SLAC1 (Q9LD83) as a Potassium Channel Inhibitor (KAT1/AKT2) — Hypothesis Evaluation

Gene: SLAC1 (Arabidopsis thaliana, UniProt Q9LD83, AT1G12480)
Term under test: GO:0019870 potassium channel inhibitor activity (MF)
Seed reference: Zhang et al. 2016, Plant Cell 28(4):949-955 (PMID:27002025)
Hypothesis slug: slac1-kat1-inhibition

Executive Judgment

Verdict: Partially supported (weak–moderate), not independently replicated.

The seed hypothesis faithfully represents what Zhang et al. 2016 report, and that
single paper provides genuinely direct evidence for the molecular activity:
SLAC1 physically interacts with KAT1, KAT2 and AKT2 (Y2H, co-IP) and, in Xenopus
oocyte voltage-clamp, functionally inhibits the inward-K⁺ (K⁺in) channels KAT1
and AKT2 but not KAT2, with guard-cell gain-of-function (overexpression)
reducing K⁺in currents and light-induced stomatal opening. This is sufficient to
qualify SLAC1 as a candidate potassium-channel inhibitor (GO:0019870).

However, the stronger clauses of the hypothesis — "independently replicated or
corroborated and not contradicted by later work" — are not met:

1. No independent functional replication. Searching the 45 PubMed "cited-by"
   records (2017–2026) found no independent lab reproducing SLAC1→KAT1/AKT2
   patch-clamp inhibition. Citations are predominantly reviews or tangential.
2. Database silence on the K⁺-channel interaction. IntAct curates 19 SLAC1
   interactions, but all are signaling kinases/phosphatases (CPK6/21/23,
   PP2CA, SnRK2s/OST1) — none with KAT1, AKT2 or KAT2. The Zhang 2016 channel
   interaction has not been captured by interaction databases.
3. The term is not currently annotated — this is an ADDITION. UniProt/TAIR/
   QuickGO carry no potassium-channel inhibitor (GO:0019870) or regulator
   (GO:0015459) MF term for SLAC1; current MF terms are anion-channel activity and
   kinase/phosphatase binding. So the curation question is whether to add a new
   MF annotation on single-study evidence, not whether to retain one.
4. Competing/contradicting in-planta directionality. Laanemets et al. 2013
   (PMID:23126621) show that slac1 loss-of-function reduces guard-cell K⁺in
   and slows opening — the opposite of the "relief of inhibition" predicted by a
   direct-inhibition model — and attribute it to elevated cytosolic [Ca²⁺] and
   increased Ca²⁺ sensitivity of K⁺in channels. The physiological role therefore
   rests mainly on gain-of-function overexpression.
5. Conceptually corroborated. The broader principle of S-type anion channels
   regulating K⁺ channels by protein–protein interaction is independently
   supported by analogous pairs (SLAH3–GORK/SKOR; ALMT12–SLAC1), lending indirect
   plausibility but not specific replication.

Curation bottom line: GO:0019870 is defensible as a non-core MF annotation
with explicit evidence/qualifier caveats (single study, heterologous oocyte
assay + overexpression), but it must not be presented as SLAC1's core function,
which remains S-type anion (Cl⁻/malate/NO₃⁻) channel activity driving stomatal
closure.

