DMI1 (DOES NOT MAKE INFECTIONS 1) is the Medicago truncatula POLLUX-type member of the CASTOR/POLLUX family of cation channels. It has four predicted transmembrane helices and a large cytosolic C-terminal region with two RCK (regulator of K+ conductance) domains that form an MthK-like, Ca2+-regulated gating ring. DMI1 resides in the nuclear envelope of root cells, preferentially in the inner nuclear membrane, where it forms a complex with the Ca2+-permeable cyclic nucleotide-gated channels CNGC15a/b/c. Within this complex DMI1 is required to activate CNGC15 and acts as a pacemaker that sets the frequency of the nuclear and perinuclear Ca2+ oscillations ("calcium spiking") induced by rhizobial Nod factors and arbuscular mycorrhizal fungal signals. Its native pore is a non-selective cation filter that conducts K+ in planar bilayers; the leading model is that DMI1 provides counter-ion flux across the nuclear envelope during Ca2+ release, although Ca2+ permeability of the related Lotus CASTOR channel has been reported. dmi1 mutants lack Nod factor-induced calcium spiking and fail to establish both root nodule and arbuscular mycorrhizal symbioses, placing DMI1 in the common symbiosis signalling pathway upstream of the Ca2+ decoder CCaMK/DMI3.
The preferred accession for this review is Q6RHR6. These relationships are manually assessed.
REFERENCE PROTEOME EQUIVALENT · Reviewed 2026-10-19
Reference proteome: UP000002051
Retain Q6RHR6 as the primary Swiss-Prot accession. The reference sequence is exactly identical across all 882 residues and is separated from CASTOR-like paralogs. Primary-record genomic cross-references provide additional mapping context. This supports equivalent DMI1 protein representation; the broad pollux-like target label is not a separate functional assertion.
MMseqs2 searched 57,093 sequences across UP000002051 at E <= 1e-5 and sensitivity 7.5, returning 13 hits. Every returned sequence was globally aligned using BLOSUM62 with gap-open -10 and extension -0.5. Scores and sequences for all 13 returned candidates are retained, with 10 detailed alignment files. The heuristic search does not establish genomic absence or unique locus identity.
file:MEDTR/DMI1/DMI1-record-evidence.yaml
target_sequence_count: 57093
file:MEDTR/DMI1/DMI1-record-evidence.yaml
candidate_count: 13
G7IN96: 882 identities across 882 global-alignment columns (100.0% identity including gaps), 0 substitutions and 0 gap columns. Query coverage is 100.0%; target coverage is 100.0%. Sequence similarity is interpreted in the manual rationale. Identity over paired residues is 100.0000% (882/882); gap columns are excluded from this denominator.
file:MEDTR/DMI1/DMI1-record-evidence.yaml
query_length: 882 target_length: 882 alignment_columns: 882 identities: 882 mismatches: 0 gap_columns: 0 identity_percent: 100.0 query_coverage_percent: 100.0 target_coverage_percent: 100.0
The primary accession is outside every searched exact-taxon reference proteome. Current UniProt records place G7IN96 in UP000002051 and give it the same taxon as the primary.
file:MEDTR/DMI1/DMI1-record-evidence.yaml
primary_in_any_reference_proteome: false
file:MEDTR/DMI1/DMI1-record-evidence.yaml
alternative_in_reference_proteome: true
file:MEDTR/DMI1/DMI1-record-evidence.yaml
same_taxon: true
Sampled UniSave observations: 09-DEC-2015, entry version 56.: no Proteomes cross-reference; 07-OCT-2020, entry version 83.: no Proteomes cross-reference; 27-NOV-2024, entry version 95.: no Proteomes cross-reference. These dated observations do not establish uninterrupted membership/absence, the exact change date, or UniProt's motivation.
