Visceral Myopathy: Clinical and Molecular Survey of a Cohort of Seven New Patients and State of the Art of Overlapping Phenotypes

Abstract

Visceral motility dysfunction is a key feature of genetic disorders such as megacystis-microcolon-intestinal hypoperistalsis syndrome (MMIHS, MIM moved from 249210 to 155310), chronic intestinal pseudo-obstruction (CIPO, MIM609629), and multisystemic smooth muscle dysfunction syndrome (MSMDS, MIM613834). The genetic bases of these conditions recently begun to be clarified with the identification of pathogenic variants in ACTG2, ACTA2, and MYH11 in individuals with visceral motility dysfunction. The MMIHS was associated with the heterozygous variant in ACTG2 and homozygous variant in MYH11, while the heterozygous variant in ACTA2 was observed in patients with MSMDS. In this study, we describe the clinical data as well as the molecular investigation of seven individuals with visceral myopathy phenotypes. Five patients presented with MMIHS, including two siblings from consanguineous parents, one had CIPO, and the other had MSMDS. In three individuals with MMIHS and in one with CIPO we identified heterozygous variant in ACTG2, one being a novel variant (c.584C>T—p.Thr195Ile). In the individual with MSMDS we identified a heterozygous variant in ACTA2. We performed the whole-exome sequencing in one sibling with MMIHS and her parents; however, the pathogenic variant responsible for her phenotype could not be identified. These results reinforce the clinical and genetic heterogeneity of the visceral myopathies. Although many cases of MMIHS are associated with ACTG2 variants, we suggest that other genes, besides MYH11, could cause the MMIHS with autosomal recessive pattern.

Introduction

Visceral motility dysfunction is characterized by symptoms related to dysmotility of gastrointestinal and urinary tract, a key feature of genetic disorders such as megacystis-microcolon-intestinal hypoperistalsis syndrome (MMIHS, MIM moved from 249210 to MIM155310); chronic intestinal pseudo-obstruction (CIPO, MIM15531); and multisystemic smooth muscle dysfunction syndrome (MSMDS, MIM613834). There is a significant overlap among these phenotypes and the recent advances regarding the genetic bases of these conditions contributed to improve the knowledge.

MMIHS is a severe condition characterized by functional obstruction in the urinary and gastrointestinal tract resulting in marked dilatation of the bladder, underdevelopment of the colon, and decreased or absent intestinal peristalsis [Gosemann and Puri, 2011]. More than 200 described cases are sporadic, and the recent identification of de novo heterozygous variant in ACTG2 confirmed the autosomal dominant (AD) pattern in many, but not in all [Thorson et al., 2014; Wangler et al., 2014]. The occurrence of MMIHS in the offspring of consanguineous parents and/or recurrence in siblings with asymptomatic parents lent support to the hypothesis of autosomal recessive (AR) inheritance [Mc Laughlin and Puri, 2013]. In addition, more recently the AR pattern was demonstrated by Gauthier et al. [2014], who reported a homozygous loss-of-function variant in MYH11 in an individual with MMIHS born to consanguineous parents.

ACTG2 is associated with a less severe phenotype of visceral myopathy named chronic intestinal pseudo-obstruction (CIPO) in familial cases with AD inheritance [Lehtonen et al., 2012; Holla et al., 2014; Klar et al., 2015]. Currently, MMIHS and CIPO are understood as part of a phenotypic spectrum related to pathogenic variants in ACTG2 [Wangler et al., 2014]. The MSMDS is a severe condition with significant overlap with MMIHS, but with congenital mydriasis and vascular abnormalities. This phenotype has been associated with de novo heterozygous variant in the codon 179 in ACTA2 [Milewicz et al., 2010].

Here, we report the clinical and molecular investigation of seven individuals with visceral smooth muscle diseases showing pathogenic variants in ACTG2 (four individuals) and ACTA2 (one individual) in sporadic cases, and two siblings with MMIHS without pathogenic variant identified. In addition, we present the state of the art of molecular bases of visceral myopathy based on a review of the related disorders.

Materials and Methods

The present study was approved by National Committee of Ethics in Research (CONEP). For all the participants and families the informed consent was duly obtained.

Patients

Seven individuals with visceral myopathy were included: Five individuals presenting typical findings of MMIHS (patients 1, 2, 3, 4, and 5), a sixth individual with CIPO phenotype (patient 6), and one with clinical features suggestive of MSMDS (patient 7)— Table I. The parents were included in the study when their DNA was available. Clinical data, including family history, and clinical follow up, were obtained from all patients.

Table I. Clinical and Molecular Findings of the Seven Evaluated Patients.

