Connexin26 hemichannels with a mutation that causes KID syndrome in humans lack sensitivity to CO2

Abstract
Main text
References
Article and author information
Metrics

Abstract

Mutations in connexin26 (Cx26) underlie a range of serious human pathologies. Previously we have shown that Cx26 hemichannels are directly opened by CO₂ (Meigh et al., 2013). However the effects of human disease-causing mutations on the CO₂ sensitivity of Cx26 are entirely unknown. Here, we report the first connection between the CO₂ sensitivity of Cx26 and human pathology, by demonstrating that Cx26 hemichannels with the mutation A88V, linked to Keratitis-Ichthyosis-Deafness syndrome, are both CO₂ insensitive and associated with disordered breathing in humans.

https://doi.org/10.7554/eLife.04249.001

Main text

Connexin26 (Cx26) is one of 21 connexin genes found in humans (Cruciani and Mikalsen, 2006). The canonical function of connexins is to form gap junctions in which two hexameric connexons, or hemichannels, in closely apposed membranes dock together to form an intercellular channel. However connexins can also function as hemichannels, thereby providing large conductance channels, which allow passage of small molecules such as ATP into the extracellular space (Stout et al., 2004; Wang et al., 2013). We have recently shown that Cx26 hemichannels are directly sensitive to CO₂ (Huckstepp et al., 2010a; Meigh et al., 2013). When CO₂ binds to Cx26, it carbamylates K125, forms a salt bridge to R104 and opens the hemichannel (Meigh et al., 2013). Cx26 hemichannels are thus a source of CO₂-gated ATP release (Huckstepp et al., 2010a).

Mutations of Cx26 are the commonest cause of non-syndromic hearing loss (Cohn and Kelley, 1999; Kelley et al., 2000; Xu and Nicholson, 2013). Some of these mutations cause loss of functional protein, while other mutations result in gap junctions and hemichannels with altered properties. However the effect of these mutations on the CO₂ sensitivity of Cx26 has never been examined. Some missense mutations of Cx26 cause serious pathologies in humans, such as the very rare ectodermal disorder, Keratitis-Ichthyosis-Deafness (KID) syndrome. KID syndrome involves a combination of deafness, visual impairment, and dermatological abnormalities (Caceres-Rios et al., 1996). About 100 cases have been reported in the literature, and of these around 70% are caused by de novo mutations in Cx26, with the remainder being inherited in an autosomal dominant manner or via germ line mosaicism (Sbidian et al., 2010). To date there are nine missense mutations that can cause KID syndrome (Xu and Nicholson, 2013). The severity of the symptoms of KID syndrome depends on the particular mutation in Cx26 (Janecke et al., 2005; Jonard et al., 2008).

The mutation, Cx26^A88V, is linked to a very severe form of KID syndrome, which is fatal in infancy (Haruna et al., 2010; Koppelhus et al., 2010). In one of the original reports linking Cx26^A88V to KID syndrome, the patient required mechanical ventilation (Koppelhus et al., 2010), suggesting a possible effect of the mutation on the neural control of breathing. In KID syndrome caused by a different missense mutation (G45E), which is fatal within the first year of life, there are also reports of breathing problems. One patient required mechanical ventilation immediately after birth (Janecke et al., 2005) and a second died from breathing failure (Sbidian et al., 2010). Nevertheless, without detailed recordings of cardiorespiratory activity, it is not possible to know whether these patients experienced inadequate central respiratory drive. For other mutations linked to KID syndrome there are no reports of abnormal breathing in the literature.

The reason why the A88V and G45E mutations should cause such pervasive and severe pathology remains unclear as subunits of Cx26^A88V and Cx26^G45E form both functional gap junctions and hemichannels (Gerido et al., 2007; Mhaske et al., 2013). Expression of Cx26^A88V in HeLa cells gives rise to enhanced hemichannel-mediated currents (compared to wild type Cx26, Cx26^WT) at positive transmembrane potentials and in the absence of extracellular Ca²⁺, leading to the suggestion that this mutation represents a gain of function (Mhaske et al., 2013). The G45E mutation, also causes enhanced hemichannel activity in the absence of extracellular Ca²⁺, and increased permeability to Ca²⁺ (Gerido et al., 2007; Sanchez et al., 2010). A gain of function has therefore been suggested as underlying the actions of this mutation too. Although the absence of extracellular Ca²⁺ opens connexin hemichannels, this condition is unlikely to occur in physiological systems. Thus the consequences of the A88V and G45E mutations on physiologically relevant gating of Cx26 remain unclear.