file:MEDTR/DMI1/DMI1-record-evidence.yaml
entry_date: 09-DEC-2015, entry version 56. proteomes: []
file:MEDTR/DMI1/DMI1-record-evidence.yaml
entry_date: 07-OCT-2020, entry version 83. proteomes: []
file:MEDTR/DMI1/DMI1-record-evidence.yaml
entry_date: 27-NOV-2024, entry version 95. proteomes: []
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005515 protein binding | IPI PMID:27230377 Nuclear-localized cyclic nucleotide-gated channels mediate s... | MODIFY | Summary: Interaction of DMI1 (via its C-terminus) with CNGC15c (A0A072VMJ3), one of the three nuclear-envelope Ca2+-permeable CNGC15 channels with which DMI1 forms a complex (Charpentier et al. 2016). Reason: The interaction is real and functionally important, but "protein binding" is uninformative. The functional content of the DMI1-CNGC15 interaction is that DMI1 regulates CNGC15 gating: DMI1 is essential for activating CNGC15 and acts as a pacemaker for the oscillation frequency, with Ca2+-dependent mobility of the DMI1 C-terminal region (the region that binds the CNGC15 N-terminus) modulating CNGC15 gating (Cook et al. 2025). This matches "ion channel regulator activity" (modulates the activity of a channel via direct interaction with it). Proposed replacements: ion channel regulator activity Supporting Evidence: PMID:27230377 We demonstrate that the cyclic nucleotide-gated channels form a complex with the postassium-permeable channel, which modulates nuclear Ca(2+) release. PMID:39814887 Together, these results demonstrate that DMI1 is not only essential for activating CNGC15, but it also acts as a pacemaker for the frequency of nuclear Ca2+ oscillations PMID:39814887 This result further suggests that the mobility of the DMI1 C-terminal region in response to Ca2+ binding contributes to the pacemaker activity by modulating CNGC15 gating. |
| GO:0005515 protein binding | IPI PMID:27230377 Nuclear-localized cyclic nucleotide-gated channels mediate s... | MODIFY | Summary: Interaction of DMI1 with CNGC15a (G7IBJ4) in the nuclear-envelope DMI1-CNGC15 channel complex (Charpentier et al. 2016). Reason: As for the CNGC15c row: the interaction underlies DMI1 regulation of CNGC15 gating (activation and pacemaking of oscillation frequency; Cook et al. 2025), so the informative term is ion channel regulator activity rather than protein binding. Proposed replacements: ion channel regulator activity Supporting Evidence: PMID:27230377 We demonstrate that the cyclic nucleotide-gated channels form a complex with the postassium-permeable channel, which modulates nuclear Ca(2+) release. PMID:39814887 Together, these results demonstrate that DMI1 is not only essential for activating CNGC15, but it also acts as a pacemaker for the frequency of nuclear Ca2+ oscillations |
| GO:0005515 protein binding | IPI PMID:27230377 Nuclear-localized cyclic nucleotide-gated channels mediate s... | MODIFY | Summary: Interaction of DMI1 with CNGC15b (G7JND3) in the nuclear-envelope DMI1-CNGC15 channel complex (Charpentier et al. 2016); the interaction is retained by the autoactive DMI1 S760N variant (Liu et al. 2022). Reason: As for the other CNGC15 rows: the interaction is the physical basis for DMI1 regulation of CNGC15 gating. Autoactive DMI1 drives spontaneous Ca2+ oscillations only in a CNGC15-dependent manner (Liu et al. 2022), and DMI1 activates and paces CNGC15 (Cook et al. 2025). Replace protein binding with ion channel regulator activity. Proposed replacements: ion channel regulator activity Supporting Evidence: PMID:35972963 Here, we demonstrate spontaneous activation of this channel complex, through gain-of-function mutations in DMI1, leading to spontaneous nuclear Ca2+ oscillations and spontaneous nodulation, in a CNGC15-dependent manner. PMID:39814887 Together, these results demonstrate that DMI1 is not only essential for activating CNGC15, but it also acts as a pacemaker for the frequency of nuclear Ca2+ oscillations |
| GO:0006811 monoatomic ion transport | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO mapping from the CASTOR/POLLUX/SYM8-like family. DMI1 is a cation channel (bilayer K+ conductance; non-selective cationic selectivity filter), so ion transport is correct but very general. Reason: DMI1 is a channel conducting cations across the nuclear envelope; the more precise process is monoatomic cation transmembrane transport. Cook et al. (2025) describe the native DMI1 filter as non-selective cationic, and Charpentier et al. (2016) describe DMI1 as potassium-permeable. Proposed replacements: monoatomic cation transmembrane transport Supporting Evidence: PMID:39814887 To confirm that the non-selective cation permeability of DMI1 is not essential for its function, we swapped the non-selective cationic filter of DMI1 that comprised the motif 'ADAGN' with the strictly potassium-selective signature 'TVGYG' PMID:14963334 The DMI1 (does not make infections) gene encodes a novel protein with low global similarity to a ligand-gated cation channel domain of archaea. |
