IMD118 is X-linked MCTS1 deficiency, a Mendelian susceptibility to mycobacterial disease in which a general translation factor defect is funnelled into a single clinical phenotype by a single vulnerable substrate. Nine patients in nine unrelated families are published: five in the 2023 report that defined the entity and four in a 2026 follow-up. MCTS1 partners with DENR to recycle post-termination 40S ribosomal subunits and to allow reinitiation on messages carrying short upstream open reading frames. Losing it should, on the face of it, disturb a great deal. It does not. Of seventeen genes of IFN-gamma immunity tested in MCTS1-knockout cells, only JAK2 translation fell — because the JAK2 5-prime untranslated region carries two ultra-short start-stop upstream open reading frames at which 40S subunits stall without MCTS1 to recycle them, blocking reinitiation at the main open reading frame. The consequence is a JAK2 level low enough to impair responses to IL-23, and partially to IL-12, while leaving other JAK2-dependent cytokine responses intact. Defective IL-23 signalling preferentially costs IFN-gamma production by the innate-like adaptive lymphocytes — MAIT cells and gamma-delta T cells — that respond to mycobacterial challenge, and IFN-gamma is the macrophage-activating cytokine that controls intracellular mycobacteria. Everything else MCTS1 does appears to be physiologically redundant, which is the founding paper's own word for it. Two clinically load-bearing consequences follow. The block is upstream of the IFN-gamma receptor and patient cells respond normally to exogenous IFN-gamma, so the lesion is bypassable — and by 2026, three patients had been given IFN-gamma-1b with reported benefit. And the disease is not the benign entity the first report suggested: one of the original five died, one of the four new patients died of disseminated BCG involving the brain, and two of the new patients have hypogammaglobulinaemia requiring immunoglobulin replacement, which the phrase "otherwise physiologically redundant" does not prepare a reader for.
Ask a research question about Immunodeficiency 118. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).
Do not include personal health information in your question. Questions and results are cached in your browser's local storage.
name: Immunodeficiency 118
category: Mendelian
creation_date: "2026-09-22T00:00:00Z"
synonyms:
- IMD118
- MCTS1 deficiency
- X-linked MCTS1 deficiency
- Mendelian susceptibility to mycobacterial disease due to MCTS1 deficiency
- MSMD due to MCTS1 deficiency
disease_term:
preferred_term: immunodeficiency 118
term:
id: MONDO:0958030
label: immunodeficiency 118
description: >-
IMD118 is X-linked MCTS1 deficiency, a Mendelian susceptibility to mycobacterial disease in
which a general translation factor defect is funnelled into a single clinical phenotype by a
single vulnerable substrate. Nine patients in nine unrelated families are published: five in
the 2023 report that defined the entity and four in a 2026 follow-up.
MCTS1 partners with DENR to recycle post-termination 40S ribosomal subunits and to allow
reinitiation on messages carrying short upstream open reading frames. Losing it should, on the
face of it, disturb a great deal. It does not. Of seventeen genes of IFN-gamma immunity tested
in MCTS1-knockout cells, only JAK2 translation fell — because the JAK2 5-prime untranslated
region carries two ultra-short start-stop upstream open reading frames at which 40S subunits
stall without MCTS1 to recycle them, blocking reinitiation at the main open reading frame.
The consequence is a JAK2 level low enough to impair responses to IL-23, and partially to
IL-12, while leaving other JAK2-dependent cytokine responses intact. Defective IL-23 signalling
preferentially costs IFN-gamma production by the innate-like adaptive lymphocytes — MAIT cells
and gamma-delta T cells — that respond to mycobacterial challenge, and IFN-gamma is the
macrophage-activating cytokine that controls intracellular mycobacteria. Everything else MCTS1
does appears to be physiologically redundant, which is the founding paper's own word for it.
Two clinically load-bearing consequences follow. The block is upstream of the IFN-gamma
receptor and patient cells respond normally to exogenous IFN-gamma, so the lesion is
bypassable — and by 2026, three patients had been given IFN-gamma-1b with reported benefit.
And the disease is not the benign entity the first report suggested: one of the original five
died, one of the four new patients died of disseminated BCG involving the brain, and two of
the new patients have hypogammaglobulinaemia requiring immunoglobulin replacement, which the
phrase "otherwise physiologically redundant" does not prepare a reader for.
parents:
- Mendelian Susceptibility to Mycobacterial Disease
- Inborn Errors of Immunity
classifications:
harrisons_chapter:
- classification_value: IMMUNE_RHEUMATOLOGIC
notes: >-
A monogenic inborn error of immunity. It presents to infectious disease or paediatrics as
a mycobacterial infection and is diagnosed by immunology and genetics.
iuis_category:
classification_value: innate immunity defect
notes: >-
IUIS places Mendelian susceptibility to mycobacterial disease among the defects of
intrinsic and innate immunity rather than among the combined immunodeficiencies. That
placement fits here: circulating leukocyte subsets develop normally, and the lesion is in
an induction pathway rather than in lymphocyte development.
references:
- reference: PMID:37875108
title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
- reference: PMID:41623352
title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
inheritance:
- name: X-linked recessive inheritance
description: >-
All nine published patients are male and hemizygous. The 2023 kindreds are from China,
Finland, Iran and Saudi Arabia; the 2026 families from Iran, Oman, China and the United
States. Consanguinity is common but not required — the American patient's family is not
described as consanguineous, and his variant is hypomorphic rather than null.
inheritance_term:
preferred_term: X-linked recessive inheritance
term:
id: HP:0001419
label: X-linked recessive inheritance
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "We report X-linked recessive MCTS1 deficiency in men with mycobacterial disease from kindreds of different ancestries (from China, Finland, Iran, and Saudi Arabia)."
explanation: >-
The inheritance mode and the geographic spread of the founding kindreds, which is what
makes this a recurrent genotype rather than a founder effect.
pathophysiology:
- name: Hemizygous MCTS1 Loss-of-Function Variants
biological_scale: MOLECULAR
mechanism_confidence: ESTABLISHED
description: >-
Eight variants are published across nine families. The 2023 five are all loss-of-expression
or complete loss-of-function: two frameshift, one splice-site, and two in-frame 3-prime
deletions both removing residues Ala133 to Lys181. The 2026 report adds p.Leu170* and
p.Glu60Lysfs*5, both loss-of-function, and p.Trp175*, which is hypomorphic when
overexpressed rather than null.
That last allele is the reason this entity now has a partial form. A nonsense variant at
codon 175 of a 181-residue protein removes very little, and the patient who carries it was
not BCG-vaccinated, stayed healthy through childhood, and presented at sixteen with weight
loss and diarrhoea and at twenty-two with disseminated Mycobacterium abscessus. Allele
severity and age at presentation move together.
genes:
- preferred_term: MCTS1
term:
id: hgnc:23357
label: MCTS1
genetic_context:
genes:
- preferred_term: MCTS1
term:
id: hgnc:23357
label: MCTS1
allele_type: frameshift, nonsense, splice-site and in-frame 3-prime deletion alleles, with one hypomorphic nonsense allele near the C-terminus
variant_origin: GERMLINE
zygosity: HEMIZYGOUS
functional_impact_category: LOSS_OF_FUNCTION
notes: >-
LOSS_OF_FUNCTION describes eight of the nine genotypes; p.Trp175* is hypomorphic on
overexpression and is the exception, recorded in this node's description rather than by
splitting the slot. The slot is single-valued and the majority genotype is a true null.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: PRIMARY_RESULT
snippet: "The p.L170* and E60Kfs5* variants are LOF, whereas p.W175* is hypomorphic when overexpressed."
explanation: >-
The three new alleles and the functional class of each, including the one that establishes
a partial form of the disease.
downstream:
- target: Failure of 40S Ribosome Recycling and Translation Reinitiation
causal_link_type: DIRECT
- name: Failure of 40S Ribosome Recycling and Translation Reinitiation
biological_scale: MOLECULAR
mechanism_confidence: ESTABLISHED
description: >-
MCTS1 heterodimerises with DENR to recycle 40S ribosomal subunits after termination and to
permit reinitiation on messages carrying short upstream open reading frames. Without it,
subunits are not released at stop codons and reinitiation downstream is impaired.
The striking part is how little this costs. The founding authors call MCTS1-dependent
recycling "otherwise physiologically redundant" in these patients — meaning that outside the
one pathway below, no phenotype attributable to a general translation defect was found. That
is a strong claim and it is theirs, not this entry's; the 2026 series' growth failure,
hypogammaglobulinaemia and gastrointestinal disease sit awkwardly against it, and this entry
records both.
molecular_functions:
- preferred_term: translation reinitiation factor activity
term:
id: GO:0003743
label: translation initiation factor activity
modifier: DECREASED
biological_processes:
- preferred_term: regulation of translational initiation
term:
id: GO:0006446
label: regulation of translational initiation
modifier: DECREASED
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: PRIMARY_RESULT
snippet: "Complete deficiency of this translation re-initiation factor impairs the translation of a subset of proteins, including the kinase JAK2 in all cell types tested, including T lymphocytes and phagocytes."
explanation: >-
The molecular consequence and its scope — a subset of proteins, in every cell type looked
at, which is what makes the narrowness of the clinical phenotype surprising.
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Surprisingly, the lack of MCTS1-dependent translation re-initiation and ribosome recycling seems to be otherwise physiologically redundant in these patients."
explanation: >-
The redundancy claim, quoted with the authors' own "seems to" intact. The 2026 series
complicates it, which is why the hedge is preserved rather than paraphrased away.
downstream:
- target: Ribosomal Stalling at the JAK2 Start-Stop Upstream Open Reading Frames
causal_link_type: DIRECT
- name: Ribosomal Stalling at the JAK2 Start-Stop Upstream Open Reading Frames
biological_scale: MOLECULAR
mechanism_confidence: ESTABLISHED
description: >-
This node is the specificity of the disease. The JAK2 5-prime untranslated region carries
three upstream open reading frames, two of them ultra-short start-stop frames. Without MCTS1
to recycle them, 40S subunits stall at those stop codons and cannot reinitiate at the JAK2
main open reading frame.