	Family 1		Family 2	Family 3	Family 4	Family 5	Family 6
	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5	Patient 6	Patient 7
Phenotype	MMIHS	MMIHS	MMIHS	MMIHS	MMIHS	CIPO	MSMDS
Parental	+	+	−	−	−	−	−
consanguinity
Familial	+ (Patient 2)	+ (Patient 1)	−	−	−	−	−
recurrence
Pathogenic
variant
ACTG2*	NE	Not identified	C.532c>T (p.Argl78Cys) Het—de novo	c770G>A (p.Arg257His) Het—de novo	.533G>T (p.Arg178Leu) Het—de novo	c.584C>T (p.Thr195lle) Het— unknown origin	Not identified
ACTA2*	NE	Not identified	NE	NE	NE	NE	c. 535c>T (p.Arg179Cys) Het—unknown origin
MYH11**	NE	Not identified	NE	NE	NE	NE	NE
Gender	M	F	M	F	M	F	F
Newborn data	Live	Live	Live	Live	Live	Live	Live
vitality
Weight [g] (centile)	1,935 (<p10)	4,150 (>90)	2,600 (10–25)	3,360 (50–75)	3,025 (75–90)	3,320 (50–75)	3,690 (75–90)
Length [cm] (centile)	38 (<p10)	50 (50–75)	42 (10)	50 (25–50)	49 (75–90)	NS	48 (10–25)
OFC [cm] (centile]	31 (p10)	34.5 (50–75)	32 (25)	36 (90)	33 (50–75)	NS	35 (75)
Gestational age [weeks]	36	38	36	39	35	37	38
Apgar score	3/8	9/10	7/10	9/10	8/9	NE	8/9/9
Age	Death (1 mo)	Death (1 y 8 mo)	Death (7 mo)	Death (11 mo)	5 mo	5y	2 y 10 mo
Paternal age (years)a	21	23	30	38	32	34	38
Maternal age (years)a	24	2B	29	34	25	27	29
Prenatal findings	Hydronephrosis, megacystis, oligohydramnios	Hydronephrosis, megacystis, polyhydramnios	Hyperechogenic kidneys, hydroureter, megacystis, club feet	Megacystis, polyhydramnios	Megacystis	Megacystis	Hydronephrosis, hydroureter, megacystis
Fetal bladder diversion	Vesicoamniotic shunt	−	−	−	−	−	−
Megacystis	+	+	+	+	+	Episodic	+
Prune-belly	+	+	+	−	−	−	−
phenotype
Microcolon	+	+	+	−	+	−	−
Intestinal hypoperistalsis	+	+	+	+	+	+	+
Mydriasis	NE	+	−	−	−	−	+
Parenteral  lnutrition  dependence	+ (Total)	+ (Total)	+ (Total)	+ (Total after 4 mo)	+ (Total)	+ (Intermittent)	+ (Intermittent)
Bladder catheterization and/or vesicostomy	+	+	+	+	+	−	+
Other findings	Hydronephrosis, sparse renal cyst, renal failure, dilated stomach, intestinal malrotation, hypoplastic terminal ileum	Hydronephrosis, intestinal malrotation, hypoplastic small intestine	Hydronephrosis, renal cysts, hydroureter, stomach dilatation		Cerebral ventriculomegaly, temporal epileptiform discharges	Stomach and large intestine dilatation	Hydronephrosis, hydroureter, intestinal malrotation, ASD, ductus arteriosus aneurysm, pulmonary valve dilatation, cerebrovascular abnormalities
Histologic findings	Renal dysplasia, chronic pyelonephritis, acute tubular necrosis, intrahepatic cholestasis	Vacuolization of muscularis propria layer (bladder and small intestine)	NE	Normal (stomach, small and large intestine)	Enterocolitis	Vacuolization and fibrosis of muscularis propria layer (stomach, gallbladder, small and large intestine)	Normal (large intestine)
Surgery	Exploratory laparotomy, gastrostomy, ileostomy, vesicostomy, appendicectomy	Gastrostomy, ileostomy, vesicostomy, multivisceral transplantation	Ileostomy, vesicostomy	Gastrostomy, ileostomy	Ileostomy	Ileostomy, partial gastrectomy with Roux loop anatomosis, cholecystectomy, exploratory laparotomy, right hemicolectomy	Ligation of ductus arteriosus, gastrostomy, sigmoid resection
Follow up	Deceased	Deceased	Deceased	Deceased	Supportive treatment	Episodic intestinal pseudo-obstruction, supportive treatment	Supportive treatment

^aparental age in years at the birth of the patient; ASD, atrial septal defect; CIPO, chronic intestinal pseudo-obstruction; F, female; Het, heterozygous; M, male; MMIHS, megacystis-microcolon-intestinal hypoperistalsis; mo, months old; MSMDS, i smooth muscle dysfunction syndrome; NE, not evaluated; OFC, occipital frontal circumference; y, years old; +, feature present; —, feature absent.

