Type 2 diabetes mellitus may be associated with a novel mitochondrial tRNAThr/tRNAPro mutation

Abstract

BACKGROUND

Mutations in mitochondrial (mt) transfer RNAs (tRNAs) represent a significant genetic risk factor for type 2 diabetes mellitus, though their underlying pathogenic mechanisms remain incompletely elucidated.

AIM

To investigate the potential pathogenic role of a novel diabetes related mt-tRNA^Thr/tRNA^Pro C15954T mutation.

METHODS

A four-generation Han Chinese pedigree exhibiting maternal inheritance of type 2 diabetes mellitus was underwent clinical, genetic and molecular analyses. Mitochondrial genome mutations were screened via polymerase chain reaction-Sanger sequencing, and mt function was determined in cybrid cells derived from four affected individuals carrying the m.C15954T mutation and four control subjects without the mutation.

RESULTS

Matrilineal relatives within the pedigree displayed heterogeneous clinical manifestations of type 2 diabetes mellitus, with the age of onset ranging from 41 years to 66 years (mean: 52 years). Whole mitochondrial genome sequencing identified a novel m.C15954T mutation located adjacent to the mt-tRNA^Thr and mt-tRNA^Pro genes. This mutation affects a phylogenetically conserved nucleotide which is critical for tRNA 3’-end processing and function. Biochemical assays demonstrated that cybrids carrying the m.C15954T mutation reduced tRNA^Thr and tRNA^Pro steady-state levels. In addition, the mt DNA copy number, adenosine triphosphate production, mt membrane potential, mt-RNA transcription, oxidative phosphorylation enzyme activities were markedly decreased. In contrast, reactive oxygen species levels were elevated.

CONCLUSION

These findings indicate that the m.C15954T mutation leads to mt dysfunctions and contributes to the pathogenesis of type 2 diabetes in this Chinese pedigree.

Key Words: Type 2 diabetes mellitus; Mitochondrial tRNA^Thr/tRNA^Pro; m.C15954T mutation; Mitochondrial dysfunctions; Adenosine triphosphate

Core Tip: This study reports a Han Chinese pedigree with maternally inherited type 2 diabetes mellitus. We identified a novel mitochondrial (mt) C15954T mutation located in the intergenic spacer between the mt-tRNA^Thr and mt-tRNA^Pro genes. Phylogenetic analysis revealed that the nucleotide at position 15954 is highly conserved across species, suggesting a potential role in the 3’ end processing of both tRNA^Thr and tRNA^Pro. Using trans-mitochondrial cybrid models, we demonstrated that the m.C15954T mutation leads to impaired mt-transfer RNA metabolism and significant mt dysfunction. These findings indicate that the m.C15954T mutation may contribute to the pathogenesis of diabetes in this pedigree.

INTRODUCTION

Diabetes mellitus (DM) is a heterogeneous metabolic disorder characterized by chronic hyperglycemia. Broadly, it can be categorized into idiopathic and hereditary forms. Idiopathic DM includes insulin-dependent (type 1 DM) and non-insulin-dependent [type 2 DM (T2DM)], which accounts for approximately 5% of the global population[1], and is frequently associated with persistent hyperglycemia and insulin resistance (IR)[1]. The pathogenesis of T2DM is influenced by both genetic predisposition and environmental factors[2]. Mutations in several candidate genes, including the AGTR1[3], TGF-β1[3], HNF1A/HNF4A[4], and ApoE[5] have been linked to disease development. Additionally, epidemiologic studies suggest a role for maternally inherited factors in T2DM[6-8], underscoring the contribution of mitochondrial (mt) genetic background on disease susceptibility. Since the initial identification of a 10.4-kb mtDNA deletion[9] and the transfer RNA (tRNA)^Leu(UUR) A3243G mutation[10] associated with T2DM, numerous studies have established that mtDNA mutations, particularly in mt-tRNA genes, play significantly role in this disease[11-13]. However, matrilineal relatives within and across families carrying such mt-tRNA mutation display considerable variability in penetrance, clinical severity, and age-of-onset of T2DM, and may even remain unaffected. This highlights the importance of genetic counseling, particularly for unaffected carriers of these mtDNA mutations[14].