| GO:0006813 potassium ion transport | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO mapping from the RCK_N domain. For DMI1 it is supported directly: DMI1 conducts K+ in planar lipid bilayers (about 64 pS in symmetrical KCl; Venkateshwaran et al. 2012), it is described as the potassium-permeable partner of CNGC15 (Charpentier et al. 2016), and a strictly K+-selective pore variant fully restores DMI1 function in planta (Cook et al. 2025). Reason: Although the RCK_N domain alone would be weak grounds for a potassium transport claim, the experimental data for DMI1 support K+ conduction, and the TVGYG swap shows that K+ permeation is sufficient for symbiotic function. The Ca2+ selectivity reported for Lotus CASTOR (Kim et al. 2019) is homolog evidence and does not refute K+ transport by DMI1. Supporting Evidence: PMID:27230377 A potassium-permeable channel is known to be required for symbiotic Ca(2+) oscillations PMID:39814887 These results demonstrate that substituting a non-selective filter with a potassium-selective filter does not impair the function of DMI1. file:MEDTR/DMI1/DMI1-deep-research-falcon.md The strongest **direct MtDMI1 ion-conduction measurement** is a single-channel conductance of approximately **64 pS in symmetrical KCl** in planar-bilayer experiments. PMID:31420535 Initial studies using yeast rescue assays and artificial lipid bilayers suggested that CASTOR, POLLUX, and DMI1 could be K+ channels |
| GO:0031965 nuclear membrane | EXP PMID:27230377 Nuclear-localized cyclic nucleotide-gated channels mediate s... | ACCEPT | Summary: DMI1 localizes to the nuclear envelope together with CNGC15 (Charpentier et al. 2016), consistent with earlier localization of functional DMI1-GFP to the nuclear envelope (Riely et al. 2007) and immunogold data showing preferential inner nuclear membrane localization (Capoen et al. 2011). Reason: Nuclear membrane localization is well established by multiple independent studies and is where DMI1 functions in generating nuclear Ca2+ oscillations. Supporting Evidence: PMID:17173544 we report here that a functional DMI1::GFP fusion is localized to the nuclear envelope in M. truncatula roots when expressed both from a constitutive 35S promoter and from a native DMI1 promoter. PMID:21825141 we found that the cation channel DMI1, which is essential for symbiotic calcium oscillations, is preferentially located on the inner nuclear membrane |
| GO:0031965 nuclear membrane | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location mapping (Nucleus membrane), based on the same experimental localization studies. Reason: Consistent with the experimental nuclear envelope / inner nuclear membrane localization of DMI1. Supporting Evidence: PMID:17173544 These new data place DMI1 in the nuclear envelope in close proximity to the origin of Nod-factor-induced calcium spiking. |
| GO:0005261 monoatomic cation channel activity | IDA PMID:22706284 The recent evolution of a symbiotic ion channel in the legum... | NEW | Summary: DMI1 forms cation channels in planar lipid bilayers (about 64 pS in symmetrical KCl, long open state; Venkateshwaran et al. 2012). Its native selectivity filter (ADAGN) is a non-selective cation filter, and a strictly K+-selective (TVGYG) variant is fully functional in planta (Cook et al. 2025). Reason: DMI1 has no molecular function annotation in GOA other than protein binding, yet direct electrophysiology shows it is a cation channel. The generic cation channel term is used, matching the treatment of the Lotus homolog CASTOR, because native selectivity (K+ versus broader cation or Ca2+ permeation) is still debated; the K+ conduction is captured by the accepted potassium ion transport annotation. Supporting Evidence: PMID:22706284 Electrophysiological characterization of DMI1 and Lj-CASTOR (wild-type and mutants) by planar lipid bilayer experiments combined with calcium imaging in Human Embryonic Kidney-293 cells expressing DMI1 (the wild type and mutants) suggest that the serine-to-alanine substitution conferred reduced conductance with a long open state to DMI1 file:MEDTR/DMI1/DMI1-deep-research-falcon.md The strongest **direct MtDMI1 ion-conduction measurement** is a single-channel conductance of approximately **64 pS in symmetrical KCl** in planar-bilayer experiments. PMID:39814887 These results demonstrate that substituting a non-selective filter with a potassium-selective filter does not impair the function of DMI1. |