^*Evaluated by Sanger sequencing in patients 2-7 (both genes were also tested by whole-exome sequencing in patient 2).

^**Evaluated by whole-exome sequencing in patient 2.

Molecular Study

The investigation strategy was defined based on the phenotype. The sporadic cases of MMIHS and CIPO were studied by Sanger sequencing of ACTG2. The molecular investigation of the family with two affected siblings with MMIHS was performed only in one patient (patient 2), because DNA from her brother was not available. When the study was designed, ACTG2 was the only gene associated with MMIHS. Despite the presence of parental consanguinity and recurrence in siblings suggesting an AR inheritance in this family, we initially tested for dominant loci based on the hypothesis of parental gonadal mosaicism. Due to overlap between the MMIHS and MSMDS phenotype, we also performed Sanger sequencing of ACTA2 in patient 2; however, since the sequencing was negative for both genes, the whole-exome sequencing (WES) was performed for the trio (patient 2 and her parents).

In the patient with signs of MSMDS, ACTG2, and ACTA2 were sequenced by the Sanger method. At first, we tested ACTG2 because there was no information about vascular abnormality in the patient at that time. This data were provided during the follow up.

The molecular tests were performed on DNA extracted from lymphocytes (patients 2, 3, 4, 5, and 7, parents of patients 1, 2, 3, 4, and 5, and mother of patient 7); and from saliva (patient 6 and her mother). DNA sample from patient 1, and the fathers of patients 6 and 7 were unavailable.

For the ACTG2 and ACTA2 Sanger sequencing, the primers were designed to encompass all coding exons and their flanking regions. The gene nucleotide numbering of ACTG2 and ACTA2 is according to the sequence NM_001615 and NM_001613, respectively. The primers and the details about Sanger sequencing of ACTG2 and ACTA2 are described in the supplementary material (SM1). For the WES, the Agilent SureSelect Human All Exon V4–51Mb kit (Agilent Technologies, Santa Clara, CA) was used to capture the target regions. The WES (paired end 100 bp reads) was performed using the Illumina HiSeq2500 platform (Illumina, Inc. San Diego, CA). The phenoDB analysis tool was applied to filter and prioritize rare functional variants (missense, nonsense, splice site variants, and indels) [Sobreira et al., 2015]. The detailed information regarding the WES is described in the supplementary material (SM2). For the novel variants, the pathogenicity was tested in different ways: (i) in silico analysis by PolyPhen-2, Mutation Taster, MutPred, Panther, SNAP, Phd-SNP, and SNPs&GO (web sources); (ii) research in different databases: dbSNP, 1000 Genomes project data, and Exome Variant Server (web sources); (iii) sequencing of the parental samples whenever possible and 100 alleles from control individuals. We used the HOPE program to analyze the structural and functional effects of novel variants [Venselaar et al., 2010].

Review of the Literature

Due to similarities between the MMIHS and MSMDS, and the recent advances in understanding their genetic bases, we reviewed both phenotypes in order to better characterize these diseases, in an attempt to verify a possible genotype–phenotype correlation. We performed the literature review through the PubMed site using the following key words: “megacystis microcolon intestinal hypoperistalsis,” “multisystemic smooth muscle dysfunction,” “ACTG2,” “ACTA2,” and “MYH11.” Only the articles reporting patients with molecular diagnosis were included. For the cases reported more than once we included the first description.

Results

Clinical Description

The clinical data and molecular results of the individuals are summarized in Table I. In addition, more detailed clinical information from each patient is presented below according to the respective phenotypes. In all cases reported here, the parents were asymptomatic.

Table II. MMIHS and MSMDS: Summary of Clinical Data and Genotype of Individuals With Molecularly Confirmed Diagnosis.