It is why a general translation defect produces a single-pathway disease. Of seventeen genes
of IFN-gamma immunity tested in MCTS1-knockout cells, only JAK2 showed a substantial
translational decrease, rescued by wild-type MCTS1 and not by patient variants. A gene's
vulnerability here is a property of its 5-prime untranslated region, not of its function.
biological_processes:
- preferred_term: translational initiation
term:
id: GO:0006413
label: translational initiation
modifier: DECREASED
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: BACKGROUND
snippet: "MCTS1 promotes JAK2 translation by reinitiating translation in the JAK2 5′UTR."
explanation: >-
The mechanism stated compactly. Graded BACKGROUND because the 2026 paper is restating the
2023 finding in its introduction rather than reporting it, and HUMAN_CLINICAL because the
work it summarises was done in patient and patient-derived cells.
downstream:
- target: Reduced JAK2 Protein
causal_link_type: DIRECT
- name: Reduced JAK2 Protein
biological_scale: MOLECULAR
mechanism_confidence: ESTABLISHED
description: >-
JAK2 falls in every cell type tested, including T lymphocytes and phagocytes. It does not
fall to zero, and that quantitative detail is what makes the phenotype selective rather than
global: the level is low enough to impair IL-23 responses and partially IL-12 responses, and
not low enough to impair the other JAK2-dependent cytokine responses.
A JAK2 null would be a different and much worse disease — JAK2 carries erythropoietin,
thrombopoietin and growth-hormone signalling. Nothing like that is seen here.
genes:
- preferred_term: JAK2
term:
id: hgnc:6192
label: JAK2
molecular_functions:
- preferred_term: non-membrane spanning protein tyrosine kinase activity
term:
id: GO:0004715
label: non-membrane spanning protein tyrosine kinase activity
modifier: DECREASED
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: PRIMARY_RESULT
snippet: "JAK2 expression is sufficiently low to impair cellular responses to interleukin-23 (IL-23) and partially IL-12, but not other JAK2-dependent cytokines."
explanation: >-
The quantitative threshold argument in the authors' own words — which cytokine responses
the residual JAK2 supports and which it does not.
downstream:
- target: Impaired IL-23 Receptor Signalling
causal_link_type: DIRECT
- name: Impaired IL-23 Receptor Signalling
biological_scale: CELLULAR
mechanism_confidence: ESTABLISHED
description: >-
IL-23 signalling is JAK2-dependent and is the response that fails here; IL-12 signalling is
partially affected. The placement of this disease on the IL-23 limb rather than the IL-12
limb of the MSMD circuit is not an inference from the biochemistry — it was measured in
patient cells, which showed impaired IL-23 responses with normal IL-12 responses.
That places IMD118 beside IL23R deficiency and IL-23 deficiency in the same functional
pathway, and it is the reason the phenotype is mycobacterial rather than the broader
susceptibility that IL-12 axis defects produce.
biological_processes:
- preferred_term: interleukin-23-mediated signaling pathway
term:
id: GO:0038155
label: interleukin-23-mediated signaling pathway
modifier: DECREASED
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: PRIMARY_RESULT
snippet: "Defective responses to IL-23 preferentially impair the production of IFN-γ by innate-like adaptive mucosal-associated invariant T cells (MAIT) and γδ T lymphocytes upon mycobacterial challenge."
explanation: >-
The step from cytokine signalling to effector function, naming the two lymphocyte
populations that carry it.
downstream:
- target: Reduced IFN-gamma Production by Innate-like Adaptive T Lymphocytes
causal_link_type: DIRECT
- name: Reduced IFN-gamma Production by Innate-like Adaptive T Lymphocytes
biological_scale: CELLULAR
mechanism_confidence: ESTABLISHED
description: >-
MAIT cells and gamma-delta T cells produce less IFN-gamma on mycobacterial challenge. These
are the lymphocytes that respond to mycobacteria without needing clonal expansion from a
naive repertoire, which is why an infant meeting BCG in the first months of life depends on
them.
Adequate JAK2 is required for exactly this, which is the 2026 authors' summary of the
pathway and the reason the entity sits within MSMD rather than beside it.
cell_types:
- preferred_term: MAIT cell
term:
id: CL:0000940
label: mucosal-associated invariant T cell
- preferred_term: gamma-delta T cell
term:
id: CL:0000798
label: gamma-delta T cell
biological_processes:
- preferred_term: interferon-gamma production
term:
id: GO:0032609
label: type II interferon production
modifier: DECREASED
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: BACKGROUND
snippet: "Adequate levels of JAK2 expression are required for IFN-γ production in response to IL-23 by innate-like adaptive T cells, such as Vδ2+ γδ T cells and mucosa-associated innate-like T cells"
explanation: >-
The pathway restated with the gamma-delta subset specified as Vdelta2-positive. Graded
BACKGROUND because the 2026 paper is summarising its predecessor here.
downstream:
- target: Failure of Macrophage Activation Against Intracellular Mycobacteria
causal_link_type: DIRECT
- name: Failure of Macrophage Activation Against Intracellular Mycobacteria
biological_scale: ORGANISM
mechanism_confidence: ESTABLISHED
description: >-
IFN-gamma is the macrophage-activating factor at the heart of antimycobacterial immunity, and
without enough of it macrophages do not restrict the intracellular growth of weakly virulent
mycobacteria. This is the step every MSMD genotype converges on, and it is established for
the syndrome rather than measured in MCTS1 patients specifically.
Its most important property here is where it sits. The lesion is upstream of the IFN-gamma
receptor and patient leukocytes respond normally to exogenous IFN-gamma, so this node can be
driven from outside — which is what makes recombinant IFN-gamma a rational therapy rather
than a hopeful one.
cell_types:
- preferred_term: macrophage
term:
id: CL:0000235
label: macrophage
biological_processes:
- preferred_term: macrophage activation
term:
id: GO:0042116
label: macrophage activation
modifier: DECREASED
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: INDIRECT
evidence_source: OTHER
quote_role: BACKGROUND
snippet: "Interferon-gamma (IFN-γ), a macrophage-activating factor (3), is at the heart of antimycobacterial immunity."
explanation: >-
The general principle this node rests on, stated in the paper's introduction with its own
citation. Graded BACKGROUND and INDIRECT because it is established immunology imported into
this disease, not a measurement in an MCTS1 patient.
downstream:
- target: Disseminated BCG Disease
causal_link_type: DIRECT
- target: Nontuberculous Mycobacterial Infection
causal_link_type: DIRECT
phenotypes:
- name: Disseminated BCG Disease
category: Immune
description: >-
The defining presentation. Eight of the nine published patients were BCG-vaccinated at or
near birth and most developed disseminated BCG disease between one and twelve months. The
2026 series shows the range: axillary adenitis with bone-marrow involvement at four months,
abscessed BCG adenitis at seven months, local armpit disease at one month, and in one
patient a BCG-osis involving brain, mastoid sinuses, lung, liver and gastrointestinal tract
that was fatal.
frequency: VERY_FREQUENT
phenotype_term:
preferred_term: Disseminated BCG disease
term:
id: HP:0020087
label: BCGosis
notes: >-
Present in most of the nine published patients. The two exceptions are informative rather
than incidental: one 2023 patient had not been BCG-vaccinated and was only two years old at
last follow-up, and the 2026 American patient was never vaccinated and presented instead with
environmental mycobacterial disease in adulthood. So the BCG phenotype requires the exposure,
and an unvaccinated carrier is not protected — only differently exposed.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Three patients had adverse reactions to the BCG vaccine, whereas another patient was not vaccinated with BCG and had an infection with Mycobacterium abscessus at 16 years of age."
explanation: >-
The split in the 2026 series between vaccinated and unvaccinated patients, which is the
evidence that the vaccine is the usual trigger rather than the cause.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "An empirical diagnosis of BCG-osis (affecting the brain, mastoid sinuses, lung, liver, and gastrointestinal tract) was established"
explanation: >-
The severe end of the range, in the patient who died — the organs involved when the disease
is not controlled.
sequelae:
- target: Lymphadenopathy
description: >-
Regional then disseminated nodal disease is how BCG spreads from the vaccination site.
- target: Granulomatous Meningitis and Seizures
description: >-
Central nervous system involvement in the patient in whom BCG disease was not controlled,
with the organism cultured from cerebrospinal fluid.
- target: Failure to Thrive
description: >-
Chronic mycobacterial infection is itself a cause of growth failure. Whether it is the only
cause here is the question recorded in this entry's redundancy discussion.
- target: Chronic Diarrhoea and Gastrointestinal Mycobacterial Disease
description: >-
The gastrointestinal tract is one of the organs BCG disseminates to, and in one patient the
ileocaecal lesion was mycobacterium-positive on biopsy.
- name: Nontuberculous Mycobacterial Infection
category: Immune
description: >-
Disseminated Mycobacterium abscessus in the unvaccinated American patient, presenting as
bulky intra-abdominal and retroperitoneal lymphadenopathy with chronic diarrhoea and weight
loss, and Mycobacterium tuberculosis complex on ileocaecal mucosal biopsy in the Chinese
patient. Environmental mycobacteria are what these patients meet when BCG is not given.
frequency: OCCASIONAL
phenotype_term:
preferred_term: Disseminated Mycobacterium abscessus infection
term:
id: HP:0032130
label: Mycobacterium abscessus abscessus infection
notes: >-
2 of 9 published patients, and both of them presented late — sixteen and twenty-two years in
the American patient, ten years in the Chinese one. The diagnostic lesson is that an adult
with disseminated environmental mycobacterial disease and a normal immunophenotype is a
candidate for this diagnosis.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "The patient was diagnosed with disseminated M. abscessus infection manifesting as bulky intra-abdominal/retroperitoneal lymphadenopathy, chronic diarrhea, and chronic weight loss."
explanation: The full presentation in the unvaccinated adult patient.
sequelae:
- target: Lymphadenopathy
description: >-
Bulky intra-abdominal and retroperitoneal nodal disease is the presenting manifestation of
the M. abscessus infection in the unvaccinated adult patient.
- target: Chronic Diarrhoea and Gastrointestinal Mycobacterial Disease
description: >-
Chronic diarrhoea accompanied the disseminated abdominal disease in that patient and the
ileocaecal tuberculous ulcer in another.
- name: Lymphadenopathy
category: Immune
description: >-
Regional adenitis at the vaccination site in the BCG patients, and bulky intra-abdominal and
retroperitoneal conglomerate lymphadenopathy in the adult with M. abscessus — in his case a
retroperitoneal mass encasing the aorta and the proximal coeliac and superior mesenteric
arteries, with necrotic change.
frequency: VERY_FREQUENT
phenotype_term:
preferred_term: Regional and intra-abdominal lymphadenopathy
term:
id: HP:0002716
label: Lymphadenopathy
notes: >-
Reported for most published patients in one form or another. It is infection-driven rather
than a primary lymphoid abnormality, which is why it resolves on antimycobacterial treatment.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "At the age of 4 mo, he presented BCG-osis with left axillary adenitis."
explanation: The regional adenitis in the youngest of the 2026 patients, with its age.
