled the preoptic area broadly rather than isolating the median preoptic nucleus.
Predicted function
Prediction: within the lamina terminalis, CNGA3 is predicted to be the conductance that makes median preoptic neurons intrinsically cold-sensitive, so that local brain cooling can trigger heat-conserving and heat-generating responses independently of skin afferent input. Conditional Cnga3 deletion restricted to MEPO glutamatergic neurons is predicted to abolish the intrinsic cold response of these cells in slices, blunt brown-adipose thermogenesis and vasoconstriction during local preoptic cooling, and widen the thermoneutral zone in vivo, while leaving skin-driven warm-defence responses intact; chemogenetic activation of Cnga3-expressing MEPO neurons should conversely drive hyperthermia.
Innovation
★★★☆☆ 3/5 The key discovery has been made at preoptic-area resolution, but nobody has tested CNGA3 specifically in the median preoptic nucleus or manipulated it conditionally in vivo, leaving the causal thermoregulatory experiment open.
Tools
Cnga3 knockout mice (cpfl5 and targeted alleles, widely available from achromatopsia research) and Cnga3 floxed lines; CNG channel blockers L-cis-diltiazem and pseudechetoxin; cGMP analogues 8-Br-cGMP and CPT-cGMP; anti-CNGA3 antibodies; Slc17a6-Cre, Adcyap1-Cre and Ptger3-based preoptic drivers plus telemetric core-temperature and brown-fat thermography readouts.
References

Direct evidence located at the region

No Allen ISH section could be located at this region's centroid for the recorded experiments.

Direct spatial evidence

MERFISH REGION LOCATOR
MEPO MERFISH region locator
Real single-cell MERFISH coordinates for the MEPO (Median preoptic nucleus) section -- our own ground-truth data, showing exactly where this region sits, independent of whether the Allen ISH section below is easy to read.

Regional expression figure

Cnga3 regional expression profile

Predicted WMB × fine-CCF profile; not direct full-transcriptome spatial measurement.

Spatial profile

MEPO
1.23
SFO
0.921
AVPV
0.721
OV
0.612
IPR
0.461
PVpo
0.442
TRS
0.432
DMH
0.419
VMPO
0.406
IPI
0.399
PVH
0.382
RCH
0.34
IPC
0.339
IPL
0.334
PVp
0.329
PVa
0.304
MPN
0.304
ARH
0.304
SBPV
0.302
SF
0.29

Evidence and specificity

Evidence status
PREDICTED_FINE
Direct sources
None; predicted localization only
Exclusive threshold
12 positive regions out of 554
Spatial specificity
0.351
Top-region fraction
0.081
Filter status
PASS

Interpretation limits