Phenotype	Megacystis-microcolon-intestinal hypoperistalsis syndrome (MMIHS)		Multisystemic smooth muscle dysfunction syndrome (MSMDS)
Gene	ACTG2a	MYH11b	ACTA2c
Number of probands	23	1	23
Parental consanguinity	−	+ (Algerian origin)	−
Inheritance	AD	AR	AD
Pathogenic variant	Heterozygous—all missense	Homozygous—stop codon	Heterozygous—all missense (codon 179)
De novo variant	16/18d	No	23/23
Familial recurrence of the phenotype	1/23e	−	−
Megacystis	21/23	+	4/23
Bladder dysfunction without	2/23	−	10/23
megacystis
Microcolon	16/23	+	−
Intestinal hypoperistalsis	21/21f	+	5/23
Prune-belly	2/20g	+	2/23
Mydriasis	−	−	23/23
Patent ductus arteriosus	−	−	23/23
Vascular involvement			23/23
Cerebrovascular	−	−	21/23
Aortic aneurism	−	−	14/23

AD, autosomal dominant, AR, autosomal recessive, MMIHS, megacystis-microcolon-intestinal hypoperistalsis syndrome; MSMDS, multisystemicsmooth muscle dysfunction syndrome; +, feature present; —, feature absent. A summary of phenotype, genotype, and family history data of all individuals with ACTG2-related disorder, including MMIHS and milder phenotypes, is described in the supplementary material (SM5).

^a[Thorson et al., 2014]; [Wangler et al., 2014]; [Tuzovic et al., 2015]; [Halim et al., 2016].

^b[Gauthier et al., 2014].

^c[Milewicz et al., 2010]; [Al-Mohaissen et al., 2012]; [Moller et al., 2012]; [Munot et al., 2012]; [Richer et al., 2012]; [Meuwissen et al., 2013]; [Moosa et al., 2013]; [Amans et al., 2014]; [Brodsky et al., 2014]; [Roulez et al., 2014]; [Yetman et al., 2015].

^dUnknown finding in five of the 23 individuals (18 evaluated patients), 16/18, one individual inherited the variant from the father, who presented with milder disease—family 34 [Wangler et al., 2014] and parental inheritance in a family due to probable gonadal mosacism—family 1 [Tuzovic et al., 2015].

^eThe recurrence of MMIHS was reported in a family with two affected siblings and normal parents—family 1 [Tuzovic et al., 2015].

^fUnknown finding in two of the 23 individuals (21 evaluated patients).

^gUnknown finding in three of the 23 individuals (20 evaluated patients).

Megacystis-Microcolon-Intestinal Hypoperistalsis Syndrome (MMIHS)

Patients 1 and 2 (family 1)

A male and a female infant, respectively, were born from consanguineous parents (F = 0.0625) and both manifested the first symptoms during the prenatal period. Hydronephrosis and megacystis were identified in both by the 13 and 20th week, respectively. While severe oligohydramnios was observed at19 weeks in patient 1, polyhydramnios was detected after 25th week in patient 2. Patient 1 was unsuccessfully treated with a vesicoamniotic shunt. After his birth, symptoms of gastrointestinal hypoperistalsis (bilious emesis, enteral feeding intolerance, and absent meconium elimination), and microcolon were observed suggesting MMIHS. He died due to renal failure at the end of the 1st month. The presence of mydriasis was not evaluated. The diagnosis of MMIHS in patient 2 was considered during the pregnancy because of megacystis plus the previously affected brother. After birth, severe gastrointestinal hypoperistalsis and microcolon were confirmed. In addition, she presented with congenital mild mydriasis with pupils slightly reactive to the light reflex. She was followed by us up to 6 months, when she transferred to another country in order to try the multivisceral transplantation, which finally occurred when she was 16 months old. However, she died due to an infection 4 months after the transplantation.

Patient 3 (family 2)

A male infant and the first child of young parents. The megacystis was identified at the 21st week of pregnancy. Hyperechogenic kidneys, hydroureter, club feet, and oligohydramnios were seen on the subsequent prenatal ultrasound examination. After birth, the patient developed symptoms of gastrointestinal hypomotility such as enteral feeding intolerance, abdominal distention, and absence of meconium elimination. Microcolon was detected and MMIHS was diagnosed. There was no evidence of mydriasis. Despite the supportive therapy he died at 7 months.

Patient 4 (family 3)

A female infant and the second child of young parents. Megacystis and polyhydramnios were detected by prenatal ultrasound at the 30th week of pregnancy. After birth, she presented with mild gastrointestinal hypomotility symptoms and tolerated oral feeding during 4 months. After this period, she developed progressive symptoms of intestinal dysmotility that included enteral feeding intolerance and chronic constipation without evidence of mechanical obstruction. She required parental nutrition and despite the supportive therapy, she died at 11 months due to sepsis. There was no evidence of mydriasis.