- name: Normal Circulating Leukocyte Subsets
category: Immune
description: >-
Routine immunophenotyping is normal. That is the diagnostic problem this disease poses: the
test a clinician runs to confirm an immunodeficiency comes back reassuring, and only
molecular diagnosis finds it.
Recorded as a phenotype in its own right because its absence is what a curator or clinician
would otherwise infer from the disease's name.
frequency: VERY_FREQUENT
phenotype_term:
preferred_term: Normal development of circulating leukocyte subsets
notes: >-
Reported for all five 2023 patients and for the American patient in 2026.
Deliberately unbound. HPO has no term for a normal immunophenotype in a patient who has an
immunodeficiency, and every candidate inverts the claim: HP:0002721 Immunodeficiency asserts
the abnormality this node exists to deny, and the lymphocyte-count terms assert a direction.
A term plus a disclaimer is still a wrong term — the disclaimer is prose and the term is
machine-readable — so `term:` is omitted and `preferred_term` carries the finding. Needs a
term request: the concept is "routine immunophenotyping normal", which is diagnostically
load-bearing across the whole of MSMD and not only here.
Searched before omitting, and re-run against `ols:hp` before committing this note.
`l~leukocyte` returns abnormality terms only (HP:0001881 Abnormal leukocyte morphology,
HP:0001974 Increased total leukocyte count, HP:0033797 Leukocyte migration defect and
siblings). `l~lymphocyte count` returns HP:0040088 Abnormal lymphocyte count and its
increased/decreased children, including HP:0005403 Decreased total T cell count.
`l~immunophenotyp` returns nothing at all. `l~normal` returns terms in which "normal" is part
of some other concept — HP:0002343 Normal pressure hydrocephalus, HP:0002372 Normal
interictal EEG, HP:0002632 Low-to-normal blood pressure — and nothing about a normal
immunological result. HPO has terms for the presence of an abnormality and none for a
documented normal finding in this domain, which is what makes this an NTR rather than a
search failure. Note HP:0002372 shows the shape a suitable term would take.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: BACKGROUND
snippet: "All the detected leukocyte subsets developed normally in the patients."
explanation: >-
The normal immunophenotype across the 2023 cohort, restated in the 2026 introduction —
which is why it is graded BACKGROUND rather than PRIMARY_RESULT.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Complete blood counts and clinical immunophenotyping results were normal (Tables S3 and S4)."
explanation: >-
The same finding measured directly in the 2026 adult patient, whose disseminated
M. abscessus infection would otherwise have prompted a search for an acquired cause.
- name: Decreased Circulating Immunoglobulin
category: Immune
description: >-
Low immunoglobulins with normal total protein and albumin in the youngest 2026 patient,
treated with intravenous immunoglobulin; hypogammaglobulinaemia in the Chinese patient, also
on replacement; moderately low B cell counts and IgG in the Omani patient.
This is the finding that sits least comfortably with the founding description of an otherwise
intact immune system. Three of the four 2026 patients received immunoglobulin replacement.
frequency: FREQUENT
phenotype_term:
preferred_term: Hypogammaglobulinaemia requiring immunoglobulin replacement
term:
id: HP:0004313
label: Decreased circulating immunoglobulin concentration
notes: >-
3 of 4 patients in the 2026 series; not reported in the 2023 cohort, whose leukocyte subsets
are described as normal without immunoglobulin data being quoted in the abstract. Whether
this is a newly recognised part of the phenotype or a difference in what was measured and
reported cannot be settled from the published record, and the entry does not try to.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Immunological analyses revealed that P1 had low immunoglobulin (IG) levels, despite having normal total serum protein and albumin levels. Intravenous IG (IVIG) treatment was therefore initiated."
explanation: >-
The finding with the controls that make it a genuine hypogammaglobulinaemia rather than a
dilutional or loss-related artefact.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "He also receives IFN-γ-1b and IVIG due to his hypogammaglobulinemia (Tables S3 and S4)."
explanation: The second instance, in the Chinese patient, alongside his IFN-gamma treatment.
sequelae:
- target: Recurrent Otitis Media
description: >-
Recurrent bacterial otitis is the expected consequence of an antibody deficit rather than
of the IFN-gamma induction defect, which is why it is drawn from here rather than from the
mycobacterial chain.
- name: Failure to Thrive
category: Growth
description: >-
Weight and height below the third percentile at three and a half years in the youngest 2026
patient; failure to thrive with height and weight below the third percentile and delayed gross
motor development in the Omani patient. Chronic weight loss in the adult with M. abscessus.
frequency: FREQUENT
phenotype_term:
preferred_term: Growth failure below the third percentile
term:
id: HP:0001508
label: Failure to thrive
notes: >-
3 of 4 patients in the 2026 series. Against the 2023 description of normal growth and
development in the surviving patients, this is the clearest divergence between the two
reports. Chronic mycobacterial infection is itself a cause of growth failure, so it is not
separable here from a direct effect of the genotype.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "At 3.5 years old, his weight was 10.5 kg, and he was 77 cm tall (both these growth parameters are below the third percentile)."
explanation: The measured growth failure, with the centile, in the youngest 2026 patient.
- name: Chronic Diarrhoea and Gastrointestinal Mycobacterial Disease
category: Gastrointestinal
description: >-
Recurrent diarrhoea from age six in the Chinese patient, leading at ten to colonoscopy that
found an ileocaecal ulcer positive for Mycobacterium tuberculosis complex; chronic diarrhoea
with abdominal distension, ascites and pancolitis in the Omani patient; chronic diarrhoea in
the American patient with M. abscessus.
The gut is a mycobacterial site in this disease, not merely a site of nonspecific symptoms —
which matters because a chronic diarrhoea in a patient already on antimycobacterial therapy
invites a search for a different cause.
frequency: FREQUENT
phenotype_term:
preferred_term: Chronic diarrhoea with gastrointestinal mycobacterial infection
term:
id: HP:0002014
label: Diarrhea
temporality: CHRONIC
notes: 3 of 4 patients in the 2026 series; not reported in the 2023 cohort.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "A mucosal biopsy specimen from this area tested positive for Mycobacterium tuberculosis complex by metagenomic next-generation sequencing."
explanation: >-
That the gastrointestinal disease is mycobacterial, established on tissue rather than
inferred from the systemic diagnosis.
- name: Recurrent Otitis Media
category: Otologic
description: >-
Recurrent otitis media from age nine in the Chinese patient, and otomastoiditis on imaging
in the Omani patient during his fatal illness.
frequency: OCCASIONAL
phenotype_term:
preferred_term: Recurrent otitis media
term:
id: HP:0000403
label: Recurrent otitis media
notes: >-
2 of 4 patients in the 2026 series. Recurrent bacterial otitis in a patient with
hypogammaglobulinaemia is the expected consequence of the antibody deficit rather than of the
mycobacterial axis, which is why it is recorded separately from the mycobacterial phenotypes.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "The local infection gradually improved, but P3 has suffered from recurrent diarrhea since the age of 6 years and recurrent otitis media since the age of 9 years."
explanation: The two chronic problems that followed the resolved BCG disease in this patient.
- name: Granulomatous Meningitis and Seizures
category: Nervous System
description: >-
Central nervous system BCG disease in the Omani patient: generalised tonic-clonic seizure at
six, imaging showing otomastoiditis, acute cerebral venous thrombosis and granulomatous
meningitis with focal white-matter lesions, pyrazinamide-resistant M. bovis BCG cultured from
cerebrospinal fluid, then breakthrough seizures, status epilepticus and new lesions. He died
at six years and nine months.
frequency: OCCASIONAL
phenotype_term:
preferred_term: Granulomatous BCG meningitis
term:
id: HP:0001287
label: Meningitis
notes: >-
1 of 9 published patients, and the one who died in the 2026 series. Recorded because it is
the worst documented course of this disease and because a reader taking the founding report's
"isolated mycobacterial disease" framing would not expect CNS involvement.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Cranial computed tomography (CT) and magnetic resonance imaging revealed otomastoiditis, acute cerebral venous thrombosis, and granulomatous meningitis with focal white matter lesions that were particularly abundant in the right hemisphere."
explanation: The imaging findings that established central nervous system involvement.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Cerebrospinal fluid (CSF) culture revealed the presence of pyrazinamide-resistant M. bovis–BCG."
explanation: >-
The organism, cultured from cerebrospinal fluid, with its resistance pattern — which is
what makes this BCG disease of the central nervous system rather than a sterile
inflammatory process.
prevalence:
- population: Worldwide
measure_type: CASES_IN_LITERATURE
prevalence_class: BELOW_1_IN_1000000
notes: >-
Nine patients in nine unrelated families: five in 2023 from China, Finland, Iran (two
families) and Saudi Arabia, and four in 2026 from Iran, Oman, China and the United States.
All male, all hemizygous, eight distinct variants.
No population estimate exists. Ascertainment is through MSMD cohorts, so the denominator is
the set of patients whose mycobacterial disease prompted genetic investigation, not the
population. rate_per_100000 is left unset.
One quantity worth recording: MCTS1 is one of twenty-two genes underlying forty-seven
genetic aetiologies of MSMD as of the 2026 report. That is the population this entity is a
slice of, and it is the reason a targeted MSMD panel rather than a single-gene test is the
diagnostic route.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: BACKGROUND
snippet: "Variants of 22 genes underlie 47 different genetic etiologies of MSMD (4)."
explanation: >-
The size of the genetic differential this entity sits inside. Graded BACKGROUND because it
is a cited count in the paper's introduction rather than its own result.
progression:
- phase: Infantile BCG disease
notes: >-
Disease appears between one and twelve months in the vaccinated patients, at the site of
vaccination first and disseminating from there. The 2026 series' youngest presented at four
months with axillary adenitis and bone-marrow involvement; another at one month with local
armpit disease.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "7 mo later, he was diagnosed with abscessed BCG adenitis (BCG-itis)."
explanation: The interval from vaccination to disease in one of the 2026 patients.
- phase: Later mycobacterial disease in the unvaccinated
notes: >-
Without BCG the disease waits for an environmental mycobacterium. The American patient had a
healthy childhood with no notable infections, began losing weight and having diarrhoea at
sixteen, and was diagnosed with disseminated Mycobacterium abscessus at twenty-two. He
carries the hypomorphic p.Trp175* allele, so allele severity and absence of BCG exposure both
point the same way and cannot be separated in one patient.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "He had a healthy childhood, without frequent or notable infections. At the age of 16 years, he began to experience weight loss and diarrhea."
explanation: >-
The healthy childhood and the age of first symptoms — the natural history of an
unvaccinated patient with a hypomorphic allele.
