Two patients diagnosed with PCD harbor bi-allelic mutations in the DNAH10 gene

A girl, 1 year and 3 months old, was born at full term via natural childbirth. She experienced respiratory distress after birth and underwent non-invasive CPAP ventilation for one week. Chest and abdominal radiographs revealed dextrocardia, lever on the left, stomach on the right. And high-resolution CT scans indicated sinusitis and patchy pulmonary infiltration, along with dextrocardia (Fig. 1A). The exhaled nitric oxide concentration through the nose was significantly reduced (patient: 13.2 nl/min, normal control of the same age: 113.4 nl/min). High-speed photography revealed that the ciliary beating in the nasal mucosa of the girl was slower compared to that of a normal individual (Supplementary Movie 1, 2). A 39-year-old male patient presented with recurrent respiratory infections, wet cough, and sought medical assistance due to infertility concerns. CT imaging revealed the presence of cystic shadows in both maxillary sinuses, while fibrosis with striated shadows and numerous small nodular shadows were observed in both lungs (Fig. 1B).

Fig. 1

The two individuals carrying DNAH10 mutations exhibited characteristic PCD lung pathophysiology. (A) Left: a nasal sinuses CT scan showing pansinusitis in the patient. Middle: a chest CT scan image revealing bronchiectasis in the patient. Right: Frontal radiography demonstrated dextrocardia (heart apex positioned on the right) in a one-year-three-month-old girl. (B) Left: CT scan analysis of the nasal sinuses revealed pansinusitis in the patient. Right: A chest CT scan image demonstrated the presence of small nodules. (C) The locations of four variants in DNAH10, identified in two PCD cases, were presented. The domains/motifs within DNAH10 are color-coded according to the NCBI browser. (D) Pedigrees of two families harboring DNAH10 variants (P01, P02) identified through whole-exome sequencing (WES) were presented. Black-filled symbols represent two individuals with PCD within these families. Double lines signify first-degree consanguinity. (E) Sanger sequencing confirmed the existence of bi-allelic DNAH10 variants in individuals II-1 of Family P01 and Family P02. The variant positions are highlighted with red boxes

WES was performed to seek the etiology of the two individuals. Whole-exome sequencing detected DNAH10 mutations in the girl, P01, specifically NM_207437: c.9263 A > G, p.E3088G, and c.6544 A > T, p.K2128* (Supplementary Table S3). No other pathogenic or suspected pathogenic variants that could explain the patient’s phenotype were detected. Analysis of the mutation sources from the parents revealed a compound heterozygous mutation, consistent with an autosomal recessive inheritance pattern. The c.6544 A > T mutation was derived from the mother and was a nonsense mutation, leading to premature termination of translation after amino acid 2182, resulting in a truncated polypeptide. The c.9263 A > G mutation, inherited from the father, was a missense mutation causing a change from glutamic acid to glycine at position 3088 (Fig. 1C-E). The amino acid at position 3088 is highly conserved among most species and is located within the microtubule-binding domain (MTBD). Structurally, it is difficult to directly predict the impact of this mutation, but it may affect the binding ability of DNAH10 (Fig. 1C, Supplementary Fig. 1A, B). The K2128* mutation led to the deletion of most of the DNAH10 motor domain, which may affect the assembly of the entire IDAf complex, and other proteins that directly interact with the DNAH10 motor domain (Fig. 1C, Supplementary Fig. 1A, B).

The male individual, P02, carried compound heterozygous missense mutations DNAH10, including c.8378G > A (p.R2793H) and c.9494 C > G (p.T3165R). The sequencing results from his parents demonstrated that the c.8378G > A and c.9494 C > G mutations were inherited from his father and mother, respectively (Fig. 1F, G). The amino acid at position 2793 and 3165 are highly conserved among most species and are located within the Hydrolytic ATP binding site of dynein motor region D4 and MTBD, respectively (Fig. 1C, Supplementary Fig. 1A).

All the identified DNAH10 variants were either rare or absent in public human databases, including the 1000 Genomes Project and gnomAD (Supplementary Table S3). Additionally, these missense variants in DNAH10 were predicted to be damaging using tools such as SIFT, MutationTaster, Proven prediction (Supplementary Table S3).