Patient 5 (family 4)

A male infant and the second child of young parents. Megacystis was identified by prenatal ultrasound at the 35th week. The amniotic fluid volume was normal. After birth, intestinal hypomotility symptoms such as vomiting, constipation, and abdominal distention and microlon were identified, supporting the diagnosis of MMIHS. There was no evidence of mydriasis. He did not tolerate oral feeding and is total parental nutrition– nutrition (TPN)-dependent since the first days of his life. Due to meningitis, the patient developed secondary hydrocephalus.

Patients 1, 2, and 3 presented with signs of prune-belly at birth. All patients were parental nutrition-dependent, but the patient 4 tolerated oral feeding for a short period.

Chronic Intestinal Pseudo-Obstruction (CIPO)

Patient 6 (family 5): A female child with two asymptomatic older brothers. Although her mother reported that bladder dilatation was observed during a prenatal ultrasound examination in the 3rd trimester, at birth there was no evidence of megacystis or visceral motility dysfunction. The symptoms of CIPO, such as recurrent episodes of vomiting, constipation, and abdominal distention, began when she was 2 years old. These acute symptoms occurred during a varicella zoster infection and were associated with urinary retention, and transitory megacystis during some episodes. The urinary symptoms spontaneously resolved after the improvement of the gastrointestinal complaints. Besides episodic megacystis, no other abnormality was identified in the urinary system. During the acute episodes of intestinal pseudo-obstruction, the treatment was conservative wherever possible and consisted of oral fasting, tube for gastrointestinal decompression, and parenteral nutrition with gradual transition to enteral feeding according to her clinical recovery. So far, her nutrition is totally enteral. Therapeutic surgical procedures (Table I) were performed only when conservative measures were inefficient. Only the mother was available for molecular study.

Multisystemic Smooth Muscle Dysfunction Syndrome (MSMDS)

Patient 7 (family 6)

She is the first child of young parents. During the pregnancy, megacystis, hydroureter, and hydronephrosis were identified by the ultrasound examination. After birth she presented with congenital mydriasis, intestinal malrotation, and symptoms of intestinal hypoperistalsis including vomiting and constipation. Cardiovascular abnormalities included atrial septal defect, ductus arteriosus aneurysm, and pulmonic valve dilatation. Surgical closure of the ductus arteriosus was performed in the 1st month of life. MRI/MRA performed at 10 months showed increased T2 and FLAIR signal intensity in the supratentorial region, predominantly in the centrum semiovale, corona radiate, and frontal subcortical region suggestive of terminal ischemic injuries. Bilateral stenosis of the extracranial carotid artery with abnormally straightened arterial course was noted, including the terminal segments and hypoplastic posterior circulation system with significant reduction of basilar flow. The posterior circulation has been maintained through the communicating system. There is no evidence of aneurysm. She was partially fed by parenteral nutrition in her first months of life and currently she receives an enteral diet by gastrostomy. Only her motherwas available for molecular study.

Molecular Study

As shown in Table I, we found four heterozygous pathogenic variants in ACTG2 (patients 3, 4, and 5 with MMIHS and patient 6 with CIPO), and one heterozygous variant in ACTA2 in patient 7 with MSMDS. The DNA chromatograms with the variants are shown in the supplementary material (Fig.—SM3). In ACTG2, the variants c.532C>T, c.770G>A, and c.533G>T were identified, respectively, in patients 3, 4, and 5. They were absent in the parents and considered apparently de novo variants. The change in the same gene in patient 6 (c.584C>T) is a novel mutation. It was not detected in her mother and the DNA from her father was not available. This variant was predicted as pathogenic by all software programs except by SNAP and SNPs&GO (see pathogenicity prediction in the supplementary material—SM4). According to the HOPE program, the substitution of threonine by isoleucine at the position 195 is probably damaging to the protein, because unlike the wild-type, this change introduces a larger molecular weight and more hydrophobic residue in the mutated protein, possibly affecting the hydrogen bond formation between the original residue and the methionine at position 191. In addition, the mutant residue is located in the surface of the protein, near a highly conserved position probably disturbing interactions with other molecules or other parts of the protein. This change was absent in 100 alleles from control individuals. Patient 7 presented with the known c.535C>T variant in ACTA2. It was not detected in her mother and the DNA from her father was unavailable.

The molecular investigation of the family with two affected sibs with MMIHS (patients 1 and 2), was performed only in patient 2. The ACTA2 and ACTG2 sequencing did not show a pathogenic variant. In the WES analysis, we initially prioritized the identification of homozygous variants. We also considered other possibilities such as AR—compound heterozygous and AD with recurrence explained by parental gonadal mosaicism. A total of 23 candidate genes were identified: AR—homozygous (10 genes), AR— compound heterozygous (four genes), and AD (nine genes), but no association between a specific gene and the phenotype has yet been proven for the candidates genes. We did not identified pathogenic variant in MYH11 in patient 2.