- phase: Outcome
notes: >-
Two of nine published patients have died of mycobacterial disease: one of the original five,
and the Omani patient at six years and nine months. The rest responded to antimycobacterial
treatment. The 2023 report describes its survivors as asymptomatic after early childhood with
normal growth and development; the 2026 series' survivors are on long-term antimycobacterial
therapy, immunoglobulin replacement and in two cases IFN-gamma-1b.
The difference between those two descriptions is the most important thing in this entry's
prognosis, and it is at least partly a difference in follow-up rather than in disease.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "He developed fever and hematemesis, leading to his collapse and death within 24 h at the age of 6 years and 9 mo."
explanation: The fatal outcome in the 2026 series, with its age.
genetic:
- name: MCTS1
relationship_type: CAUSATIVE
variant_origin: GERMLINE
gene_term:
preferred_term: MCTS1
term:
id: hgnc:23357
label: MCTS1
notes: >-
The single causal gene. MCTS1 is also a well-established oncogene, overexpressed in several
tumour types, which makes the narrowness of the germline loss-of-function phenotype the
surprising part of this disease: nine hemizygous null or near-null males with mycobacterial
disease and no reported malignancy.
A note on the identifier. The deep-research report committed alongside this entry gives MCTS1
as HGNC:7838, which is NKX3-1, an unrelated homeobox gene. The correct identifier is
hgnc:23357. The same run's HLASA report made the same class of error for LARS2. Gene CURIEs
are unchecked in the research lane (the recipe passes --term-skip-prefix HGNC) and only
self-consistency-checked on the KB side, which is dismech issue #10948; the report's JAK2
identifier, by contrast, was correct.
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "These findings suggest that X-linked recessive human MCTS1 deficiency underlies isolated mycobacterial disease by impairing JAK2 translation in innate-like adaptive T lymphocytes, thereby impairing the IL-23-dependent induction of IFN-γ."
explanation: The gene-disease claim and the mechanism in one sentence, as the authors state it.
- name: JAK2
relationship_type: MODIFIER
variant_origin: GERMLINE
gene_term:
preferred_term: JAK2
term:
id: hgnc:6192
label: JAK2
notes: >-
JAK2 is not mutated in this disease. It is recorded here because it is the substrate whose
translation fails, and because the structure of its 5-prime untranslated region — not its
function — is what makes it the vulnerable one. The relationship type is MODIFIER as the
nearest available value; there is no relationship type for "gene whose translation is
selectively impaired by the causal gene's loss", and calling it CAUSATIVE would assert a
JAK2 variant that no patient has.
evidence:
- reference: PMID:37875108
reference_title: "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria."
supports: SUPPORT
directness: DIRECT
evidence_source: IN_VITRO
quote_role: PRIMARY_RESULT
snippet: "JAK2 expression is sufficiently low to impair cellular responses to interleukin-23 (IL-23) and partially IL-12, but not other JAK2-dependent cytokines."
explanation: >-
JAK2's role as the limiting substrate, and the quantitative reason the impairment is
selective for IL-23 rather than global. Graded IN_VITRO, as the same sentence is on the
pathophysiology node: the cellular responses it describes were measured in knockout cell
lines and reporter systems, not in a patient.
diagnosis:
- name: Routine immunophenotyping
presence: ABSENT
description: >-
Recorded as a negative because it is the trap. Circulating leukocyte subsets develop normally
and complete blood counts and clinical immunophenotyping are normal, so the test that would
ordinarily confirm an immunodeficiency comes back reassuring. Molecular diagnosis is the only
route.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Complete blood counts and clinical immunophenotyping results were normal (Tables S3 and S4)."
explanation: The normal routine immunology in the adult patient, whose diagnosis came from sequencing.
- name: Whole-exome sequencing
description: >-
How every published patient was diagnosed. Given that MSMD has twenty-two genes and
forty-seven aetiologies, a targeted MSMD panel or exome is the rational first test in a boy
with disseminated BCG or environmental mycobacterial disease and a normal immunophenotype.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Whole-exome sequencing of the probands revealed hemizygosity for rare germline MCTS1 variants."
explanation: The diagnostic method used across the 2026 series.
- name: Mycobacterial culture and molecular identification from the affected site
description: >-
Identifying the organism is what converts a granulomatous or necrotic lesion into a
mycobacterial diagnosis, and in this series it required a range of methods: PCR for the
M. bovis complex on axillary and bone-marrow specimens, cerebrospinal fluid culture,
mycobacterial culture from lymph node biopsy, and metagenomic next-generation sequencing on
an ileocaecal mucosal biopsy. Histology alone was insufficient in at least one case —
the Chinese patient's biopsy showed epithelial damage, crypt distortion and gastritis
without granulomas.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Histopathology revealed small intestinal villous widening with epithelial damage and increased intraepithelial lymphocytes, focal crypt distortion in the colon, and mild chronic gastritis without granulomas"
explanation: >-
A mycobacterial gastrointestinal infection with non-granulomatous histology — the reason
organism identification rather than histology is the diagnostic step recorded here.
treatments:
- name: Antimycobacterial Therapy
therapeutic_modality: SMALL_MOLECULE
description: >-
The primary treatment, and mostly effective. Regimens in the published patients range from
isoniazid and rifampicin for eighteen months, through isoniazid, rifampicin, ethambutol and
pyrazinamide, to a four-drug combination of imipenem with ceftaroline, omadacycline,
clofazimine and linezolid for the M. abscessus patient. One isolate was
pyrazinamide-resistant M. bovis BCG, which is expected — M. bovis is intrinsically
pyrazinamide-resistant — and is worth knowing before a standard antituberculous regimen is
started.
Some patients also carry long-term antibacterial prophylaxis: the youngest 2026 patient is on
daily azithromycin.
treatment_term:
preferred_term: antimycobacterial pharmacotherapy
term:
id: NCIT:C15986
label: Pharmacotherapy
therapeutic_agent:
- preferred_term: isoniazid
term:
id: CHEBI:6030
label: isoniazide
- preferred_term: rifampicin
term:
id: CHEBI:28077
label: rifampicin
- preferred_term: ethambutol
term:
id: CHEBI:4877
label: ethambutol
target_mechanisms:
- target: Disseminated BCG Disease
description: >-
Kills the organism the immune defect cannot control. It addresses the infection and leaves
the IFN-gamma induction defect intact, which is why several patients need it long-term and
why exposure to a new mycobacterium remains a risk.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "P1 was treated with isoniazid and rifampicin, initially for 4 mo, but with subsequent prolongation for 18 mo based on his clinical response and microbiological status."
explanation: >-
A worked regimen with its duration and the basis on which it was extended — the practical
detail a clinician needs.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: REFUTE
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Despite intravenous phenytoin and dexamethasone treatment and anti-infection treatment with ceftriaxone, vancomycin, and acyclovir, the patient remained febrile and lethargic."
explanation: >-
Graded REFUTE against a reading of antimicrobial treatment as reliably effective. In the
patient who died, broad empirical cover failed because the organism was mycobacterial and
the regimen was not — the antimycobacterial drugs were started eleven days after admission.
- name: Recombinant Interferon Gamma-1b
therapeutic_modality: PROTEIN_REPLACEMENT
description: >-
The mechanistically indicated therapy, and the one this entry's pathograph predicts. The
lesion is upstream of the IFN-gamma receptor and patient cells respond normally to exogenous
IFN-gamma, so supplying it bypasses the block rather than working around it.
Three of the four 2026 patients received it. In the Omani patient subcutaneous IFN-gamma-1b
improved activity, mobility and nutritional status before he was lost to follow-up and it
was stopped after six weeks; in the American patient, increasing doses over a year and a
half, reaching 400 micrograms three times weekly, produced substantial improvement after
about a year.
Those are two uncontrolled observations, one of them in a patient who then died after the
drug was stopped. They are recorded as such.
treatment_term:
preferred_term: interferon gamma therapy
term:
id: NCIT:C15986
label: Pharmacotherapy
therapeutic_agent:
- preferred_term: interferon gamma-1b
term:
id: NCIT:C100089
label: Interferon Gamma-1b
target_mechanisms:
- target: Failure of Macrophage Activation Against Intracellular Mycobacteria
description: >-
Exogenous IFN-gamma activates macrophages directly, downstream of the IL-23-to-IFN-gamma
induction step that fails. This is the node the treatment targets, and the reason it
targets that node rather than a phenotype is that the bypass is a mechanistic claim the
published cell data support.
notes: >-
therapeutic_modality is PROTEIN_REPLACEMENT rather than the immunomodulation reading the
drug's usual context invites. The lesion here is an inability to produce IFN-gamma: IL-23
signalling fails, so MAIT and gamma-delta cells do not make the cytokine. Giving it replaces
a protein the patient cannot make, in a patient whose response to it is intact. That is
replacement in the same sense immunoglobulin replacement is, and it is why the entry's
pathograph predicts the therapy rather than merely accommodating it.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Treatment with increasing doses of IFN-γ-1b for 1.5 years has started to show substantial improvements in the patient’s condition after ~1 year of treatment at a dosage of 400 µg three times per week."
explanation: >-
The dose, the schedule, the duration before response, and the response — the only
quantitative treatment observation in this disease.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "Subcutaneous IFN-γ1b treatment was initiated, leading to improvements in the patient’s activity, mobility, and nutritional status."
explanation: >-
The second reported response, in the patient with central nervous system disease. The same
paragraph records that follow-up was lost and the drug stopped after six weeks; that is in
this treatment's description rather than in the snippet, which reports only what was
observed on treatment.
- name: Immunoglobulin Replacement Therapy
therapeutic_modality: PROTEIN_REPLACEMENT
description: >-
Given to three of the four 2026 patients — monthly intravenous immunoglobulin in the youngest,
intravenous immunoglobulin in the Chinese patient for hypogammaglobulinaemia, and in the
Omani patient on the basis of severe illness with moderately low B-cell counts and IgG even
though immunoglobulin levels were normal.
That last case is worth distinguishing: it was a decision made on clinical severity rather
than on a demonstrated antibody deficiency, and it is recorded that way here.
treatment_term:
preferred_term: immunoglobulin replacement therapy
term:
id: NCIT:C62710
label: Immunoglobulin Therapy
target_mechanisms:
- target: Decreased Circulating Immunoglobulin
description: >-
Substitutes for antibody the patient does not make. It has no relationship to the
IFN-gamma induction defect, which is why it is recorded against the humoral phenotype
rather than against a mechanism node.
evidence:
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "He is on prophylaxis with azithromycin daily and IVIG monthly."
explanation: >-
The maintenance regimen in the youngest 2026 patient, covering both the antibacterial
prophylaxis and the replacement.