We identified a patient with KID syndrome caused by a heterozygous Cx26 A88V mutation. This patient failed to breathe spontaneously at birth and initially required mechanical ventilation. Later when he started to breathe spontaneously, he continued to demonstrate periods of apnea and bradycardia. A pneumogram performed at a post-menstrual age of 40 weeks showed abnormal persistence of central apnea lasting ≥20 s and accompanied by periods of bradycardia and prolonged oxygen desaturation (Figure 1). This respiratory pattern is abnormal for the age of the infant and is suggestive of blunted chemosensory control of breathing. Given the previously described role of Cx26 in mediating the CO₂-dependent drive to breathe (Huckstepp et al., 2010b; Wenker et al., 2012), we considered whether the mutation A88V might alter the CO₂-sensitivity of Cx26.

Figure 1

Download asset Open asset

Incidence of central sleep apnea in a patient with Cx26^A88V.

Recording of cardiorespiratory activity during sleep from an infant at a post-menstrual age of 40 weeks diagnosed with KID syndrome. Traces of nasal air flow, thoracic movement, electrocardiogram (ECG), heart rate (HR) and arterial O₂ saturation show that this patient exhibited a prolonged period during which no effort was made to breathe and this was followed by pronounced bradycardia and arterial O₂ desaturation, all of which are characteristic of central sleep apnea. Unfortunately, at 2 months of age this patient died from overwhelming sepsis.

https://doi.org/10.7554/eLife.04249.002

We introduced the A88V mutation into Cx26 and then tested the CO₂ sensitivity of Cx26^A88V hemichannels expressed in HeLa cells via an established and sensitive dye-loading protocol (Huckstepp et al., 2010a; Meigh et al., 2013). Under conditions of normal extracellular Ca²⁺, HeLa cells expressing wild type Cx26 hemichannels readily load with carboxyfluorescein when exposed to a moderately hypercapnic saline (PCO₂ 55 mmHg) (Huckstepp et al., 2010a; Meigh et al., 2013). However HeLa cells expressing Cx26^A88V showed no such CO₂-dependent dye loading even when exposed to higher levels of PCO₂ (70 mmHg, Figure 2). The failure to exhibit CO₂-dependent dye loading was not due to a lack of functional hemichannels as the positive control of removing extracellular Ca²⁺, which opens all connexin hemichannels, caused robust dye loading (Figure 2). HeLa cells transfected with an empty vector do not show any dye loading in response to a CO₂ stimulus or removal of extracellular Ca²⁺ (Figure 2—figure supplement 1). Surprisingly therefore, the conservative mutation A88V caused Cx26 hemichannels to lose their sensitivity to CO₂. As this mutation is far from the residues involved in CO₂ binding (K125 and R104), the mechanism for the loss of CO₂ sensitivity is unclear.

Figure 2 with 1 supplement see all

Download asset Open asset

Cx26^A88V hemichannels are no longer sensitive to CO₂.

(Top) Images of HeLa cells expressing Cx26^A88V under control, hypercapnic and zero Ca²⁺ conditions. The cells were exposed to 200 µM carboxyfluorescein (CBF) for 5 min under each condition before being washed. Some low background loading of CBF is seen under control conditions. In presence of CO₂ no loading is seen. The positive control of removal of extracellular Ca²⁺ causes robust dye loading demonstrating the presence of functional hemichannels. (Bottom) Cumulative probability distributions of pixel intensity of HeLa cells expressing Cx26^A88V under control, hypercapnia (two levels of PCO₂) and zero Ca²⁺. Only the removal of extracellular Ca²⁺ causes dye loading as shown by the rightward shift of the curve to higher pixel intensities (p = 0.004, Mann Whitney U test compared to control). These distributions show all of the measurements made (minimum 40 cells each from five independent repetitions).