As key adapter molecules that translate genetic code into amino acid sequences, mt-tRNAs are indispensable for mt protein synthesis and respiratory chain function[15]. Despite comprising only approximately 10% of the mtDNA coding capacity, mt-tRNA genes harbor a disproportionately high number of pathogenic mutations. In fact, the MITOMAP database indicates that they account for over half of all known disease-causing mtDNA variants[16]. This highlights their exceptional functional importance. Structurally, most canonical mt-tRNA adopt a cloverleaf conformation consisting of an acceptor arm, a D-arm, an anticodon stem, a variable region, and a TψC loop[16]. However, due to the multifactorial etiology of T2DM, the precise role of mt-tRNA mutation in the pathogenesis of the disease remains largely elusive.

In a recent systematic mutation screening of 250 T2DM patients and 255 control subjects from Hangzhou, Zhejiang Province, we identified several mutations including mt-tRNA^Ala A5587G[17], mt-tRNA^Trp A5514G[18], and mt-tRNA^Leu(UUR) A3243G and ND6 T14502C mutations[19] that potentially modulate the risk of T2DM. In this study, we reported the clinical, genetic, and biochemical features of a Han Chinese family affected by T2DM. Molecular analysis identified a novel m.C15954T located in the intergenic spacer between mt-tRNA^Thr and mt-tRNA^Pro. Notably, a previously reported m.A4401G mutation, situated at the junction of mt-tRNA^Gln and mt-tRNA^Met, has been shown to disrupt mt-tRNA processing and cause abnormal tRNA metabolism[20]. Similarly, the m.C15954T mutation occurs near the 3’ ends of mt-tRNA^Thr and mt-tRNA^Pro. Mutations in this region may interfere with RNase Z-mediated 3’ end processing and CCA addition catalyzed by tRNA nucleotidyl transferase 1[21,22]. We therefore hypothesize that the m.C15954T mutation could similarly disrupt tRNA metabolism, leading to mt dysfunction relevant to T2DM. To test this hypothesis, we established cytoplasmic hybrid (cybrid) cell lines from four patients carrying the m.C15954T mutation and four control individuals without the mutation.

MATERIALS AND METHODS

Pedigree information

A Han Chinese family was recruited through Hangzhou First People’s Hospital (Figure 1A). Additionally, 255 healthy individuals from the same geographic region (100 males and 155 females, aged 44-55 years, mean age 49 years) were enrolled as controls. The study was approval by the Ethics Committee of Hangzhou First People’s Hospital (approval No. KY-20240327-0100-01), and written informed consent was obtained from all participants.

Figure 1 Molecular and genetic features of one Chinese pedigree with type 2 diabetes. A: One family with maternally transmitted diabetes, arrow indicates the proband; B: Sequence analysis of the m.C15954T mutation. WT: Wild type; MT: Mutant.

Diagnosis of T2DM was based on the American Diabetes Association criteria[23], which include any of the following: (1) Fasting plasma glucose ≥ 7.0 mmol/L; (2) 2-hour plasma glucose ≥ 11.1 mmol/L during an oral glucose tolerance test (OGTT); (3) Glycated hemoglobin (HbA1c) level ≥ 6.5%; or (4) Random plasma glucose ≥ 11.1 mmol/L. Control subjects were excluded if they had a family history of cardiovascular events, organ dysfunction, long-term medication use, a history of major surgery or trauma, pregnancy, or any other condition that could interfere with outcome assessment.

Biochemical determinations

Demographic, anthropometric, clinical, and medical history data were collected from matrilineal relatives of the pedigree. Blood pressure (BP) was measured using a mercury sphygmomanometer, with systolic and diastolic values determined by the first and fifth Korotkoff, respectively[24]. Body mass index was calculated as weight (kg) divided by height squared (m²). Venous blood samples were collected after an overnight fast at 7:00. HbA1c was quantified by high-performance liquid chromatography (Bio-Rad, United States). Plasma glucose (0-hour), triacylglycerol, total cholesterol, and serum creatinine were assessed using standard assays (Beckman Coulter, Japan). Fasting insulin was determined via electrochemiluminescence immunoassay (Roche Cobas e601, China). IR was evaluated using the homeostasis model assessment (HOMA-IR), calculated as [fasting plasma glucose (mmol/L) × fasting insulin (mU/L)]/22.5, a HOMA-IR value ≥ 2.69 indicated IR. A 75-g OGTT was performed, with plasma glucose measured at 0 hour and 2 hours. Urinary biomarkers including microalbumin and α1-microglobulin were analyzed on a BN^TM II system (Siemens, Germany). Estimated glomerular filtration rate was derived using the chronic kidney disease epidemiology collaboration equation[25]. Additionally, a questionnaire was administered to record the age at onset of T2DM for each affected individual in the pedigree.