Evidence Matrix

GO Curation Implications

• Current state (verified this run): GO:0019870 is not currently annotated to
  SLAC1 in UniProt/TAIR/QuickGO. The decision is therefore whether to ADD a new
  MF term on single-study evidence — a higher bar than retaining an existing one.
• Term GO:0019870 (potassium channel inhibitor activity, MF): Supportable only
  as a cautious, explicitly-caveated non-core lead. There is direct functional
  evidence (oocyte voltage-clamp inhibition of KAT1 and AKT2) plus physical
  interaction. Appropriate evidence code would be IDA (functional, from heterologous
  assay) optionally paired with IPI (physical interaction). Mandatory caveats:
  single study, single lab, heterologous oocyte system, gain-of-function
  overexpression, no independent replication, no database curation of the
  interaction, and a directionally opposite in-planta loss-of-function phenotype.
  Note it does not inhibit KAT2 despite binding it. A conservative curator could
  reasonably decline the addition pending independent replication.
• Do not treat as core function. SLAC1's primary MF is S-type anion channel
  activity (voltage-gated anion/chloride channel; the gene drives stomatal
  closure). The K⁺-channel-inhibitor activity is a secondary/regulatory
  "moonlighting" activity. If the review is deciding whether GO:0019870 should be
  a prominent/core term, the answer is no — keep it, but mark non-core.
• Alternative/parent term consideration: GO:0015459 (potassium channel
  regulator activity) is the more conservative parent if a curator judges
  "inhibitor" too strong given the single-study, heterologous basis. Given the
  data do show inhibition (not activation), GO:0019870 is biologically correct;
  the conservatism concern is about evidential strength, not direction.
• Associated BP: negative regulation of stomatal opening / regulation of
  stomatal movement is reasonable but is largely a downstream/physiological
  outcome and is confounded by the Laanemets loss-of-function result.

Mechanistic Scope

• Immediate molecular activity being tested: SLAC1 protein binding to Shaker
  inward-K⁺ channel subunits (KAT1, AKT2) and reduction of their K⁺in conductance
  — i.e., a channel–channel inhibitory protein–protein interaction.
• Directly supported (in the one study): physical interaction (Y2H/co-IP) and
  current inhibition in oocytes.
• Downstream / inferred, not the molecular activity itself: suppression of
  light-induced stomatal opening; drought/ABA physiology. These are pathway-level
  consequences shown mainly by overexpression and are confounded in loss-of-function
  by Ca²⁺-sensitivity feedback.
• Selectivity note: binding ≠ inhibition. SLAC1 binds KAT2 in Y2H yet does not
  inhibit it; this dissociation argues for a specific functional, not merely
  steric/co-localization, effect on KAT1/AKT2 — a point in the hypothesis's favor.

Conflicts and Alternatives

1. Directional contradiction (Laanemets 2013, PMID:23126621). Loss of SLAC1
   lowers, not raises, K⁺in. Explained by elevated cytosolic [Ca²⁺] increasing
   Ca²⁺-sensitivity-mediated K⁺in inhibition — an alternative mechanism that
   does not require direct SLAC1–KAT1 contact.
2. Kinase-route alternative (Acharya 2013, PMID:24033256). OST1 interacts with
   both SLAC1 and KAT1 and inhibits K⁺in; K⁺in suppression during closure may be
   dominated by kinase signaling rather than direct anion-channel binding.
3. Autoinhibition reinterpretation (Qin 2024, PMID:38985761). The SLAC1 N/C
   termini that Zhang 2016 implicate in KAT1 binding are the same cytosolic
   regulatory domains that mediate SLAC1 autoinhibition; constructs/overexpression
   effects attributed to KAT1 binding deserve scrutiny.
4. Heterologous-system artifact. Oocyte overexpression of a membrane protein can
   non-specifically depress co-expressed channel currents (membrane crowding,
   trafficking competition); KAT2 serving as a negative control mitigates but does
   not eliminate this.
5. Paralog/term carry-over risk: SLAH3 (SLAC1 homolog) inhibits different K⁺
   channels (GORK/SKOR). Curators should avoid propagating GO:0019870 across the
   SLAC/SLAH family by homology — the specific K⁺ targets differ per paralog.

Knowledge Gaps

• Independent functional replication (highest priority). Checked: 45 citing
  papers; none reproduce SLAC1→KAT1/AKT2 patch-clamp inhibition independently.
  Resolution: an independent guard-cell patch-clamp or oocyte study.
• Clean loss-of-function test. Does slac1 enhance K⁺in once the Ca²⁺ feedback
  is clamped out (Laanemets' <10 nM [Ca²⁺] condition)? This is the decisive in-planta
  test of direct inhibition and has not been reported.
• Native interaction evidence. IntAct has no SLAC1 records; no reported
  endogenous co-IP from guard-cell tissue (as opposed to oocyte overexpression).
• AKT2 specificity. Full-text confirms AKT2 K⁺in inhibition (Suppl. Fig. 2), but
  this rests on one supplemental figure; independent confirmation absent.
• Structural basis. No structure of a SLAC1–KAT1 complex; unknown whether
  binding occludes the pore/voltage sensor or alters gating/trafficking.