Review of the Literature

Sixteen articles including clinical and molecular data from individuals with MMIHS or MSMDS were reviewed. As summarized in Table II, for the individuals with MMIHS and pathogenic variants in ACTG2 (23 probands), the intestinal hypoperistalsis (21/21), and the bladder dysfunction are essential symptoms (23/23). The vesical hypocontractility often manifests as megacystis (91.3%— 21/23), and microcolon appears in 69.5% (16/23). Mydriasis and PDA were not identified in patients. The single reported individual with loss-of-function variant in MYH11 presented with the typical signs of MMIHS. The main findings in individuals with MSMDS and proven pathogenic variants in ACTA2 (probands) are mydriasis (23/23), PDA (22/23), and vascular abnormalities (22/23). Visceral dysfunction is mainly due to bladder involvement with (4/23) or without (10/23) megacystis, while intestinal hypoperistalsis is less frequent (5/23), and microlon is absent in patients with MSMDS. Although the prune-belly phenotype is not common, this feature was described in both phenotypes.

Discussion

In the present study, we report the clinical data and molecular investigation of seven individuals from six families presenting different phenotypes of visceral myopathy. Pathogenic variants were found in five families, each with only one affected individual.

In patients with MMIHS, we identified three individuals with heterozygous variants in ACTG2. All were sporadic and the respective variants have been previously described—c.532C>T [Thorson et al., 2014; Wangler et al., 2014; Halim et al., 2016], c.770G>A [Tuzovic et al., 2015], and c.533G>T [Thorson et al., 2014; Halim et al., 2016]. For the family with two siblings (typical phenotype of MMIHS and family history suggestive of AR pattern), the molecular investigation showed no pathogenic variants in ACTG2, ACTA2, or MYH11. Although we have identified 23 candidate genes by WES, a pathogenic variant could not yet be proven. Through the analysis of the exome, we did not find a pathogenic variation in the following genes: CHRM3 (associated with urinary dysfunction and mydriasis in humans) [Weber et al., 2011], and other genes such as CHRNA3, CHRNB2, or CHRNB4 (based on MMIHS-like phenotype—mydriais, megacystis, hypoperistalsis, gastric, and intestinal distention observed in mice) [Xu et al., 1999a,b].

Most of the published cases of MMIHS are sporadic and caused by de novo heterozygous variants in ACTG2. Recurrence in siblings related to ACTG2 was recently reported under the hypothesis of gonadal mosaicism because the parents were asymptomatic and did not present the variant in the sequencing [Tuzovic et al., 2015]. In another patient, the variant in ACTG2 had paternal origin, but the father had a milder phenotype [Wangler et al., 2014]. However, some individuals with MMIHS, but without pathogenic variant in ACTG2 [Wangler et al., 2014; Halim et al., 2016], including two patients whose parents are consanguineous [Halim et al., 2016], suggest genetic heterogeneity and AR pattern in some. AR inheritance has been suggested because of the presentation of a number of families with recurrence in siblings and/or parental consanguinity, and asymptomatic parents [Mc Laughlin and Puri, 2013]. The molecular evidence supporting this inheritance pattern was recently demonstrated by Gauthier et al. [2014] in an individual with MMIHS and a homozygous variant (c.3598A>T—p.LysK1200Ter) in MYH11, whose parents are consanguineous. Although a functional study has not yet been performed, this variant was absent in the genome database, as well as in 323 control individuals. This variant is predicted as likely to be deleterious by in silico analysis [Gauthier et al., 2014]. These authors defended the hypothesis that this variant could interfere with dimerization of the protein or inducing the degradation of the transcript through nonsense-mediated decay. The association of MYH11 with MMIHS is supported by animal models in which the mice with homozygous deletion of Myh11 presented with symptoms such as giant bladder, abnormal intestinal movement, and delay in the closure of ductus arteriosus [Morano et al., 2000].

The individual with CIPO has the novel heterozygous variant c.584C>T in ACTG2. Its pathogenicity was suggested by in silico analysis and reinforced by its absence in different databases, and in 100 alleles from control individuals.

Until now, 31 families with ACTG2-related disorders and 15 different pathogenic variants in ACTG2 were described, all but one being missense variants—supplementary material (SM5) [Lehtonen et al., 2012; Holla et al., 2014; Thorson et al., 2014; Wangler et al., 2014; Klar et al., 2015; Tuzovic et al., 2015; Halim et al., 2016]. The only exception is an in tandem base substitution reported by Klar et al. [2015]. Many missense variants (10/14 = 71.4%) resulting in a substitution of an arginine for another amino acid in different codons [Lehtonen et al., 2012; Holla et al., 2014; Thorson et al., 2014; Wangler et al., 2014; Tuzovic et al., 2015; Halim et al., 2016].