Mutational analysis of mitochondrial genomes

Genomic DNA was extracted from peripheral blood using the QIAmp Blood Kit (QIAGEN, Hilden, Germany). Following established protocols[26], the complete mitochondrial genomes from four matrilineal relatives (II-4, II-6, III-4, III-10) and 255 control subjects were amplified via polymerase chain reaction (PCR) using 24 primer sets. The resulting amplicons were subjected to Sanger sequencing, and the sequences were compared against the revised Cambridge reference sequence (rCRS) (GenBank Accessible No. NC9.012920.1)[27]. Sequence variants and mutations were analyzed using DNA STAR software version 5.01 (United States).

Data analysis

To assess the potential pathogenicity of the identified mtDNA variants, phylogenetic conservation analysis was performed as previously described[28]. The conservation index (CI) was calculated by aligning human mtDNA sequences with those from 16 other species. A CI ≥ 75% was considered indicative of functional significance[29].

mtDNA copy number analysis

mtDNA copy number in peripheral blood was determined by quantitative PCR based on the 2^-∆∆Ct method, following an established protocol[30]. The primers used were as follows: Nuclear β-globin gene: Forward 5’-GAAGAGCCAAGGACAGGTAC-3’ and reverse 5’-CAACTTCATCCACGTTCACC-3’; mt-ND1 gene, forward 5’-AACATACCCATGGCCAACCT-3’ and reverse 5’-AGCGAAGGGTTGTAGTAGCCC-3’.

Generation of cybrid cell models

Lymphoblastoid cell lines were established from four diabetic matrilineal members (II-4, II-6, III-4, III-10) carrying the tRNA^Thr/tRNA^Pro C15954T mutation and four control subjects (C1-C4) without this mutation. Cells were cultured in RPMI 1640 medium (corning) with 10% fetal bovine serum (FBS). The 143B.TK^- cell line was maintained in high-glucose Dulbecco’s modified eagle medium (4.5 mg/mL glucose, 0.11 mg/mL pyruvate) containing 100 μg/mL bromodeoxyuridine and 5% FBS. Its mtDNA-deficient derivative, the ρ⁰ 206 line, was cultured under the same conditions with the addition of 50 μg/mL uridine. Transformation of ρ⁰ 206 cells was achieved by fusion with immortalized lymphoblastoid cells according to the established procedures[31]. This process yielded four control cybrids (C1-C4) and four mutant cybrids (II-4, II-6, III-4, III-10), which were subsequently used for biochemical assays. To verify successful cybrid generation, the presence of the m.C15954T mutation was assessed by PCR-Sanger sequencing using the following primers: Forward 5’-TGAAACTTCGGCTCACTCCT-3’ and reverse 5’-GAGTGGTTAATAGGGTGATAG-3’. Amplified products were purified, sequenced, and compared to the rCRS (GenBank Accessible No. NC012920.1)[27].

Northern blot analysis

To determine whether the m.C15954T mutation affected tRNA metabolism, Northern blot analysis was performed to assess the steady-state levels of tRNA^Thr and tRNA^Pro. Total RNA was extracted from cybrid cells using the totally RNA kit from Ambion (Thermo Fisher, Shanghai, China). Aliquots of 2 μg RNA were separated on a 10% polyacrylamide/7M urea gel and electroblotted onto a positively charged nylon membrane (Roche). The membrane was then hybridized with specific digoxigenin (DIG)-labeled oligodeoxynucleotide probes[32]. The probe sequences were as follows: tRNA^Thr: 5’-TGTCCTTGGAAAAAGGTTTTCATCTCCGG-3’, tRNA^Pro: 5’-CAGAGAAAAAGTCTTTAACTCCACCATTAG-3’; 5S rRNA (loading control): 5’-GGGTGGTATGGCGGTAGA C-3’. Hybridization and band-density quantification were carried out as described previously[32].