- reference: PMID:41623352
reference_title: "Complete and partial forms of X-linked MCTS1 deficiency in patients with mycobacterial disease."
supports: NO_EVIDENCE
directness: DIRECT
evidence_source: HUMAN_CLINICAL
quote_role: PRIMARY_RESULT
snippet: "IG levels were normal, but IVIG was administered due to the patient’s severe medical condition, moderately low B cell counts and IgG levels"
explanation: >-
Graded NO_EVIDENCE for this treatment's indication rather than SUPPORT. The sentence
records that the drug was given and states explicitly that the immunoglobulin levels did
not indicate it — so it bears on the decision, not on the claim that these patients have an
antibody deficiency needing replacement.
discussions:
- discussion_id: imd118_redundancy_claim
kind: CONTROVERSY
status: OPEN
prompt: >-
Is MCTS1-dependent translation reinitiation really "otherwise physiologically redundant", or
did the founding cohort simply have less follow-up than the second one?
rationale: >-
The 2023 report's most striking claim is that losing a general translation factor costs
nothing outside the mycobacterial axis: normal immunity, growth and development in the
survivors. It is a claim worth taking seriously — nine hemizygous null males with no
malignancy is real evidence that an established oncogene is dispensable.
The 2026 series does not contradict it directly and does sit awkwardly with it. Three of
four new patients have growth failure below the third percentile, three have
hypogammaglobulinaemia or low IgG prompting replacement, one has delayed gross motor
development, and three have chronic gastrointestinal disease. Some of that is attributable
to chronic mycobacterial infection, which causes growth failure on its own. Some of it may
not be.
The two readings make different predictions. If the extra features are infection-driven,
they should improve with mycobacterial control; if they are a direct consequence of the
translation defect, they should not. The American patient — healthy childhood, no BCG,
disease only from sixteen — is the informative case, and what is reported for him is chronic
weight loss dating from the onset of his infection rather than from birth.
This entry records both the founding claim and the later series without resolving them, and
the quoted hedge in the mechanism node is deliberate.
attaches_to:
- pathophysiology#Failure of 40S Ribosome Recycling and Translation Reinitiation
- phenotypes#Failure to Thrive
- phenotypes#Decreased Circulating Immunoglobulin
- discussion_id: imd118_ifng_bypass
kind: OPEN_QUESTION
status: OPEN
prompt: >-
Should recombinant IFN-gamma be started at diagnosis in MCTS1 deficiency rather than after
antimycobacterial therapy fails?
rationale: >-
The mechanistic case is unusually clean for a disease this rare. The defect is in the
induction of IFN-gamma, not in the response to it; patient leukocytes respond normally to
exogenous IFN-gamma and IFN-alpha; so the lesion is bypassable in principle, and the
pathograph in this entry predicts exactly where the bypass enters.
The evidence is two uncontrolled observations. In the American patient, IFN-gamma-1b at 400
micrograms three times weekly produced substantial improvement — after about a year, which
is worth noting for anyone judging an early response. In the Omani patient it improved
activity, mobility and nutrition, was stopped after six weeks when follow-up was lost, and
he died some months later; that sequence is suggestive and is not evidence of causation in
either direction.
What would settle it is not obtainable at nine patients. What is obtainable is a consistent
record of whether IFN-gamma was offered, at what dose, and when relative to the
antimycobacterial regimen — which the 2026 report supplies for its patients and the 2023
report does not for its five.
attaches_to:
- treatments#Recombinant Interferon Gamma-1b
- pathophysiology#Failure of Macrophage Activation Against Intracellular Mycobacteria
- discussion_id: imd118_bcg_in_at_risk_families
kind: OPEN_QUESTION
status: OPEN
prompt: >-
Does an unvaccinated male relative of an MCTS1-deficient proband need BCG withheld, screening
for environmental mycobacteria, or both?
rationale: >-
Eight of nine published patients were BCG-vaccinated at or near birth, and most of their
disease is BCG disease. The obvious inference is that withholding BCG from at-risk males
prevents the disease.
The American patient shows the limit of that inference. He was never vaccinated, had a
healthy childhood, and presented at twenty-two with disseminated Mycobacterium abscessus and
a retroperitoneal mass encasing the aorta. Withholding BCG did not protect him; it changed
which mycobacterium found him, and when. One 2023 patient was likewise unvaccinated and was
only two at last follow-up, so his eventual course is unknown.
The practical question this raises is not answerable from nine patients but is worth posing
explicitly: an X-linked disease with a vaccine-avoidable early presentation and an
unavoidable late one needs a surveillance answer, not only a vaccination answer. No published
source addresses it.
attaches_to:
- phenotypes#Disseminated BCG Disease
- phenotypes#Nontuberculous Mycobacterial Infection
notes: >-
Patient count and sources. Nine patients in nine unrelated families: five in PMID:37875108
(2023) and four in PMID:41623352 (2026). The 2026 paper's own count of five prior reported
variants matches the 2023 cohort, so the two sets are disjoint and the total is nine.
Scope of the 2023 abstract. Only the abstract and the opening of the 2023 paper are in the
reference cache here; its five patients' individual clinical courses are not quotable from
what is cached. Every clinical phenotype in this entry that carries a per-patient detail is
therefore evidenced from the 2026 paper, whose full text is cached. Where this entry states
something about the 2023 cohort — normal leukocyte subsets, one death, the redundancy claim —
it is quoted either from the 2023 abstract or from the 2026 paper's summary of it, and graded
BACKGROUND in the latter case.
The gene identifier. The deep-research report gives MCTS1 as HGNC:7838, which is NKX3-1. The
correct identifier is hgnc:23357. This is the second wrong gene CURIE in the same batch of
five openscientist runs — the HLASA report gave LARS2 as HGNC:21353, which is KLHL31. The
research recipe passes `--term-skip-prefix HGNC` so gene CURIEs are unchecked where they are
emitted, and `just validate-terms` compares a CURIE against its own label rather than against
the gene the entry names (dismech#10948). Both were caught here only because the entries write
`label:` out. The same report's JAK2 identifier was correct, so the error rate is not uniform
and cannot be assumed either way.
Two curator errors, both caught by term validation, both worth recording because neither felt
like guessing at the time: NCIT:C1428 was written for recombinant interferon gamma and is
Stavudine; the entry now uses NCIT:C100089 Interferon Gamma-1b. CHEBI:6030 was written with
the label "isoniazid" where CHEBI's canonical label is "isoniazide".
GeneReviews baseline: `just check-genereviews` reports NO_CHAPTER for both GeneReviews and
StatPearls against the committed Bookshelf index (snapshot 2026-09-10).
JAK2 in the `genetic:` section. JAK2 is not mutated in this disease and is recorded with
`relationship_type: MODIFIER` because there is no value for "gene whose translation the causal
gene's loss selectively impairs". CAUSATIVE would assert a JAK2 variant no patient has. This
is a schema gap rather than a curation choice, and the nearest available value is used with
the reason stated in that record's notes.
Review round 1. Two IMPORTANT findings, both taken. The mechanism chain was graded
HUMAN_CLINICAL throughout on the reasoning that the experiments used patient and
patient-derived cells. That is the inference CLAUDE.md's definition rules out: IN_VITRO covers
cultured cells whether human or not, and "patient-derived" says where a cell came from rather
than the setting the experiment ran in. The work behind those six items is MCTS1-knockout HeLa
transfection and reporter assays, 40S and 80S footprinting in HeLa, CRISPR knockout in THP-1,
HEK-Blue IL-23 reporter lines, gamma-2A fibrosarcoma transfection, and SV40-immortalised
patient fibroblasts transduced with wild-type MCTS1. The founding paper says it in its own
words, describing the impaired BCG response of patient leukocytes as measured in vitro. Those
six are now IN_VITRO; the general-immunology statement about IFN-gamma as macrophage-activating
factor is now OTHER; the phenotype and treatment items stay HUMAN_CLINICAL, which is what they
are.
This matters beyond bookkeeping. CLAUDE.md's rule that model and in-vitro evidence should not
be the sole support for a human claim only works if the grading is honest, and as written the
entry asserted that a chain established in three immortalised cell lines was direct human
clinical evidence.
The second finding was a binding that inverted its own claim: Normal Circulating Leukocyte
Subsets was bound to HP:0002721 Immunodeficiency, so any consumer reading HP terms would take
the entry to assert the abnormality that node exists to deny. The original notes flagged it as
a binding of convenience, which does not help — the disclaimer is prose and the term is
machine-readable. The term is dropped, preferred_term carries the finding, and the searches
behind that omission are recorded in the phenotype's own notes and were re-run before being
written down.
Not curated. The deep-research report cites a body of MCTS1 tumour-biology work — oncogenic
roles in laryngeal squamous cell carcinoma, breast-cancer prognostic signatures, the
OTUD6B-LIN28B axis. That is overexpression biology in cancer, not a mechanism of germline
loss, and the most interesting thing about it is precisely that it does not manifest in these
patients. It is discussed in the gene record rather than curated as pathophysiology.
Deep research results are used as seeds for research; they do not undergo the same validation as the main records and may contain errors. How we use deep research.