Impact of biallelic variants on DNAH10 expression

To assess the in vivo effects of these variants on DNAH10, we initially employed RT-PCR to examine the transcriptional expression levels of DNAH10 in ciliated epithelial cells of nasal mucosa. The results indicated a significant decrease in DNAH10 expression in the patient P01 (Fig. 2A). Western blot analysis demonstrated nearly absent DNAH10 protein expression in ciliated epithelial cells of patient P01, in contrast to those without PCD (Fig. 2B). Subsequently, we conducted immunofluorescence analysis to investigate alterations in DNAH10 expression and localization. Notably, in contrast to the nasal mucosa cilia of healthy individuals, where distinct DNAH10 staining was observed along the ciliary axoneme, no DNAH10 staining was detected in the nasal mucosa cilia of the two subjects (Fig. 2C). This finding suggested that the biallelic variants may lead to a decline in DNAH10 expression.

Fig. 2

Bi-allelic DNAH10 variants induced declined expression of DNAH10. (A) RT-PCR was employed to evaluate the expression of DNAH10 at the RNA level in nasal mucosal tissue collected from P01. *P < 0.05. n = 3 per group (B) Western blot analysis was conducted to assess the expression of DNAH10 in nasal mucosal tissue. (C) Representative images were obtained from the nasal mucosa of control individuals (NC) and individuals harboring bi-allelic DNAH10 variants. These images were stained with the anti-DNAH10 antibody, anti-α-tubulin antibody, and DAPI. Scale bars = 5 mm. Red, Ace-tubulin; Green, DNAH10, respectively

Dnah10 KO mice display typical PCD phenotypes consistent with IDA defects

To investigate the impact of DNAH10 deletion on cilium function in vivo, we first developed a CRISPR-Cas9-based Dnah10 KO mouse model (Supplementary Figure S2A). The genotypes of the mice were verified through Sanger sequencing and PCR analysis of genomic tail DNA (Supplementary Figure S2B, C). Western blot analysis further revealed that DNAH10 protein was largely undetectable in KO mice (Supplementary Figure S2D). Additionally, RT-PCR analysis demonstrated the absence of the Dnah10 transcript in the lung tissues of KO mice (Supplementary Figure S2E). Furthermore, immunofluorescence staining revealed a significant reduction in ciliary DNAH10 expression in ciliated cells of Dnah10 KO mice lung tissues (Supplementary Figure S3).

Mouse pulmonary function was determined by following parameters, including inspiratory-to-expiratory time (Ti/Te), PEF, Mv, Rpef, Volbal, PenH, and EF50. The results showed that the above parameters were significantly reduced in the Dnah10 KO mice, indicating increasing airway obstruction and worsening bronchoconstriction (Fig. 3A, B). Hematoxylin and eosin (H&E) staining demonstrated chronic lung infection, which induced inflammatory cell infiltrates and pulmonary interstitial hyperplasia (Fig. 3C). An increased number of neutrophils and mononuclear cells was observed, including lymphocytes and macrophages, in the lung of Dnah10 KO mice (Fig. 3D). Western blot analysis demonstrated significantly elevated levels of inflammatory cytokine IFN-β and complement regulatory molecule CD46, accompanied by increased expression of immune cell-specific markers, including CD68 (monocyte), CD45 (T cell), CD63 (neutrophil), and CD86 (dendritic cell) (Fig. 3E). The fluorescence intensity of these inflammatory cell markers was significantly intensified in the lungs of Dnah10 KO mice, showing severe respiratory tract infections in these animals (Fig. 3F).