Analysis of mt-RNA transcription

The total RNA was isolated from eight cybrids using the TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, United States). Reverse transcription was performed with 5 μg RNA using a commercial kit (Takara, Kusatsu, Shiga, Japan). Quantitative PCR was then conducted with SYBR Green (Bio-Rad, Hercules, CA, United States) according to an established protocol[33].

Adenosine triphosphate analysis

Adenosine triphosphate (ATP) levels in mutant and control cybrids were measured using the CellTiter-Glo^® luminescent cell viability assay (Promega, Madison, WI, United States) in accordance with the manufacturer’s instructions[34].

Reactive oxygen species analysis

Reactive oxygen species (ROS) levels were evaluated by incubating 2 × 10⁶ cells with 2,7-dichlorodihydrofluorescein for 30 minutes. Fluorescence intensity was recorded using a plate reader, as previously reported[35].

Mt membrane potential assessments

Mt membrane potential (MMP) was assessed in both control and mutant cell lines using JC-10 dye (Life Technologies, CA, United States), following a published method[35]. Fluorescence of JC-10 monomers (excitation/emission: 490/529 nm) and J-aggregates (excitation/emission: 490/590 nm) was detected by flow cytometry.

Analysis of oxidative phosphorylation enzymatic activities

The activities of mt respiratory complexes I-IV in cybrid cells were determined according to a standardized procedure[36], and normalized to citrate synthase activity.

Pathogenicity scoring

The potential pathogenicity of the m.C15954T mutation was evaluated using a previously described scoring system[37].

Statistical analysis

All mt functional data are presented as mean ± SD. The homogeneity of variance was first examined using the F-test. Differences between unpaired groups were analyzed with Student’s t-test using SPSS version 22.0, with statistical significance set at P < 0.05.

RESULTS

Clinical and biochemical characteristics of the T2DM pedigree

A Han Chinese family with T2DM was recruited from Hangzhou First People’s Hospital. The proband (III-4), a 44-year-old female from Hangzhou, Zhejiang Province, had been diagnosed with diabetes two years prior to the study and presented for routine T2DM management. Laboratory assessments, including HbA1c and OGTT, confirmed her diabetic status (Table 1). Physical examinations indicated obesity and hypertension (BP: 155/80 mmHg). A review of familial medical history revealed that several matrilineal relatives (II-4, II-6, III-10) also affected by T2DM. Specifically, individual II-4 developed diabetes at age 66, followed by renal complications. Earlier generations (I-2 and II-1) had died years earlier from diabetes-related complications.

Table 1 Summary of clinical and biochemical data of several matrilineal relatives in this pedigree with type 2 diabetes mellitus.

Subjects	Gender	BMI (kg/m2)	Age at onset (year)	Age at test (year)	HbA1c (%)	Glucose (0-hour, mmol/L)	Glucose (2-hour, mmol/L)	Fast insulin (mU/L)	HOMA-IR	TG (mmol/L)	TC (mmol/L)	Serum Cr (μmol/L)	mAlb (mg/L)	α1-mG (mg/L)	eGFR (mL/minute)	BP (mmHg)
II-4	F	23.0	66	72	6.9	7.3	12.6	14.2	4.61	1.22	4.71	122.0	977.2	39.7	69	145/90
II-6	F	22.2	59	63	7.2	7.9	13.5	11.66	4.11	1.09	3.30	139.0	1055.6	47.3	53	155/75
III-4	F	26.2	42	44	6.6	8.8	11.9	7.10	2.79	3.20	5.90	88.0	10.9	9.4	100	155/80
III-10	F	24.3	41	46	6.5	7.2	12.0	6.6	2.11	2.60	2.99	59.0	7.7	8.05	88	140/70
III-9	M	20.5		39	5.2	4.8	7.8	4.2	0.89	1.30	3.48	77.0	6.3	11.1	119	125/75
Normal range		20-22			4.0-6.1	3.9-6.1	3.9-7.8	2.6-12.0	< 2.5	< 1.7	< 5.69	41-81	0-19	0-12.5	80-130	< 140/90

BMI: Body mass index; HbA1c: Glycosylated hemoglobin; HOMA-IR: Homeostatic model assessment for insulin resistance; TG: Triglycerides; TC: Total cholesterol; Cr: Creatinine; mAlb: Microalbumin; α1-mG: Α1-microglobulin; eGFR: Estimated glomerular filtration rate; BP: Blood pressure; F: Female; M: Male.