| GO:0009877 nodulation | IMP PMID:14963334 Medicago truncatula DMI1 required for bacterial and fungal s... | NEW | Summary: dmi1 mutants are unable to form nodules with rhizobia: they show no Nod factor-induced calcium spiking and no nodule formation. DMI1 is a working component of the nuclear Ca2+ oscillator that transduces the Nod factor signal (it activates and paces the CNGC15 Ca2+ channels), and gain-of-function DMI1 alleles drive spontaneous nodulation. Reason: Participation: DMI1 does part of the work of nodulation signalling, as a channel and CNGC15 regulator in the Ca2+ oscillator; it is not a substrate of the process. Comparator check: CNGC15A and CNGC15B, the partner channels in the same complex and in the same pathway role, carry IMP nodulation annotations (from PMID:27230377), as do CCaMK and CYCLOPS, the next steps in the pathway. DMI1 lacks this annotation despite being the founding gene of the oscillator, which looks like a gap rather than a deliberate convention. Supporting Evidence: PMID:14963334 Several of the plant genes required for transduction of rhizobial signals, the Nod factors, are also necessary for mycorrhizal symbiosis. Here, we describe the cloning and characterization of one such gene from the legume Medicago truncatula. PMID:35972963 Here, we demonstrate spontaneous activation of this channel complex, through gain-of-function mutations in DMI1, leading to spontaneous nuclear Ca2+ oscillations and spontaneous nodulation, in a CNGC15-dependent manner. |
| GO:0036377 arbuscular mycorrhizal association | IMP PMID:14963334 Medicago truncatula DMI1 required for bacterial and fungal s... | NEW | Summary: DMI1 is a common symbiosis pathway gene required for arbuscular mycorrhizal (AM) symbiosis as well as nodulation. Medicago DMI1 also restores AM colonization in a dmi1 mutant of the liverwort Marchantia paleacea, showing conservation of this role across land plants. Reason: Same participation and comparator arguments as for nodulation: the nuclear Ca2+ oscillator generated by DMI1 and CNGC15 is used by both symbioses, and CNGC15A and CNGC15B carry IMP arbuscular mycorrhizal association annotations from the same work (PMID:27230377). Keeping this distinct from any defense-response term is appropriate; DMI1 has no demonstrated role in immunity. Supporting Evidence: PMID:14963334 Several of the plant genes required for transduction of rhizobial signals, the Nod factors, are also necessary for mycorrhizal symbiosis. file:MEDTR/DMI1/DMI1-deep-research-falcon.md found that MtDMI1 restored mycorrhizal colonization in a DMI1-deficient liverwort, *Marchantia paleacea*. |
Loading supporting content…
Download this section (compressed HTML)Q: What ions does native, full-length DMI1 conduct in the inner nuclear membrane, and in which direction? The K+-selective TVGYG pore variant is fully functional, while the Lotus homolog CASTOR has been reported as Ca2+-selective; should the core MF be potassium channel activity rather than generic cation channel activity?
Q: Does DMI1 regulate CNGC15 gating primarily through direct protein contact (RCK gating ring to CNGC15 N-terminus) or electrically, through changing nuclear membrane potential? The answer determines whether "ion channel regulator activity" is the right MF term.
Q: Should a GO term exist for generation of symbiotic nuclear calcium oscillations, so that DMI1, CNGC15 and MCA8 can be annotated to the oscillator step rather than only to nodulation and arbuscular mycorrhizal association?
Experiment: Patch-clamp of native inner nuclear membranes (nuclear patch) from wild-type and dmi1 roots, or of full-length DMI1 reconstituted with CNGC15, to measure K+/Ca2+ permeability ratios under physiological gradients.
Experiment: Separation-of-function DMI1 mutants that keep channel conductance but disrupt the CNGC15-binding interface (and vice versa), tested for Nod factor-induced spiking frequency, to separate the conduction and channel-regulator roles.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)