Fig. 3

Dnah10 KO mice represented compromised lung function and accumulated inflammation. (A) Breathing curve illustrating lung function parameters in WT and Dnah10 KO mice. (B) Lung function parameters indicate airway obstruction and chronic bronchitis in Dnah10 KO mice. PEF: peak expiratory flow rate (mL/s); Mv: minute ventilation (mL); Rpef: ratio of time to PEF to Rt, reflecting respiratory muscle strength and small airway obstruction; Rt: time to expire 65% of tidal volume; Volbal: respiration ratio; Penh: measure of bronchoconstriction; EF50: expiratory flow at 50% exhalation (mL/s), indicating obstructive ventilatory dysfunction. *P < 0.05; n = 6/group. (C) H&E staining of lung sections from 40-day-old WT and Dnah10 KO mice revealed increased infiltration of inflammatory cells in Dnah10 KO mice compared to WT mice. Mononuclear cells, including lymphocytes and macrophages, as well as neutrophils, were significantly elevated, as illustrated in the scatter plot. The black dotted boxes highlight the magnified areas of the lung sections. Lymphocytes, neutrophils, and macrophages were indicated by green, black, and blue arrows, respectively. Scale bar = 20 μm. (D) Cell counting for neutrophils and mononuclear cells from left HE staining. Equal-sized 5% areas of the sections were selected. The statistical analysis was presented in the accompanying panel. *P < 0.05; n = 6/group. (E) Western blot analysis was conducted to evaluate the expression levels of inflammatory factors, including IFN-β, CD68, CD45, CD63, and CD86, in Dnah10 KO and WT mice. (F) Immunostaining of immunocytes, including CD45 (leucocyte marker), CD68 (macrophage marker), CD86 (marker for monocytes, T, and B lymphocytes), and CD63 (marker for activated basophils), revealed enhanced positive signals in the lung sections of the Dnah10 KO mouse model. Green, CD45; yellow, CD68; red, CD63; purple, CD86; blue, DAPI. Scale bar = 50 μm

Notably, the Dnah10 KO mice exhibited heterotaxy syndrome. The heart was located on the right side of Dnah10 KO mice, whereas the other organs appeared to be appropriately distributed (Fig. 4A). This suggested that the defect in DNAH10 disrupts the node cilia. To investigate the phenotypic alterations in the airway motile cilia, we conducted H&E staining. Our findings revealed severe vacuolation in the multiciliated airway epithelial structure of Dnah10 KO mice, accompanied by a more crooked appearance of the cilia (Fig. 4B). SEM was employed to visualize the morphology of the cilia. In contrast to the straight and uniformly directed cilia protruding from the cell plasma membrane in WT mice, the cilia in Dnah10 KO mice displayed curvature, lacked directional orientation, and aberrantly protruded outside the cells and into the disrupted respiratory epithelium (Fig. 4C). Furthermore, the distal regions of the motile cilia in Dnah10 KO mice displayed marked curvature (Fig. 4C). Additionally, TEM analysis revealed the absence of the IDA ultrastructure and disorganization of the outer doublet microtubules in the cilia region of the KO mice (Fig. 4D). Furthermore, high-speed video microscopy demonstrated a significant reduction in ciliary beat frequency (CBF) in the KO mice compared to the WT mice, suggestive of a phenotypic recapitulation of patients with PCD, thus confirming the impairment of ciliary motility due to DNAH10 deletion (Fig. 4E, F; Supplementary movie 1–2).

Fig. 4

Dnah10 KO mice showed typical PCD phototype. (A) Heterotaxy of the heart and intestines was observed in Dnah10 KO mice. (B) H&E staining of bronchi in lung sections from 40-day-old WT and Dnah10 KO mice revealed that, compared to WT mice, the tracheal epithelia of Dnah10 KO mice exhibited a more clustered distribution of cilia. Scale bar = 10 μm. (C) The SEM images depicted the surface of tracheal epithelia from both WT and Dnah10 KO mice. Scale bar = 9 μm. (D) TEM images of tracheal epithelium cilia were obtained from both WT and Dnah10 KO mice. Compared to WT mice showing normal “9 + 2” axonemal structure, and intact IDAs (Green arrows), KO mice exhibited disruption of some ciliary axoneme structures, showing single microtubules, loss of a set of doublet microtubules (red dashed box), and absence of IDAs (the red arrow). Scale bar = 200 nm. (E) A representative kymograph of nasal cilia from 40-day-old WT and Dnah10 KO mice was obtained through high-speed video microscopy. Scale bar = 0.005s. (F) The ciliary beat frequency (CBF) of Dnah10 KO mice was significantly reduced compared to that of WT mice (*P < 0.05). Each group consisted of six mice