Analysis of mtDNA mutations/variants

The maternal inheritance pattern suggested an underlying mtDNA mutations involvement. PCR-Sanger sequencing identified 33 variants classified within the East Asian mtDNA haplogroup F1b1a1[38] (Table 2). These included five variants in the D-Loop, three in 12S rRNA, two in 16S rRNA, one affecting both mt-tRNA^Thr and tRNA^Pro, with the remaining located in genes related to oxidative phosphorylation (OXPHOS). Nine missense mutations were detected: ND1 T4216C [tyrosine (Tyr)-histidine], ND2 C5178A [leucine (Leu)-methionine], CO1 C8414T (Leu-phenylalanine), G8584A [alanine (Ala)-threonine (Thr)], A8701G (Thr-Ala), A8860G (Thr-Ala), ND3 A10398G (Thr-Ala), ND6 C14766T (Thr-isoleucine), and CytB A15326G (Thr-Ala). Phylogenetic conservation analysis across mice[39], cattle[40], and Xenopus laevis[41] showed that most variants were poorly conserved, with the exception of m.C15954T (Figure 1B). This mutation was absent in all 255 control samples, indicating a potential role in T2DM pathogenesis.

Table 2 Mitochondrial DNA variants in matrilineal relatives of this type 2 diabetes mellitus pedigree.

Gene	Position	Alterations	Amino acid changes	rCRS	Conservation (H/B/M/X)	Reported1
D-loop	73	A to G		A		Yes
	263	A to G		A		Yes
	310	Ins C		T		Yes
	16132	T to C		T		Yes
	16224	T to C		T		Yes
12S rRNA	750	A to G		A	A/A/A/-	Yes
	1107	T to C		T	T/C/T/T	Yes
	1438	A to G		A	A/A/A/G	Yes
16S rRNA	2706	A to G		A	A/G/A/A	Yes
	3107	del N		N	N/T/T/-	Yes
ND1	4200	A to T		A		Yes
	4216	T to C	Tyr to His	T	Y/Y/H/H	Yes
ND2	5178	C to A	Leu to Met	C	L/T/T/T	Yes
CO1	7028	C to T		C		Yes
A6	8414	C to T	Leu to Phe	C	L/F/M/W	Yes
	8584	G to A	Ala to Thr	G	A/V/V/I	Yes
	8701	A to G	Thr to Ala	A	T/S/L/Q	Yes
	8860	A to G	Thr to Ala	A	T/A/A/T	Yes
CO3	9540	T to C		T		Yes
	9545	A to G		A		Yes
ND3	10397	A to G		A		Yes
	10398	A to G	Thr to Ala	A	T/T/T/A	Yes
	10400	C to T		C		Yes
ND4	11722	G to A		G		Yes
	11917	G to A		G		Yes
ND5	12705	C to T		C		Yes
ND6	14318	T to C		T		Yes
	14766	C to T	Thr to Ile	C	T/S/T/S	Yes
CytB	14927	A to G		A		Yes
	15043	G to A		G		Yes
	15301	G to A		G		Yes
	15326	A to G	Thr to Ala	A	T/M/I/I	Yes
tRNAThr/tRNAPro	15954	C to T		C	C/C/C/C	No

¹Please see online MITOMAP (http://www.MITOMAP.org), mtDB (http://www.genpat.uu.se/mtDB) databases.

rCRS: Reversed Cambridge reference sequences; H: Human; B: Bovine; M: Mouse; X: Xenopus laevis; Tyr: Tyrosine; His: Histidine; Leu: Leucine; Met: Methionine; Phe: Phenylalanine; Ala: Alanine; Thr: Threonine; Ile: Isoleucine.