Due to diversity of the manifestations of visceral myopathy related to ACTG2, Wangler et al. [2014] proposed a phenotypic spectrum of ACTG2-disorders ranging from MMIHS, the most severe presentation characterized by early onset of symptoms (prenatal or neonatal) with severe dysmotility of bowel and bladder, and parenteral nutrition dependence, to milder forms with variable degree of gastrointestinal, and urinary dysfunction, including CIPO, also referred as familial visceral myopathy [Lehtonen et al., 2012]. Indeed, CIPO was the first phenotype associated with ACTG2 variants [Lehtonen et al., 2012]. It is not possible to define a genotype–phenotype correlation, since a same variant was observed in MMIHS and in individuals with a milder phenotype [Thorson et al., 2014; Wangler et al., 2014; Tuzovic et al., 2015; Halim et al., 2016]. However, cases resulting from de novo variants are frequently more severe than inherited cases [Wangler et al., 2014]. Although the penetrance seems to be complete, there is a considerable intra-familial variability and milder presentation may not be recognized [Lehtonen et al., 2012; Wangler et al., 2014]. Even in the severe phenotype, some variability related to the onset (prenatal or neonatal), voiding dysfunction and TPN dependence, have been observed [Wangler et al., 2014].

Beyond gastrointestinal and urological manifestations, others symptoms identified in individuals with pathogenic variants in ACTG2, including biliary complications (cholecystistis and cholelithiasis), and impaired uterine contraction, suggest that the myopathy may not be restricted to the gastrointestinal and urinary tracts, thus, expanding the phenotype [Lehtonen et al., 2012; Klar et al., 2015]. This is reinforced by the histologic findings of myopathy detected in the gallbladder of the patient with CIPO here reported.

In the patient with MSMDS, we identified a previously reported ACTA2 variant, c.535C>T [Meuwissen et al., 2013]. Heterozygous variants in ACTA2 were first described in individuals with thoracic aortic aneurysm with dissection (TAAD) [Guo et al., 2007], coronary artery disease, stroke, and Moyamoya disease [Guo et al., 2009]. Then, a severe phenotype named multisystemic smooth muscle dysfunction syndrome (MSMDS) including PDA, congenital mydriasis, hypotonic bladder, ascending aorta aneurysms, and cerebrovascular disease was associated with a de novo heterozygous variant in the codon 179 [Milewicz et al., 2010]. The p.Arg179His variant is the most common substitution associated with this severe phenotype; however, other variants in the same codon (p.Arg179Cys and p.Arg179Lys) result in a similar phenotype [Moller et al., 2012; Munot et al., 2012; Meuwissen et al., 2013]. Patient 7, described in the present study with MSMDS, is the second individual identified with the c.535C>T (p.Arg179Cys) in ACTA2. The recurrent association between a severe clinical presentation and the arginine substitution in the 179 position suggests a genotype-phenotype correlation. The variant in this codon could cause a severe systemic disease because of the arginine in the 179 position, that is, near to a keyprotein–protein interaction site [Milewicz et al., 2010]. Interestingly, one individual reported by Roder et al. [2011] presenting the p.Arg179His variant has Moyamoya disease without other manifestations observed in MSMDS.

In order to better understand the overlap between MMIHS and MSMDS, two severe phenotypes with visceral myopathy, the review of published cases with proven pathogenic variants shown in Table II highlights more clearly the features characterizing each phenotype. While the visceral dysfunction in individuals with MMIHS manifests mainly as intestinal hypoperistalsis and megacystis, bladder involvement without megacystis is observed in patients with MSMDS. Some findings seem quite specific of each phenotype, as microcolon in MMIHS and mydriasis, PDA, and vascular abnormalities in MSMDS. Therefore, these findings may direct the molecular investigation of a patient with signs of visceral smooth muscle dysfunction. On the other hand, when the phenotype is more suggestive of MMIHS, there are no clear clinical signs for the differentiation between autosomal dominant (ACTG2) and recessive (MYH11) inheritance. Autosomal recessive inheritance seems to be less frequent than sporadic and autosomal dominant, and further studies are needed for a better etiologic definition of MMIHS. Although PDA had not been identified in individuals with homozygous MYH11 variant [Gauthier et al., 2014], it remains a possibility due to involvement of ductus arteriosus in Myh11 mice and in individuals with heterozygous variants in MYH11 [Zhu et al., 2006]. Although the prune-belly appearance is not common in either MMIHS or MSMDS, this phenotype should be taken into account during the initial evaluation of a child with clinical signals of visceral motility dysfunction.