Record notes
Patient count and sources. Nine patients in nine unrelated families: five in PMID:37875108 (2023) and four in PMID:41623352 (2026). The 2026 paper's own count of five prior reported variants matches the 2023 cohort, so the two sets are disjoint and the total is nine. Scope of the 2023 abstract. Only the abstract and the opening of the 2023 paper are in the reference cache here; its five patients' individual clinical courses are not quotable from what is cached. Every clinical phenotype in this entry that carries a per-patient detail is therefore evidenced from the 2026 paper, whose full text is cached. Where this entry states something about the 2023 cohort — normal leukocyte subsets, one death, the redundancy claim — it is quoted either from the 2023 abstract or from the 2026 paper's summary of it, and graded BACKGROUND in the latter case. The gene identifier. The deep-research report gives MCTS1 as HGNC:7838, which is NKX3-1. The correct identifier is hgnc:23357. This is the second wrong gene CURIE in the same batch of five openscientist runs — the HLASA report gave LARS2 as HGNC:21353, which is KLHL31. The research recipe passes `--term-skip-prefix HGNC` so gene CURIEs are unchecked where they are emitted, and `just validate-terms` compares a CURIE against its own label rather than against the gene the entry names (dismech#10948). Both were caught here only because the entries write `label:` out. The same report's JAK2 identifier was correct, so the error rate is not uniform and cannot be assumed either way. Two curator errors, both caught by term validation, both worth recording because neither felt like guessing at the time: NCIT:C1428 was written for recombinant interferon gamma and is Stavudine; the entry now uses NCIT:C100089 Interferon Gamma-1b. CHEBI:6030 was written with the label "isoniazid" where CHEBI's canonical label is "isoniazide". GeneReviews baseline: `just check-genereviews` reports NO_CHAPTER for both GeneReviews and StatPearls against the committed Bookshelf index (snapshot 2026-09-10). JAK2 in the `genetic:` section. JAK2 is not mutated in this disease and is recorded with `relationship_type: MODIFIER` because there is no value for "gene whose translation the causal gene's loss selectively impairs". CAUSATIVE would assert a JAK2 variant no patient has. This is a schema gap rather than a curation choice, and the nearest available value is used with the reason stated in that record's notes. Review round 1. Two IMPORTANT findings, both taken. The mechanism chain was graded HUMAN_CLINICAL throughout on the reasoning that the experiments used patient and patient-derived cells. That is the inference CLAUDE.md's definition rules out: IN_VITRO covers cultured cells whether human or not, and "patient-derived" says where a cell came from rather than the setting the experiment ran in. The work behind those six items is MCTS1-knockout HeLa transfection and reporter assays, 40S and 80S footprinting in HeLa, CRISPR knockout in THP-1, HEK-Blue IL-23 reporter lines, gamma-2A fibrosarcoma transfection, and SV40-immortalised patient fibroblasts transduced with wild-type MCTS1. The founding paper says it in its own words, describing the impaired BCG response of patient leukocytes as measured in vitro. Those six are now IN_VITRO; the general-immunology statement about IFN-gamma as macrophage-activating factor is now OTHER; the phenotype and treatment items stay HUMAN_CLINICAL, which is what they are. This matters beyond bookkeeping. CLAUDE.md's rule that model and in-vitro evidence should not be the sole support for a human claim only works if the grading is honest, and as written the entry asserted that a chain established in three immortalised cell lines was direct human clinical evidence. The second finding was a binding that inverted its own claim: Normal Circulating Leukocyte Subsets was bound to HP:0002721 Immunodeficiency, so any consumer reading HP terms would take the entry to assert the abnormality that node exists to deny. The original notes flagged it as a binding of convenience, which does not help — the disclaimer is prose and the term is machine-readable. The term is dropped, preferred_term carries the finding, and the searches behind that omission are recorded in the phenotype's own notes and were re-run before being written down. Not curated. The deep-research report cites a body of MCTS1 tumour-biology work — oncogenic roles in laryngeal squamous cell carcinoma, breast-cancer prognostic signatures, the OTUD6B-LIN28B axis. That is overexpression biology in cancer, not a mechanism of germline loss, and the most interesting thing about it is precisely that it does not manifest in these patients. It is discussed in the gene record rather than curated as pathophysiology.
Create: Immunodeficiency 118 (MCTS1) · 2026-09-22T18:16:47Z · View source
De novo curation of IMD118 (MCTS1, MONDO:0958030) from the 2023 Cell report that defined the entity and a 2026 four-family follow-up. Deep research: one openscientist run (16 references, 16/16 resolved, 5/16 judged on topic, term_validation needs_review true). The report is the strongest of the five in this batch on mechanism and supplied the JAK2 start-stop uORF account, the IL-23 limb placement and the seventeen-gene reporter screen. Its gene CURIE for MCTS1, HGNC:7838, is NKX3-1; the correct identifier hgnc:23357 was substituted and the error recorded, the second such error in this batch after LARS2 in the HLASA run. Its JAK2 identifier was correct. The substantive divergence from the report is prognostic: the report follows the 2023 framing of isolated mycobacterial disease with otherwise normal immunity and generally favourable outcome, and the 2026 series adds a death from CNS BCG disease, growth failure below the third percentile in three of four, hypogammaglobulinaemia prompting replacement in three of four, and chronic gastrointestinal mycobacterial disease. Both readings are recorded, with the founding redundancy claim quoted with its own hedge intact and the tension carried as a CONTROVERSY discussion rather than resolved. Two REFUTE and one NO_EVIDENCE evidence item keep counter-evidence and a non-indication in place. Two curator errors caught by term validation: NCIT:C1428 written from memory for recombinant interferon gamma is Stavudine, now NCIT:C100089; CHEBI:6030's canonical label is isoniazide not isoniazid. Validation on the final tree: just validate, validate-terms, validate-disorders, count-verified-snippets 37/37, check-entity-refs, check-causal-targets, check-qualifier-terms, check-duplicate-keys, check-folded-hyphens, check-snippet-length, check-title-snippets, check-snippet-grading, check-enum-values, check-coarse-phenotypes, check-reference-titles, check-case-collisions, check-genereviews (NO_CHAPTER), list-disconnected-phenotypes 7/9. Review round 1 (ai4c-reviewer, CHANGES_REQUESTED, two IMPORTANT findings, both taken, plus one of two suggestions). First: the whole mechanism chain was graded HUMAN_CLINICAL on the reasoning that the experiments used patient and patient-derived cells, which is the inference CLAUDE.md's definition rules out - IN_VITRO covers cultured cells whether human or not. The work behind those items is MCTS1-knockout HeLa transfection and reporter assays, 40S/80S footprinting, CRISPR knockout in THP-1, HEK-Blue IL-23 reporter lines, gamma-2A fibrosarcoma transfection, and SV40-immortalised patient fibroblasts; the founding paper itself describes the impaired BCG response of patient leukocytes as measured in vitro. Seven items regraded to IN_VITRO and one general-immunology statement to OTHER; phenotype and treatment items left HUMAN_CLINICAL. Second: Normal Circulating Leukocyte Subsets was bound to HP:0002721 Immunodeficiency, a binding that inverts the claim the node exists to make. The original notes flagged it as a binding of convenience, which does not help, since the disclaimer is prose and the term is machine-readable. The term is dropped, preferred_term carries the finding, and the phenotype's notes record the searches - l~leukocyte, l~lymphocyte count, l~immunophenotyp, l~normal against ols:hp - which were re-run before the note was written, per the CLAUDE.md rule that naming a query does not make the note true. HPO has terms for abnormality and none for a documented normal result in this domain; HP:0002372 Normal interictal EEG is cited in the note as the shape a suitable term would take. Also added a notes: paragraph on why IFN-gamma-1b is PROTEIN_REPLACEMENT rather than immunomodulation. Declined the suggestion to promote hepatomegaly, splenomegaly, osteomyelitis, fever and abscess from prose to phenotype nodes: the 2023 abstract lists them without per-patient attribution and the 2026 full text attributes its own, so promoting them would mean denominators this entry cannot source. One gate catch worth recording: regrading the JAK2 sentence on the pathophysiology node left the same sentence graded HUMAN_CLINICAL on the JAK2 gene record, and check-snippet-grading failed - which is precisely the divergence that gate exists for. Both now IN_VITRO. Post-round validation: count-verified-snippets 37/37, all gates green.
Disease: Immunodeficiency 118 (IMD118) MONDO ID: MONDO:0958030 · OMIM: #301115 · Gene: MCTS1 (Xq24) Category: Mendelian — X-linked recessive inborn error of immunity (a subtype of Mendelian Susceptibility to Mycobacterial Disease, MSMD) Primary reference: Bohlen J, Zhou Q, Philippot Q, et al. "Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria." 2023 (PMID: 37875108).
Immunodeficiency-118 (IMD118) is an ultra-rare, X-linked recessive inborn error of immunity caused by hemizygous loss-of-function mutations in MCTS1 on chromosome Xq24. The disease was defined by a single landmark study — Bohlen, Zhou, Béziat, Casanova and colleagues (2023, PMID: 37875108) — which identified complete MCTS1 deficiency in 5 unrelated males from kindreds of different ancestries (China, Finland, Iran, and Saudi Arabia). Clinically, IMD118 belongs to the MSMD spectrum: affected males present in infancy with disseminated infection by weakly virulent mycobacteria, most commonly after Bacillus Calmette-Guérin (BCG) vaccination, while otherwise displaying essentially normal immunity, growth, and development.
The mechanism is a specific and elegant translational defect. MCTS1 (Malignant T-cell-amplified sequence 1) is a translation re-initiation and 40S ribosome-recycling factor that forms a heterodimer with DENR. Complete MCTS1 deficiency impairs the translation of a small subset of proteins — most importantly the tyrosine kinase JAK2 — in all cell types tested. JAK2 loss cripples signaling downstream of IL-23 (and partially IL-12), which in turn reduces IFN-γ production by innate-like adaptive MAIT (mucosal-associated invariant T) and γδ T lymphocytes upon mycobacterial challenge. Because IFN-γ is the central macrophage-activating cytokine controlling intracellular mycobacteria, its failure produces the characteristic mycobacterial susceptibility. The selectivity of the JAK2 defect is explained by two ultra-short "start-stop" upstream open reading frames (stuORFs) in the JAK2 5′UTR that render JAK2 translation uniquely dependent on MCTS1-mediated ribosome recycling.
Prognosis is generally favorable but not benign. Among the 5 reported patients, one died of disseminated mycobacterial disease while the other four responded to antimycobacterial therapy and remained asymptomatic after early childhood, with normal growth and development. Critically, patient leukocytes retained normal responses to exogenous IFN-γ and IFN-α, meaning the block is upstream of the IFN-γ receptor and is potentially bypassable — providing a clear rationale for recombinant IFN-γ as adjunctive therapy, as used successfully in other MSMD genotypes. A notable paradox is that MCTS1 is an established oncogene, yet its complete germline loss produces only a narrow immunological phenotype rather than developmental defects or cancer, indicating that its translational function is largely physiologically redundant except along the IL-23→JAK2→IFN-γ axis.
IMD118 (OMIM #301115) is caused by hemizygous loss-of-function variants in MCTS1 (Xq24) and was described in 5 unrelated males from four countries. As the primary report states: "We report X-linked recessive MCTS1 deficiency in men with mycobacterial disease from kindreds of different ancestries (from China, Finland, Iran, and Saudi Arabia)" (PMID: 37875108).
The five patients carried a spectrum of loss-of-function variant types: 2 frameshift variants, 1 splice-site variant, and 2 in-frame 3′ deletions, the latter two both yielding a truncated protein lacking residues Ala133–Lys181 (p.Ala133_Lys181del). All variants result in complete MCTS1 deficiency. The molecular consequence is defined by the authors: "Complete deficiency of this translation re-initiation factor impairs the translation of a subset of proteins, including the kinase JAK2 in all cell types tested, including T lymphocytes and phagocytes." The downstream immunological consequence connects the translation defect to disease: "Defective responses to IL-23 preferentially impair the production of IFN-γ by innate-like adaptive mucosal-associated invariant T cells (MAIT) and γδ T lymphocytes upon mycobacterial challenge."