As illustrated in Figure 2, the m.C15954T mutation is situated adjacent to the 3’-end spacer of mt-tRNA^Pro within a cytosine-rich light-strand transcript, and near mt-tRNA^Thr in a guanine-rich heavy-strand transcript[42]. Maturation of mt tRNAs from primary transcripts involves precise endonucleolytic cleavage: RNase P (comprising MRPP1, MRPP2, and MRPP3 subunits) processes the 5’ end, while RNase Z (encoded by ELAC2) cleaves the 3’ end[43,44]. Therefore, the m.C15954T mutation is predicted to disrupt 3’ end processing of both mt-tRNA^Thr and mt-tRNA^Pro precursors a mechanism analogous to that reported for the T2DM-associated mt-tRNA^Cys/tRNA^Tyr A5826G mutation[45].

Figure 2 Secondary structures of mitochondrial-tRNA^Thr and mitochondrial-tRNA^Pro genes, arrow indicates the position of 15954. Processing sites in mitochondrial (mt)-tRNA^Thr and mt-tRNA^Pro precursors were determined for tRNase Z.

The m.C15954T mutation affected tRNA metabolism

To assess the impact of the m.C15954T mutation on the stability of tRNA^Thr and tRNA^Pro, total mt-RNA from control and mutant cybrids was analyzed by Northern blot under denaturing conditions using DIG-labeled oligodeoxynucleotide probes specific for tRNA^Thr, tRNA^Pro, and 5S rRNA (loading control). As shown in Figure 3A, the steady-state levels of both tRNA^Thr and tRNA^Pro were markedly lower in mutant cybrids than in controls. Quantitative analysis revealed that the average levels of tRNA^Thr and tRNA^Pro in mutant cybrids were 53.2% and 45.7%, respectively, of those in control cybrids (P < 0.0001 for both; Figure 3B).

Figure 3 Mitochondrial-transfer RNAs analyses. A: Analysis of the steady-state levels of mitochondrial (mt)-tRNA^Thr and mt-tRNA^Pro in control and mutant cell lines by Northern blot; B: Qualifications of mt-transfer RNA levels, the average relative transfer RNAs content were normalized to the average content per cell of 5S rRNA. tRNA: Transfer RNA.

Mt-RNA transcription was impaired by the m.C15954T mutation

We further examined mt-RNA transcription in cell lines with and without the m.C15954T mutation. Compared with controls, mutant cell lines showed significantly reduced expression of multiple mt transcripts, including mt-ND1 (P = 0.00258), mt-ND2 (P = 0.005), mt-ND3 (P = 0.0096), mt-ND5 (P = 0.0015), mt-CO3 (P = 0.0263), and mt-ATP6 (P = 0.0164) (Figure 4). These results suggest that the mutation partially disrupts mt-RNA transcription.

Figure 4 Analysis of mitochondrial-RNA transcription in control and mutant cell lines.

Mitochondrial functions were impaired in 15954T cybrids

To evaluate the effect of the m.C15954T mutation on mitochondrial functions, we compared mtDNA copy number, ATP levels, MMP, and ROS production between control and mutant cybrids. As illustrated in Figure 5, the m.C15954T mutation was associated with significant reductions in mtDNA copy number (approximately 31.5%), ATP content (approximately 22.6%), and MMP (approximately 32.4%; P < 0.05). Conversely, ROS levels were increased by approximately 42.7% (P < 0.001) relative to controls. Additionally, activities of respiratory chain complexes I and IV were substantially lower in 15954T cybrids than in controls (P < 0.001, Figure 6). Together, these data indicate that the m.C15954T mutation leads to impaired mt function.

Figure 5 Analysis of mitochondrial DNA copy number, adenosine triphosphate, mitochondrial membrane potential and reactive oxygen species levels in 15954C and 15954T cybrids. A: Mitochondrial DNA copy number; B: Adenosine triphosphate; C: Mitochondrial membrane potential; D: Reactive oxygen species. mtDNA: Mitochondrial DNA; ATP: Adenosine triphosphate; MMP: Mitochondrial membrane potential; ROS: Reactive oxygen species.

Figure 6 Analysis of oxidative phosphorylation enzymatic activities in 15954C and 15954T cybrids. CI-CIV: Complex I-IV; C: Control; M: Mutant.