https://doi.org/10.7554/eLife.04249.004

As the missense mutations which underlie KID syndrome act in a dominant manner (Jonard et al., 2008; Xu and Nicholson, 2013), we tested whether the expression of Cx26^A88V subunits might have a dominant negative action on the CO₂ sensitivity of Cx26^WT. We transfected HeLa cells that stably expressed Cx26^WT with the Cx26^A88V subunit and documented their sensitivity to CO₂ following transfection. 4 days after transfection with Cx26^A88V, the HeLa cells still exhibited sensitivity to CO₂ (Figure 3A), but this was reduced compared to the Cx26^WT HeLa cells that had not been transfected with Cx26^A88V (Figure 3B). 5 and 6 days after transfection, the HeLa cells showed no sensitivity to CO₂ (Figure 3A). Nevertheless functional hemichannels were still present as the removal of extracellular Ca²⁺ caused dye loading (Figure 3A). The loss of CO₂ sensitivity was not simply a consequence of days in culture, as Cx26^WT HeLa cells that had not been transfected with Cx26^A88V retained their sensitivity to CO₂ over the whole period examined (Figure 3B). We therefore conclude that Cx26^A88V subunits are able to act in a dominant negative manner to cause loss of CO₂ sensitivity from wild type Cx26 hemichannels.

Figure 3

Download asset Open asset

Cx26^A88V hemichannels act in a dominant negative manner to remove CO₂ sensitivity from Cx26^WT.

(A) Cumulative probability distributions for CO₂-dependent dye loading in HeLa cells that stably express Cx26^WT, which have been transfected with Cx26^A88V. 4 days after transfection with Cx26^A88V the cells still exhibit significant sensitivity to 55 mmHg PCO₂ stimulus (p = 0.048 CO₂ compared to control, Mann Whitney U test). 5 and 6 days after transfection the CO₂ sensitivity of the HeLa cells was abolished. On all 3 days, the positive control of zero Ca²⁺ caused dye loading, demonstrating the presence of functional hemichannels. The graphs show all of the measurements made from 5 independent repetitions of the experiment. (B) Comparison of the sensitivity to CO₂ of HeLa cells stably expressing Cx26^WT which have been transfected with Cx26^A88V (Cx26^WT + Cx26^A88V, n = 5) with those that have not (Cx26^WT, n = 7). In the absence of transfection, the Cx26^WT-expresssing HeLa cells retain sensitivity to CO₂ on all 3 days. By contrast Cx26^A88V causes significantly depressed CO₂ sensitivity 4 days after transfection (p = 0.001), and loss of sensitivity on days 5 (p = 0.024) and 6 (p = 0.001). Comparisons of Cx26^WT with Cx26^WT + Cx26^A88V via Mann Whitney U test, and False Discovery Rate procedure for multiple comparisons (Curran-Everett, 2000). Error bars upper and lower quartiles.

https://doi.org/10.7554/eLife.04249.005

This is the first instance in which a mutation linked to serious human pathologies has been demonstrated to abolish the CO₂ sensitivity of Cx26. This in turn suggests that Cx26-mediated CO₂ sensing may be important for human physiology in the range of contexts that are associated with the diverse pathologies linked to this mutation. In the closely related β connexin, connexin30 (Cx30), the mutation A88V connected to Clouston's Syndrome (Bosen et al., 2014), may result in constitutively open Cx30 hemichannels (Essenfelder et al., 2004). However Cx30 is also opened by CO₂ (Huckstepp et al., 2010a) and the effect of this mutation on the CO₂ sensitivity of Cx30 has not yet been investigated. There are no reports in the literature of disordered breathing in patients with Clouston's syndrome.

Previous studies suggesting that the A88V mutation gave a gain of function in Cx26, examined hemichannel function in the absence of extracellular Ca²⁺ (Mhaske et al., 2013). As the CO₂ sensitivity of the mutated hemichannel was not specifically examined in this previous study, it is likely that both sets of findings are correct—an enhancement of macroscopic hemichannel currents (Mhaske et al., 2013), and a loss of CO₂ sensitivity. However under physiological conditions of normal extracellular Ca²⁺ and in the presence of physiological CO₂/HCO₃⁻ buffering, we suggest that A88V should be considered as a loss-of-function mutation that effectively removes the capacity for CO₂-evoked ATP release via Cx26 hemichannels.