The main features present in MMIHS and MSMDS can be identified in other phenotypes such as the X-linked chronic intestinal pseudo-obstruction (CIPO-X), and urinary bladder disease/prune-belly caused by FLNA and CHRM3 variants, respectively [Kapur et al., 2010; Weber et al., 2011]. In CIPO-X, besides CIPO, other manifestations including gastrointestinal abnormalities as intestinal malrotation, short small intestine, microcolon, and pyloric stenosis are described. Extra enteric findings like PDA, vascular abnormality in CNS, megacystis, periventricular nodular heterotopy, thrombocytopenia, and dysmorphic facies were observed in some patients [Clayton-Smith et al., 2009; Kapur et al., 2010]. The phenotype called urinary bladder disease/prune-belly, caused by homozygous CHRM3 variants, is associated with bladder dysfunction, mydriasis, and dry mouths without intestinal abnormality [Weber et al., 2011].

The similarity of these diseases can be explained by the involvement of different parts of visceral contractile apparatus, including contractile filaments as α and γ-actin, myosin heavy chain 11, and filamin A (ACTA2, ACTG2, MYH11, and FLNA respectively), and muscarinic acetylcholine receptor (CHRM3). However, unexpected findings highlight interesting points regarding the expression of these proteins. For instance, the intestinal and bladder dysfunction in patients with ACTA2 variants indicate that the α-actin isoform probably has a more important role in enteric smooth muscle cell than previously supposed, where the γ-actin is the major isoform [Milewicz et al., 2010]. The absence of vascular manifestations in the ACTG2-related disorders reinforces the concept that ACTG2 does not have an important function in vascular smooth muscle cell contraction.

With respect to mydriasis, it is usually described in MSMDS (ACTA2) as well as in urinary bladder disease/prune-belly (CHRM3). However, the pupils and their responses to light, and cholinergic agents seem to be different in these disorders. While the pupils in MSMDS are usually nonreactive to the light and the pilocarpine [Moller et al., 2012; Richer et al., 2012; Roulez et al., 2014], the pupils of the individuals with variants in CHRM3 are described as having impaired constriction to light [Weber et al., 2011]. The presence of a weak light reflex and a very little change in pupil size when pilocarpine is instilled in Chrm3 mutant mice support the hypothesis that other mechanisms could be related to the contraction of the pupillary sphincter muscle [Matsui et al., 2000]. Finally, despite CHRM3 being the main muscarinic receptor involved in visceral smooth muscle contraction, gastrointestinal symptoms were not evidenced in patients reported by Weber et al. [2011], suggesting a non-M3 acetylcholine receptor response or possible upregulated non cholinergic mechanisms of contraction maintaining gastrointestinal function, similar to those observed in animal models [Uchiyama and Chess-Williams, 2004].

In conclusion, a clinical and molecular study of seven individuals from six families with disorders of visceral smooth muscle contraction showed mutations in ACTG2 (four) and in ACTA2 (one) in five patients, and failed to find pathogenic variant, especially in MYH11, in the only consanguineous family, suggesting the existence of other gene(s) related to autosomal recessive inheritance. The results reinforce the clinical and genetic heterogeneity of the visceral myopathies. The review of MMIHS and MSMDS, similar phenotypes, suggest that some features can direct the molecular investigation of a patient with visceral myopathy. Finally, considering that the genes related to visceral myopathies in humans and/or in mouse models are genes encoding contractile proteins or acetylcholine receptors in enteric smooth muscle, further analysis should prioritize genes related to visceral contractile apparatus.

Addendum Added During Revision of the Manuscript

After the submission of this manuscript, two articles were published in on line version describing patients with ACTG2-related disorder:MMIHS [Lu et al.,2016] and MMIHS, and CIPO [Matera et al., 2016].

Web Sources

PolyPhen-2: http://genetics.bwh.harvard.edu/pph2/

Mutation Taster: http://www.mutationtaster.org/

MutPred: http://mutpred.mutdb.org/

Panther: http://www.pantherdb.org/

SNAP: https://rostlab.org/services/snap/

Phd-SNP and SNPs&GO: http://snps.biofold.org/snps-and-go/snps-and-go.html

dbSNP: http://www.ncbi.nlm.nih.gov/projects/SNP/

1000 Genomes project data: http://browser.1000genomes.org/index.html

Exome Variant Server: http://evs.gs.washington.edu/EVS/