The m.C15954T was “definitely pathogenic” for T2DM

Using the revised pathogenicity scoring system[37], we assigned a total score of 11 points to the m.C15954T mutation, classifying it as “definitely pathogenic” for T2DM (Table 3).

Table 3 The pathogenic role of the m.C15954T mutation.

Scoring criteria	m.C15954T mutation	Score/20	Classification
More than one independent report	No	0	≤ 6 points: Neutral polymorphisms; 7-10 points: Possibly pathogenic; 11-13 points (not including evidence from single fiber, steady-state level, or trans-mitochondrial cybrid studies): Probably pathogenic; ≥ 11 points (including evidence from single fiber, steady-state level or trans-mitochondrial cybrid studies): Definitely pathogenic
Evolutionary conservation of the base pair	No changes	2
Variant heteroplasmy	No	0
Segregation of the mutation with disease	Yes	2
Biochemical defect in complex I, III or IV	Yes	2
Evidence of mutation segregation with biochemical defect from single fiber studies	No evidence	0
Mutant mt-tRNA steady-state level or evidence of pathogenicity in trans-mitochondrial cybrid studies	Strong evidence	5
Maximum score	Definitely pathogenic	11

mt: Mitochondrial; tRNA: Transfer RNA.

DISCUSSION

The principal findings of this study are as follows: (1) The m.C15954T mutation occurs at a highly conserved nucleotide within the intergenic spacer region between mt-tRNA^Thr and mt-tRNA^Pro, which may impair the 3’ end processing of both tRNAs; (2) The mutation reduces the steady-state levels of tRNA^Thr and tRNA^Pro, induced mt dysfunction, and compromises OXPHOS activity; and (3) Incomplete penetrance and variable clinical phenotypes among matrilineal relatives suggest that the m.C15954T mutation alone is insufficient to cause overt T2DM, implying additional modifying factors such as epigenetic alterations or nuclear genetic background.

In the present study, we performed clinical, genetic, and biochemical analyses of a four-generation Han Chinese pedigree with maternally inherited T2DM. The exclusive occurrence of the disease among matrilineal relatives strongly suggested a mt etiology. Clinical evaluation revealed considerable variability in disease severity and age of onset among the four affected individuals, who developed diabetes at ages 66, 59, 42, and 41 (mean 52 years). Sequencing of the complete mt genome identified 33 variants belonging to the Eastern Asian haplogroup F1b1a1[38]. Among these, 32 were considered polymorphisms due to low evolutionary conservation and lack of established functional relevance. Notably, a homoplasmic C-to-T transition at position 15954, located in the spacer region immediately adjacent to the 3’ ends of mt-tRNA^Thr and mt-tRNA^Pro, was of particular interest. This nucleotide is highly conserved across primates and was absent in 255 Han Chinese controls, supporting its potential role in T2DM pathogenesis.

In human mitochondria, the 22 mt-tRNAs, together with 13 messenger RNAs and 2 rRNAs, are transcribed as polycistronic precursors from heavy and light strands[46]. Although mt-tRNA^Thr and mt-tRNA^Pro are encoded on opposite strands, both reside within long precursor transcripts that require extensive processing to become functional[47]. Key maturation steps include endonucleolytic cleavage, 5’ and 3’ trimming, splicing, base modification, and CCA addition, the latter being essential for tRNA stability and aminoacylation[48,49]. We propose that the m.C15954T mutation disrupts 3’ end processing of both mt-tRNA^Thr (on the heavy strand) and mt-tRNA^Pro (on the light strand). This defect is consistent with previous reports linking impaired 3’ end maturation to human disease, such as the cardiomyopathy-associated A12265G mutation in mt-tRNA^Ser(AGY)[50] and the m.A4295G mutation in mt-tRNA^Ile[51].

Using cybrid cell models, we observed that this mutation reduced the steady-state levels of tRNA^Thr and tRNA^Pro by approximately 46.8% and 54.3%, respectively, compared to control cells. This level of reduction falls below the threshold typically required to produce a clinical phenotype, a pattern similar to that reported for the tRNA^Lys A8344G mutation[52]. The resulting disturbance in mt-tRNA metabolism subsequently impaired mt-RNA transcription and compromised OXPHOS enzymatic activities. Furthermore, cell lines carrying the m.C15954T mutation exhibited decreased mtDNA copy number, diminished ATP production, reduced MMP, impaired OXPHOS enzyme activity, and a marked increase in ROS levels. Collectively, these findings indicate that the 15954T mutation induces mt dysfunction, which underlies the observed clinical manifestations.