This establishes IMD118 as a single-gene, X-linked recessive disorder in which a general translation factor defect is funneled into a strikingly narrow clinical outcome via a single critical substrate (JAK2) and a single critical pathway (IL-23→IFN-γ).
MCTS1 (also called MCT-1; NCBI Gene 28985; Xq24) is a translation factor that partners with DENR (density-regulated reinitiation factor). Together MCTS1/DENR are the mammalian counterparts of the yeast Tma20/Tma22 heterodimer, with eIF2D (yeast Tma64) providing a parallel activity. These factors recycle post-termination 40S ribosomal subunits at stop codons and modulate reinitiation on mRNAs containing short upstream open reading frames (uORFs).
The functional role was demonstrated in yeast and in vitro systems: "Tma64 (eIF2D), Tma20 (MCT-1), and Tma22 (DENR) function as 40S recycling factors in vitro" (PMID: 30146315). The reinitiation-promoting property that is mechanistically central to IMD118 was shown by Jendruchová et al.: "MCTS1/DENR enhance reinitiation at short upstream open reading frames (uORFs) harboring penultimate codons that confer dependence on these factors in bulk 40S recycling" (PMID: 38903097). When MCTS1 is absent, 80S ribosomes queue behind stop codons and aberrant reinitiation occurs in 3′UTRs — the general failure that, at the JAK2 locus, specifically prevents translation of the main JAK2 open reading frame.
A striking feature of IMD118 is the mismatch between MCTS1's known oncogenic biology and the narrow phenotype of its germline loss. MCTS1/DENR are established oncogenes: "DENR and MCTS1 have been identified as oncogenes in several different tumor entities. The heterodimeric DENR·MCTS1 protein complex promotes translation of mRNAs containing upstream Open Reading Frames (uORFs)" (PMID: 35115540). The complex is cell-cycle regulated (DENR Ser73 phosphorylated by Cyclin B/CDK1 and Cyclin A/CDK2), and MCTS1 overexpression drives proliferation in laryngeal squamous cell carcinoma via the OTUD6B-LIN28B axis and LARP7 stabilization (PMID: 37634410, PMID: 35274760), and appears in breast-cancer prognostic gene signatures (PMID: 35441810, PMID: 35120331).
Despite this oncogenic role, the 5 hemizygous MCTS1-null males exhibited only isolated mycobacterial disease with otherwise normal immunity, growth, development, and no reported malignancy. The authors note: "the lack of MCTS1-dependent translation re-initiation and ribosome recycling seems to be otherwise physiologically redundant in these patients" (PMID: 37875108). This tells us that, in humans, MCTS1's translational function is largely dispensable for viability and development, and that its physiological non-redundancy is confined to the anti-mycobacterial IFN-γ axis.
IMD118 is mechanistically embedded within the well-characterized IL-12/IL-23/IFN-γ circuit whose disruption defines MSMD. The essential role of IL-23 was established by Philippot et al.: "Human IL-23 is essential for IFN-γ-dependent immunity to mycobacteria" (PMID: 36763636). Independent genetic validation comes from IL23R deficiency: a homozygous R381X mutation in IL23R causes MSMD, where "impaired IL-23 immunity caused by a homozygous R381X mutation in IL23R underlies MSMD" (PMID: 35829840), producing impaired IL-23-mediated STAT3 phosphorylation, reduced IFN-γ secretion, and isolated disseminated BCG/NTM disease.
The effector cells are innate-like T cells. MAIT cells recognize mycobacterial vitamin-B metabolites presented on MR1 and produce IFN-γ, TNF, and granzyme B, and are reduced/exhausted in active tuberculosis (PMID: 37147816, PMID: 41707313). MCTS1 deficiency impairs JAK2 → IL-23 signaling → IFN-γ from MAIT/γδ cells, placing IMD118 firmly in the same functional pathway as IL12B, IL12RB1, IL23R, IFNGR1/2, STAT1, and the other MSMD genes.
The OMIM #301115 Clinical Synopsis, derived from Bohlen et al. 2023, describes the clinical picture of the 5 unrelated males (aged 3–18 years at report). Disseminated mycobacterial disease developed between 3 and 12 months of age, usually after BCG vaccination. Reported organ-system features and suggested HPO terms include:
| Feature | HPO term |
|---|---|
| Hepatomegaly | HP:0002240 |
| Splenomegaly | HP:0001744 |
| Osteomyelitis | HP:0002754 |
| Fever | HP:0001945 |
| Lymphadenopathy / enlarged lymph nodes | HP:0002716 |
| Abscess | HP:0025615 |
| Disseminated mycobacteriosis | (BCG/NTM) |
Immunologically, patients had normal numbers of circulating leukocyte subsets — routine immune workup is typically unrevealing, a hallmark of MSMD that necessitates molecular diagnosis. Most patients recover with treatment and remain asymptomatic after early childhood. The source data underlie the OMIM synopsis: "We report X-linked recessive MCTS1 deficiency in men with mycobacterial disease from kindreds of different ancestries" (PMID: 37875108).
The molecular reason JAK2 is uniquely vulnerable among the many proteins translated in the cell was resolved by 5′UTR reporter assays. Among 17 MSMD/IFN-γ-immunity genes tested in MCTS1-KO HeLa cells, only JAK2 showed a >65% decrease in translation, which was rescued by wild-type MCTS1 but not by patient variants or a synthetic loss-of-function control (A109D). The JAK2 5′UTR contains three uORFs, two of which are ultra-short start-stop "stuORFs" (uORF1, uORF2). In the absence of MCTS1, 40S ribosomes stall (a "roadblock") at these stuORF stop codons, blocking re-initiation at the JAK2 main ORF — confirmed directly by accumulation of 40S ribosomal footprints. CRISPR MCTS1-knockout in THP-1 monocytes reproducibly lowered endogenous JAK2 across 5 independent clones.
Functionally, patient leukocytes had impaired IFN-γ production after BCG stimulation and impaired IL-23 responses (but normal IL-12 responses), yet retained NORMAL responses to exogenous IFN-γ and IFN-α. This localizes the defect upstream of the IFN-γ receptor and predicts that exogenous IFN-γ can bypass the block.
The protein architecture underpinning MCTS1's role was characterized earlier: "MCT-1 contains the PUA domain, a recently described RNA-binding domain that is found in several tRNA and rRNA modification enzymes... MCT-1 protein interacts with the cap complex through its PUA domain and recruits the density-regulated protein (DENR/DRP), containing the SUI1 translation initiation domain" (PMID: 16982740).
Of the 5 hemizygous MCTS1-deficient males, one died of disseminated mycobacterial disease; the remaining four responded to antimycobacterial treatment, had no other infections, and showed normal growth and development. The redundancy of MCTS1 outside the mycobacterial axis is emphasized by the authors: "Surprisingly, the lack of MCTS1-dependent translation re-initiation and ribosome recycling seems to be otherwise physiologically redundant in these patients" (PMID: 37875108). The preserved response to exogenous IFN-γ and IFN-α provides both a favorable prognostic indicator and a therapeutic target.
Branch point (bypass): Because the lesion is upstream of the IFN-γ receptor, exogenous IFN-γ can activate macrophages directly (patient cells respond normally to IFN-γ and IFN-α) → therapeutic bypass and generally favorable prognosis.
MCTS1 LoF (Xq24, hemizygous male)
│ loss of 40S recycling / reinitiation factor
▼
JAK2 stuORF roadblock in 5'UTR ──► 40S ribosome stall
│ selective translational block (>65% ↓ JAK2)
▼
↓ JAK2 protein (all cell types)
│
▼
impaired IL-23 signaling (± partial IL-12) [IL-12 response NORMAL]
│
▼
↓ IFN-γ from MAIT & γδ T cells (mycobacterial challenge)
│
▼
macrophages not activated ──► uncontrolled mycobacteria
│ ▲
▼ │ BYPASS
disseminated mycobacterial disease exogenous IFN-γ (response intact)
(fever, HSM, lymphadenopathy,
abscess, osteomyelitis; infancy)
IMD118 is an X-linked recessive inborn error of immunity within the MSMD spectrum, characterized by selective susceptibility to weakly virulent mycobacteria (BCG vaccine strain and non-tuberculous mycobacteria). Identifiers: OMIM #301115; MONDO:0958030; gene MCTS1 (Xq24). ICD-11 maps to inborn errors of immunity / MSMD categories; no dedicated ICD-10 code. Synonyms: MCTS1 deficiency, X-linked MCTS1 deficiency, MSMD due to MCTS1 deficiency. The information is derived from an aggregated disease-level primary report (Bohlen et al. 2023, 5 individual patients) and OMIM, not from EHR-scale data.
Causal factor: monogenic — hemizygous loss-of-function variants in MCTS1. Genetic risk factor: male sex (X-linked recessive; a single mutant X allele is sufficient in males). Environmental trigger: BCG vaccination and exposure to environmental (non-tuberculous) mycobacteria precipitate disease; the mutation is necessary, mycobacterial exposure is the environmental precipitant. No protective genetic variants are described; exogenous IFN-γ acts as an acquired protective/therapeutic factor because responses to it are intact. Gene–environment interaction: the classic MSMD interaction — an inherited IL-23→IFN-γ axis defect that becomes clinically manifest only upon mycobacterial exposure (especially live BCG vaccine).
Predominantly infectious/physical manifestations rather than behavioral. Onset is infantile (3–12 months), typically post-BCG. Features (with qualitative frequencies given the 5-patient series): disseminated mycobacteriosis (defining), fever (HP:0001945), lymphadenopathy (HP:0002716), hepatomegaly (HP:0002240), splenomegaly (HP:0001744), abscess (HP:0025615), osteomyelitis (HP:0002754). Severity is variable — from recoverable disseminated disease to fatal dissemination. Progression is episodic/acute during active infection, with recovery and asymptomatic status after early childhood in survivors. Quality-of-life impact is concentrated in infancy/early childhood during active infection; survivors have generally normal function thereafter.
Causal gene: MCTS1 (Xq24; NCBI Gene 28985; HGNC:7838). Variant types among 5 patients: 2 frameshift, 1 splice-site, 2 in-frame 3′ deletions (both → p.Ala133_Lys181del). Classification: pathogenic loss-of-function (ACMG). Zygosity/origin: germline, hemizygous in males. Functional consequence: complete loss of function → loss of 40S recycling/reinitiation → selectively reduced JAK2 translation. Allele frequency: extremely rare/private; not established as recurrent in gnomAD. Modifier genes: none formally identified. Epigenetics/chromosomal abnormalities: none reported; the mechanism is translational, not epigenetic.