DNAH10 defect impaired IDAf complex

To investigate the role of DNAH10 in cilia motility, we utilized ciliary structure data obtained through cryogenic electron microscopy (cryo-EM) to gain structural insights into the IDAf complex. The IDAf complex comprises intermediate chains (DNAI3, DNAI4, DNAI7), light chains (DYNLRB1, DYNRB2, DYNLL1), and heavy chains (DNAH2, DNAH10, DNAH12). The spatial organization of the IDAf complex involves the assembly of DNAH2, DNAI3, DNAI4, DNAI7, DYNLRB1, DYNLRB2, DYNLL1, DYNLT2B, DYNLT1, CFAP43, CFAP44, CFAP57, CFAP73, CFAP100, and DNAH12 (Fig. 5A). To validate the structural information, we selected three proteins (CCDC73, CFAP57, DYNLL1) and conducted co-immunoprecipitation experiments using lung tissues. The results confirmed the interactions between DNAH10 and CCDC73, CFAP57, and DYNLL1, respectively (Fig. 5B). Western blot analysis revealed decreased expression levels of these proteins (Fig. 5C). Immunofluorescence staining revealed localization of CCDC73 and DYNLL1 in both cilia and cytoplasm, whereas CFAP57 was exclusively localized to the cilia. The expression levels of these proteins were found to be reduced in the cilia of knockout mice (Fig. 5D-F)and patient (Fig. 5G). To delve deeper into the potential impact of DNAH10 absence on proteins neighboring the IDAf, immunostaining was conducted to meticulously assess markers associated with the ODA (DNAH5, DNAH9), IDA light chain (DNAI1), dynein arm assembly (LRRC6), inner dynein arm assembly (CCDC39), outer dynein arm assembly (CCDC114), and the axoneme central apparatus protein SPAG6 in multiciliated cells derived from KO mice nasal tissue. Notably, our immunostaining results revealed a significant reduction in the expression of DNAH10, DNAH5, DNAI1, and CCDC39 in cilia upon deletion of DNAH10. In contrast, DNAH9, CCDC114, LRRC6, and SPAG6 were not significantly affected (Fig. 6). This finding suggested a structural alteration in the IDA-associated axonemal structure of Dnah10 KO mice.

Fig. 5

Cryo-electron microscopy identified interactors of DNAH10 in ciliated cells from Dnah10 KO mice. (A) DNAH10 interacts with DNAH2 (Uniprot ID: Q9P225), DNAI3 (Q8IWG1), DNAI4 (Q5VTH9), DNAI7 (Q6TDU7), DYNLRB1 (Q9NP97), DYNLRB2 (Q8TF09), DYNLL1 (P63167), DYNLT2B (Q8WW35), DYNLT1 (P63172), CFAP43 (Q8NDM7), CFAP44 (Q96MT7), and DNAH12 (E9PG32) to form the IDAf complex using cryo-electron microscopy. (B) Co-IP analysis confirmed the interaction between DNAH10 and CCDC73/CFAP57/DYNLL1 in mouse lung tissues. (C) Western blot analysis was performed to assess the expression of CCDC73/CFAP57/DYNLL1, which was diminished in Dnah10 KO mice. D-F. Immunofluorescence analysis revealed the expression of CCDC73/CFAP57/DYNLL1 was reduced in ciliated cells from the nasal mucous membrane of patient P02. Scale bars = 5 mm. Red, Ace-tubulin; Green, CCDC73, CFAP57, DYNLL1, respectively; blue, DAPI. G. Immunostaining analysis was performed to investigate the expression of CCDC73, CFAP57, and DYNLL1 in ciliated cells derived from Dnah10 KO nasal epithelium. Red, ace-tubulin; green, CCDC73, CFAP57, and DYNLL1; blue, DAPI; Scale bar = 10 μm

Fig. 6

Immunostaining analysis was performed to investigate the expression of DNAH10, DNAH9, DNAH5, DNAI1, CCDC114, CCDC39, LRRC6, and SPAG6 in ciliated cells derived from Dnah10 KO nasal epithelium. Green, ace-tubulin; red, DNAH10, DNAH9, DNAH5, DNAI1, CCDC114, CCDC39, LRRC6, and SPAG6; blue, DAPI; Scale bar = 10 μm