Our report is the first to document altered central respiratory drive in a KID syndrome patient. In rodents, CO₂-sensitivity of Cx26 contributes to the chemosensory control of breathing (Huckstepp et al., 2010b; Wenker et al., 2012). Although we do not know if the loss of CO₂ sensitivity in Cx26 contributed to the aberrant respiratory drive exhibited by this patient, these results are consistent with this possibility, and represent the first evidence to suggest that Cx26 hemichannels are a requisite component of the drive to breathe in humans. Overall the ability of physiological levels of PCO₂ to permit ATP release via Cx26 hemichannels may be important in the epidermis, cochlea and brain. Investigation of whether the absence of this mechanism of ATP release in patients with Cx26^A88V contributes to the serious pathological abnormalities that they suffer would seem to be warranted.

Materials and methods

Case study

Request a detailed protocol

The Institutional Review Board of the Connecticut Children's Medical Center considered this under the category of a case report and thus exempt from formal review.

Mutant connexin production

Request a detailed protocol

Puc19 Cx26^A88V was produced from wild type Cx26 via the Quikchange protocol using the following primers: forward 5′ TGT CCA CGC CGG TCC TCC TGG TAG C 3′ reverse 5′ GCT ACC AGG AGG ACC GGC GTG GAC A 3′. Cx26^A88V was subcloned into a pCAG-GS mCherry vector for mammalian cell transfection. Successful mutation of Cx26 was confirmed by sequencing which also verified that apart from the desired mutation the sequence was identical to the wild type.

HeLa cell culture

Request a detailed protocol

HeLa cells were cultured by standard methods in DMEM, 10% FCS with addition of 3 mM CaCl₂. For experimentation, cells were plated onto coverslips at a density of 5 × 10⁴ cells per well. Transient transfections were performed using the genejuice protocol.

Solutions used

Control aCSF

Request a detailed protocol

124 mM NaCl, 26 mM NaHCO₃, 1.25 mM NaH₂PO₄, 3 mM KCl, 10 mM D-glucose, 1 mM MgSO₄, 1 mM CaCl₂.

Zero Ca²⁺ aCSF

Request a detailed protocol

124 mM NaCl, 26 mM NaHCO₃, 1.25 mM NaH₂PO₄, 3 mM KCl, 10 mM D-glucose, 1 mM MgSO₄, 1 mM MgCl₂, 1 mM EGTA.

Hypercapnic (55 mmHg CO₂) aCSF

Request a detailed protocol

100 mM NaCl, 50 mM NaHCO₃, 1.25 mM NaH₂PO₄, 3 mM KCl, 10 mM D-glucose, 1 mM MgSO₄, 1 mM CaCl₂.

Hypercapnic (70 mmHg CO₂) aCSF

Request a detailed protocol

70 mM NaCl, 80 mM NaHCO₃, 1.25 mM NaH₂PO₄, 3 mM KCl, 10 mM D-glucose, 1 mM MgSO₄, 1 mM CaCl₂.

Hypercapnic aCSF was saturated with sufficient CO₂ (the remaining balance being O₂) to adjust the final pH (pH 7.5) to that of the control aCSF removing any potential effects of changes in extracellular pH.

All other solutions were saturated with 95% O₂/5% CO₂.

Dye loading protocols

Request a detailed protocol

Coverslips plated with HeLa cells transiently transfected with Cx26^A88V were exposed to Hypercapnic aCSF (55 mmHg or 70 mmHg) containing 200 µM CBF for 10 min. This was followed by control aCSF with 200 µM CBF for 5 min and a 30 min wash with control aCSF to ensure that all dye is removed from the outside of the cells.

A control comparison was used to establish any baseline loading occurring in the absence of a stimulus. HeLa cells expressing Cx26^A88V were exposed to 200 µM CBF in control aCSF for 15 min, followed by 30 min of washing.

A zero Ca²⁺ positive control was also performed to ensure functional connexin hemichannels were being expressed. Cx26^A88V expressing HeLa cells were exposed to 200 µM CBF in zero Ca²⁺ aCSF for 10 min. This was followed by control aCSF with 200 µM CBF for 5 min and 30 min of washing with aCSF.