Discriminating Tests

1. Ca²⁺-clamped slac1 patch clamp: measure guard-cell K⁺in in slac1 vs WT
   with cytosolic [Ca²⁺] buffered <10 nM. Direct-inhibition model predicts higher
   K⁺in in slac1; Ca²⁺-feedback model predicts rescue to WT levels.
2. Non-conducting SLAC1 point mutants: test whether a pore-dead SLAC1 (anion
   transport abolished) still inhibits KAT1 — separates the inhibitor activity from
   anion-channel function.
3. Interface-mutant SLAC1: map the KAT1-binding surface and test a binding-dead,
   transport-competent SLAC1 for loss of K⁺in inhibition while retaining closure.
4. Endogenous proximity assay: split-luciferase/BiFC or co-IP from native guard
   cells (not overexpression) ± drought to show the interaction occurs at
   physiological levels.
5. Quantitative stoichiometry: compare SLAC1:KAT1 abundance in guard cells to
   judge whether direct binding can plausibly titrate KAT1 in planta.

Curation Leads (require curator verification)

• Lead 1 — Retain GO:0019870 as a non-core MF annotation with an explicit
  caveat that evidence is from a single study using heterologous oocyte voltage
  clamp plus guard-cell overexpression (IDA), supported by IPI (Y2H/co-IP), and is
  not independently replicated. Verify snippet: "Voltage-clamp results demonstrated
  that SLAC1 inhibited KAT1 dramatically, but did not inhibit KAT2." (PMID:27002025).
• Lead 2 — Add a NOT/contrast note on specificity: SLAC1 binds KAT2 (Y2H) but
  does not inhibit it; do not annotate KAT2 inhibition. Verify snippet: full-text
  "SLAC1 Strongly Inhibits KAT1, but Not KAT2, in Xenopus laevis Oocytes."
• Lead 3 — Flag competing mechanism: cite PMID:23126621 as a qualifier that
  in-planta slac1 K⁺in reduction is Ca²⁺-mediated, not a relief-of-direct-inhibition
  effect. Verify snippet: "severely reduced activity of inward K+ (K+in) channels in
  slac1 guard cells."
• Lead 4 — Do not homology-propagate GO:0019870 to SLAH3/other SLAC paralogs;
  SLAH3's K⁺ targets (GORK/SKOR, outward) differ (PMID:36653170).
• Candidate alternative term if curator deems "inhibitor" too strong: GO:0015459
  potassium channel regulator activity (direction still supports "inhibitor").
• Suggested curator question: Is a single-lab, heterologous + overexpression
  result sufficient for an experimental MF annotation given the opposite-direction
  loss-of-function phenotype? If policy requires independent corroboration for core
  terms, keep GO:0019870 as non-core only.

Supported vs Refuted Clauses of the Seed Hypothesis

Analyses run across iterations: NCBI E-utilities (efetch/elink/esummary) for the
reference paper, full-text AKT2/patch-clamp verification from PMC4863386, cited-by
screen of all 45 papers citing Zhang 2016, IntAct web-service interactor query for
Q9LD83 (19 interactors, all kinases/phosphatases), and UniProt/QuickGO MF-annotation
checks confirming GO:0019870 is not currently annotated.

Provenance artifacts (computed this run): evidence_matrix.csv,
go_decision_table.csv, hypothesis_clause_verdicts.csv in the job directory.

Artifacts

• OpenScientist evidence matrix
• OpenScientist final report
• OpenScientist final report
• OpenScientist go decision table
• OpenScientist hypothesis clause verdicts