MtDNA haplogroups have been proposed to modulate T2DM susceptibility and expression across different ethnic populations. For example, multiple experimental studies indicated that European haplogroups J/T or T may increase diabetes risk[53]. Achilli et al[54] also reported that haplogroups H, H3, U3, and V are significantly linked to a higher incidence of diabetic complications. Notably, a recent study demonstrated that mtDNA haplogroup N9a elevates T2DM susceptibility in Chinese populations by affecting mt function and intracellular signaling[55]. Complete mt genome sequencing of the affected matrilineal relatives in this pedigree identified 33 variants belonging to the haplogroup F1b1a1[38]. To further explore the potential influence of mtDNA haplogroups on T2DM penetrance and expressivity, we compared our data with eight additional reported T2DM pedigrees (Table 4). Several primary mt-tRNA mutations, such as haplogroup G4-specific mt-tRNA^Glu A14687G[56], haplogroup B5-specific mt-tRNA^Glu A14692G[57], haplogroup D4-specific mt-tRNA^Cys/tRNA^Tyr A5826G[45], haplogroup F2-specific mt-tRNA^Ala T5587C[17], haplogroup G2a-specific mt-tRNA^Trp A5514G[58], haplogroup D4b1-specific mt-tRNA^Thr G15897A[59], and haplogroup M11b-specific mt-tRNA^Gly T10003C[60] have been directly linked to diabetes onset and appear to enhance the penetrance and clinical severity in these families.

Table 4 Summary of clinical and molecular data for several Chinese diabetic pedigrees carrying the primary mitochondrial-transfer RNA mutations.

Pedigree number	Number of matrilineal relatives	Penetrance of diabetes (%)	mt-tRNA mutations	mtDNA haplogroup	Ref.
1	12	50	tRNAThr/tRNAPro C14954T	F1b1a1	This study
2	11	45.4	tRNAGlu A14687G	G4	Levinger et al[51]
3	11	18.2	tRNAGlu A14692G	B5	Moslemi et al[52]
4	12	41.6	tRNACys/tRNATyr A5826G	D4	Roe et al[41]
5	12	25	tRNALeu(UUR) A3243G	M7c	Jiang et al[17]
6	8	50	tRNAAla T5587C	F2	Brandon et al[15]
7	19	31.5	tRNATrp A5514G	G2a	Crispim et al[53]
8	19	26.3	tRNAThr G15897A	D4b1	Achilli et al[54]
9	11	45.5	tRNAGly T10003C	M11b	Fang et al[55]

mt: Mitochondrial; tRNA: Transfer RNA.

Based on our findings, we propose the following molecular mechanisms through which the m.C15954T mutation contributes to T2DM: First, the mutation disrupts the 3’ end processing, CCA addition, and steady-state levels of mt-tRNA^Thr and mt-tRNA^Pro. This impairment in mt-tRNA metabolism compromises mt protein synthesis and respiratory chain function. Second, the resulting respiratory dysfunction elevates oxidative stress and uncouples OXPHOS, leading to reduced ATP production[61]. These abnormalities may promote pancreatic beta-cell dysfunction and apoptosis, thereby diminishing insulin secretion[62]. Ultimately, impaired insulin action fails to suppress hepatic glucose output or facilitate peripheral glucose uptake, resulting in IR and the development of T2DM.

CONCLUSION

Our finding demonstrated that the novel m.C15954T mutation altered the structural and functions of tRNA^Thr and tRNA^Pro, leading to mt dysfunction that contributed to the pathogenesis of T2DM. A key limitation of this study is its small sample size. Future research involving larger cohorts, as well as analyzing the effects of mtDNA haplogroup on T2DM progression are essential to validate these observations.

ACKNOWLEDGEMENTS

We thanked the members of our laboratory for discussion; we are also grateful to Dr. Jiang ZC from the Second Affiliated Hospital of Zhejiang University for critical reading of this manuscript.