The key environmental factors are mycobacterial exposures — live attenuated BCG vaccine (Mycobacterium bovis BCG) and environmental non-tuberculous mycobacteria (NTM). No toxin, radiation, pollution, occupational, or lifestyle factors are implicated. Infectious agents are causative triggers acting on the genetic substrate rather than independent causes.
See the Mechanistic Model section above for the full ordered causal chain. In brief: MCTS1 LoF → failure of 40S ribosome recycling/reinitiation → JAK2 stuORF roadblock → selective JAK2 depletion → impaired IL-23 (and partial IL-12) signaling → reduced IFN-γ from MAIT/γδ T cells → failed macrophage activation → disseminated mycobacterial disease. Molecular pathway: JAK-STAT (JAK2) / IL-23→IFN-γ axis. Cellular process: cytokine signaling and antimicrobial macrophage activation. Protein dysfunction: loss of function of a ribosome-recycling factor (MCTS1) causing loss of a downstream kinase (JAK2). Immune involvement: immunodeficiency (not autoimmunity).
Primary organs: reticuloendothelial/lymphoid system — liver (UBERON:0002107), spleen (UBERON:0002106), lymph nodes (UBERON:0000029), bone/bone marrow (osteomyelitis; UBERON:0001474). Body systems: immune/hematopoietic and, via sites of mycobacterial dissemination, potentially multi-organ. Cell populations: MAIT cells (CL:0000940), γδ T cells (CL:0000798), macrophages/monocytes (CL:0000235/CL:0000576) — because the JAK2 defect is present in all cell types tested, including phagocytes. Subcellular: cytosolic ribosome/translation machinery (GO:0022626). Lateralization: not applicable (systemic/bilateral dissemination).
Onset: infantile, 3–12 months, typically after BCG vaccination; onset pattern acute/subacute. Course: episodic active disease during infection with recovery in survivors; survivors are typically asymptomatic after early childhood. Critical period: early infancy around BCG exposure is the window of vulnerability and of opportunity for early recognition/intervention. Remission: treatment-induced with antimycobacterial therapy (± IFN-γ).
Inheritance: X-linked recessive; affects males, mothers are obligate carriers. Penetrance: appears high in exposed hemizygous males but based on only 5 patients. Expressivity: variable (recovery to death). Epidemiology: ultra-rare — 5 reported patients worldwide; prevalence/incidence not estimable. Populations: four ancestries reported (Chinese, Finnish, Iranian, Saudi Arabian) — no single founder population. Sex ratio: essentially all affected are male (X-linked recessive). Consanguinity: relevant for autosomal recessive MSMD genes but less so here given X-linkage; the geographic spread suggests independent (private) mutations rather than a founder effect. Carrier frequency: unknown/very low.
Routine immune workup is typically normal (normal circulating leukocyte subsets), so diagnosis rests on molecular genetics. Recommended approach: whole-exome or whole-genome sequencing / targeted MSMD gene panels including MCTS1, prompted by disseminated BCG/NTM disease in an otherwise healthy infant. Functional confirmation: reduced JAK2 protein, impaired IFN-γ after BCG stimulation, impaired IL-23 (but normal IL-12) responses, and preserved responses to exogenous IFN-γ/IFN-α — the last both diagnostic and therapeutically informative. Microbiology: blood/tissue mycobacterial culture and, increasingly, metagenomic next-generation sequencing (mNGS) to identify unculturable NTM. Differential diagnosis: other MSMD genotypes — IL12RB1, IL12B, IFNGR1/2, STAT1, ISG15, IRF8, IL23R, RORC, NEMO, CYBB — distinguished by their specific gene defects and, in some, broader infectious susceptibility.
Generally favorable but potentially fatal: 1 of 5 patients died of disseminated mycobacterial disease; 4 recovered with antimycobacterial therapy, had no other infections, and showed normal growth and development. Prognostic factors: early diagnosis and prompt antimycobacterial (± IFN-γ) therapy; intact response to exogenous IFN-γ is a favorable feature. Long-term survivors are typically asymptomatic after early childhood. Given the tiny cohort, survival statistics are indicative rather than definitive.
Antimycobacterial therapy is the cornerstone (multidrug regimens directed at BCG/NTM). Recombinant IFN-γ is a rational adjunct because patient cells respond normally to it, bypassing the upstream block — a strategy with established benefit in other MSMD genotypes (e.g., partial IFNGR1 and IL12RB1 defects). Hematopoietic stem cell transplantation is a consideration in severe/refractory MSMD generally, though not specifically reported for IMD118. No pharmacogenomic, gene, RNA, or targeted small-molecule therapies exist yet for MCTS1 deficiency. Suggested NCIT terms: recombinant interferon gamma; antimycobacterial agent; hematopoietic stem cell transplantation.
Primary prevention: in families with a known MCTS1 variant, avoid live BCG vaccination in at-risk male infants until immune status is clarified — BCG is the principal precipitant. Secondary prevention: early recognition of disseminated mycobacterial disease and prompt therapy. Genetic counseling: X-linked recessive inheritance — carrier mothers have a 50% chance of transmitting the variant; sons of carriers have a 50% risk of being affected, daughters a 50% chance of being carriers. Carrier/prenatal testing and cascade testing are available once the familial variant is known. Tertiary prevention: antimycobacterial prophylaxis and monitoring in affected individuals.
Species affected: human (Homo sapiens, NCBI Taxon 9606) — the disease is described only in humans. Mechanistic conservation is strong: the MCTS1/DENR–eIF2D recycling system has direct yeast counterparts (Tma20/Tma22/Tma64), demonstrating deep evolutionary conservation of the underlying translation machinery (PMID: 30146315, PMID: 38903097). Orthologous Mcts1 exists in mouse and other vertebrates. No naturally occurring animal disease (OMIA) is reported. No zoonotic component.
No dedicated Mcts1-deficient mammalian disease model of IMD118 has been reported. In vitro/cellular models were central to the discovery: MCTS1-knockout HeLa cells (5′UTR reporter screens), CRISPR MCTS1-knockout THP-1 monocytes (endogenous JAK2 reduction across 5 clones), and patient-derived leukocytes. Yeast (Saccharomyces cerevisiae, Tma20/Tma22/Tma64) is a powerful model for the ribosome-recycling/reinitiation mechanism itself. These systems recapitulate the molecular defect (JAK2 translational block, ribosome stalling) but not the whole-organism immunophenotype; a conditional or humanized mouse expressing the JAK2 stuORF-containing 5′UTR would be needed to model the immunological disease in vivo.
| PMID | Title (abbrev.) | Role in this report |
|---|---|---|
| 37875108 | Human MCTS1-dependent translation of JAK2 is essential for IFN-γ immunity to mycobacteria | Primary/defining source. Identifies IMD118, the 5 patients, variant spectrum, JAK2 stuORF mechanism, IL-23→IFN-γ/MAIT-γδ axis, outcomes (1 death), preserved exogenous IFN-γ response. |
| 30146315 | Tma64/eIF2D, Tma20/MCT-1, Tma22/DENR recycle post-termination 40S subunits in vivo | Establishes MCTS1/DENR/eIF2D as 40S recycling factors; evolutionary conservation. |
| 38903097 | Impacts of yeast Tma20/MCTS1, Tma22/DENR and Tma64/eIF2D on translation reinitiation | Shows MCTS1/DENR enhance reinitiation at uORFs — mechanism underlying JAK2 dependence. |
| 16982740 | MCT-1 protein interacts with the cap complex... | Defines MCTS1 PUA domain, cap-complex interaction, DENR recruitment. |
| 35115540 | Cyclin B/CDK1 and Cyclin A/CDK2 phosphorylate DENR... | Establishes MCTS1/DENR as oncogenes and cell-cycle-regulated uORF translation factors. |
| 37634410 / 35274760 | MCTS1 in laryngeal squamous cell carcinoma | Oncogenic role of MCTS1 (contrast with benign germline-loss phenotype). |
| 36763636 | Human IL-23 is essential for IFN-γ-dependent immunity to mycobacteria | Validates the IL-23→IFN-γ axis that MCTS1/JAK2 loss disrupts. |
| 35829840 | Homozygous stop mutation in IL23R causes MSMD | Independent genetic validation that disrupting IL-23 signaling causes MSMD. |
| 37147816 / 41707313 | MAIT cells in mycobacterial immunity | Effector-cell biology: MAIT cells produce IFN-γ via MR1, reduced in active TB. |
| 40656276 / 41786143 / 41048447 | MSMD overviews & management | Context for MSMD classification, diagnostics, and IFN-γ/antimycobacterial/HSCT management. |
Evidence source types: IMD118-defining data are human clinical + in vitro/cellular (Bohlen 2023); mechanistic recycling/reinitiation data are in vitro + model organism (yeast); pathway validation is human genetic (IL23R, IL-23); oncogenic role is in vitro / tumor genomics.
Report compiled from 7 confirmed findings and 35 reviewed papers over 5 investigation iterations. All disease-specific clinical and mechanistic claims trace primarily to Bohlen et al. 2023 (PMID: 37875108); mechanistic and pathway claims are corroborated by the cited translation-machinery and MSMD literature.
Checked with linkml-reference-validator 0.2.1.
| Outcome | Count |
|---|---|
| References checked | 16 |
| Resolved | 16 |
| Unresolved (possible confabulation) | 0 |
| Unverifiable | 0 |
| References weighed for topical relevance | 16 |
| On topic | 5 |
| Off topic | 0 |
All extracted references resolved successfully.
Checked with linkml-term-validator 0.4.5, through the ols: adapter.
| Outcome | Count |
|---|---|
| Terms checked | 21 |
| Resolved | 19 |
| Unresolved (possible confabulation) | 0 |
| Obsolete | 0 |
| Unverifiable | 2 |
| Terms whose name was checked | 12 |
| Terms named correctly | 7 |
| Terms named as a different term | 2 |
| Terms whose name is worth a second look | 3 |
These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:
MONDO:0958030 (3 mentions) - the report calls it "IMD118"; MONDO calls it immunodeficiency 118UBERON:0002107 (2 mentions) - the report calls it "UBERON anatomy: liver", "liver"; UBERON calls it liver**The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:
HP:0002716 (2 mentions) - the report calls it "Lymphadenopathy / enlarged lymph nodes"; HP calls it LymphadenopathyGO:0006446 (1 mention) - the report calls it "GO biological process: regulation of translational initiation"; GO calls it regulation of translational initiation**GO:0022626 (2 mentions) - the report calls it "GO cellular component: cytosolic ribosome"; GO calls it cytosolic ribosome**The report gives these identifiers more than one name of its own:
HGNC:7838 - called "MCTS1", "Gene/protein:* MCTS1"UBERON:0002107 - called "UBERON anatomy:** liver", "liver"