Danh10 KO mice presented with mitochondrial metabolic rewiring and pulmonary fibrosis

To further investigate PCD pathophysiology caused by deletion of DNAH10, we employed a proteomics approach for WT mice and Dnah10 KO. A total of 4570 proteins were quantified, revealing 349 upregulated proteins and 90 downregulated proteins (Fig. 7A, B, Supplementary Fig. 4). Heat map donut charts provided detailed visualizations of the upregulated and downregulated proteins influenced by the deletion of Dnah10 (Fig. 7B). GO enrichment analysis demonstrated significant alterations in gene categories in response to the deletion of Dnah10. The upregulated GO terms suggested a significant enrichment of collagen-containing extracellular matrix, super-molecular fiber organization, and innate immune response pathway in Dnah10 KO mice (Fig. 7C). The downregulated proteins were associated with metabolic processes, cellular component assembly, and macromolecule localization (Fig. 7D). We observed that some of the downregulated proteins belonged to mitochondria (Fig. 7E). Furthermore, a network diagram was utilized to analyze the interactions of differential proteins, known to play crucial roles in fibrosis in humans and/or mice, between Dnah10 KO mice and WT mice. The results revealed complex interactions among proteins involved in these processes (Fig. 7F).

Fig. 7

LC-MS/MS proteomics identified upregulated and downregulated proteins in lung tissues from Dnah10KO mice. (A) Volcano plot depicted protein alterations in lung tissues of WT and Dnah10 KO mice. (B) The circular heatmap illustrates differential gene expression patterns between WT and Dnah10 KO mice. C, D. GO pathway analysis revealed enrichment of 147 significantly downregulated and 916 upregulated proteins, based on functional annotation from OmicShare. The upregulated proteins were primarily associated with collagen-rich extracellular matrix, supramolecular fiber organization, and immune response, whereas the downregulated proteins were linked to metabolic processes. E. The Gene Ontology (GO) Cellular Component Ontology described many decreased proteins belong to mitochondrial components. F. The correlation between down-regulated proteins from GO pathway analysis was analyzed

Subsequently, we investigated the pulmonary fibrosis phenotype in Dnah10 KO mice. ECM remodeling serves as a pivotal process in pathologies, including fibrosis [18, 19]. Fibroblasts represent the predominant cell type in connective tissues throughout the body and are the primary source of ECM components [20]. Fibroblasts primarily function in maintaining and synthesizing new fibrillar collagens, thereby preserving tissue homeostasis. Upon activation, fibroblasts undergo transdifferentiation into myofibroblasts, triggered by chemical signals promoting either proliferation or cellular differentiation. This process leads to excessive collagen deposition and tissue remodeling. Consequently, myofibroblasts are implicated in the increased stiffness of the ECM, a characteristic observed in fibroproliferative diseases. Myofibroblasts uniformly express α-smooth muscle actin (α-SMA), a 42 kDa actin isoform typically found in stem and precursor cells [21]. α-SMA serves as a well-established biomarker for the assessment of activated fibroblasts across various tissues and organs, including the lung. Immunofluorescence and Western blot revealed a significant increase in α-SMA staining in Dnah10 KO mice (Fig. 8A-C). Additionally, Sirius red staining demonstrated a notable elevation in both type I and type II collagen fibers in Dnah10 KO mice (Fig. 8D, E). Western blot also demonstrated elevated expression of collagen I protein (Fig. 8F). Collectively, our findings demonstrated that DNAH10 deletion led to pulmonary fibrosis in Dnah10 KO mice.

Fig. 8

Dnah10 KO mice exhibited pulmonary fibrosis phenotype. (A) Immunofluorescence revealed a significant increase in α-SMA staining in Dnah10 KO mice. α-smooth muscle actin (α-SMA) is a biomarker for the assessment of activated fibroblasts. Scale bar = 50 μm. Green, α-SMA; Blue, DAPI. (B) Results of immunofluorescence images quantification corresponding to α-SMA in (A). (C) Western blot analysis confirmed the increased expression of α-SMA. D, E. Sirius red staining demonstrated a notable elevation in both type I and type II collagen fibers in Dnah10 KO mice. Scale bar = 50 μm. F. Western blot analysis confirmed the increased expression of collagen I