Imaging and analysis

Request a detailed protocol

For each condition cells were imaged by epifluorescence (Scientifica Slice Scope, Cairn Research OptoLED illumination, 60× water Olympus immersion objective, NA 1.0, Hamamatsu ImageEM EMCCD camera, Metafluor software). Using ImageJ, the extent of dye loading was measured by drawing a region of interest (ROI) around individual cells and calculating the mean pixel intensity for the ROI. The mean pixel intensity of the background fluorescence was also measured in a representative ROI, and this value was subtracted from the measures obtained from the cells. All of the images displayed in the figures reflect this procedure in that the mean intensity of the pixels in a representative background ROI has been subtracted from every pixel of the image. The analysis of the CO₂ sensitivity of Cx26^A88V was performed as five independent repetitions in which at least 40 cells were measured in each condition, and the mean pixel intensities plotted as cumulative probability distributions.

References

1. Bosen F
2. Schutz M
3. Beinhauer A
4. Strenzke N
5. Franz T
6. Willecke K
(2014) The Clouston syndrome mutation connexin30 A88V leads to hyperproliferation of sebaceous glands and hearing impairments in mice
FEBS Letters 588:1795–1801.

https://doi.org/10.1016/j.febslet.2014.03.040
- Google Scholar
(1996) Keratitis, ichthyosis, and deafness (KID syndrome): review of the literature and proposal of a new terminology
Pediatric Dermatology 13:105–113.

https://doi.org/10.1111/j.1525-1470.1996.tb01414.x
- Google Scholar
1. Cohn ES
2. Kelley PM
(1999) Clinical phenotype and mutations in connexin 26 (DFNB1/GJB2), the most common cause of childhood hearing loss
American Journal of Medical Genetics 89:130–136.

https://doi.org/10.1002/(SICI)1096-8628(19990924)89:3<130::AID-AJMG3>3.0.CO;2-M
- Google Scholar
1. Cruciani V
2. Mikalsen SO
(2006) The vertebrate connexin family
Cellular and Molecular Life Sciences 63:1125–1140.

https://doi.org/10.1007/s00018-005-5571-8
- Google Scholar
1. Curran-Everett D
(2000)
Multiple comparisons: philosophies and illustrations

American Journal of Physiology. Regulatory, Integrative and Comparative Physiology 279:R1–R8.
- Google Scholar
(2004) Connexin30 mutations responsible for hidrotic ectodermal dysplasia cause abnormal hemichannel activity
Human Molecular Genetics 13:1703–1714.

https://doi.org/10.1093/hmg/ddh191
- Google Scholar
(2007) Aberrant hemichannel properties of Cx26 mutations causing skin disease and deafness
American Journal of Physiology. Cell Physiology 293:C337–C345.

https://doi.org/10.1152/ajpcell.00626.2006
- Google Scholar
1. Haruna K
2. Suga Y
3. Oizumi A
4. Mizuno Y
5. Endo H
6. Shimizu T
7. Hasegawa T
8. Ikeda S
(2010) Severe form of keratitis-ichthyosis-deafness (KID) syndrome associated with septic complications
The Journal of Dermatology 37:680–682.

https://doi.org/10.1111/j.1346-8138.2010.00839.x
- Google Scholar
1. Huckstepp RT
2. Eason R
3. Sachdev A
4. Dale N
(2010a) CO₂-dependent opening of connexin 26 and related beta connexins
The Journal of Physiology 588:3921–3931.

https://doi.org/10.1113/jphysiol.2010.192096
- Google Scholar
1. Huckstepp RT
2. Id Bihi R
3. Eason R
4. Spyer KM
5. Dicke N
6. Willecke K
7. Marina N
8. Gourine AV
9. Dale N
(2010b) Connexin hemichannel-mediated CO₂-dependent release of ATP in the medulla oblongata contributes to central respiratory chemosensitivity
The Journal of Physiology 588:3901–3920.

https://doi.org/10.1113/jphysiol.2010.192088
- Google Scholar
1. Janecke AR
2. Hennies HC
3. Gunther B
4. Gansl G
5. Smolle J
6. Messmer EM
7. Utermann G
8. Rittinger O
(2005) GJB2 mutations in keratitis-ichthyosis-deafness syndrome including its fatal form
American Journal of Medical Genetics. Part A 133A:128–131.

https://doi.org/10.1002/ajmg.a.30515
- Google Scholar
1. Jonard L
2. Feldmann D
3. Parsy C
4. Freitag S
5. Sinico M
6. Koval C
7. Grati M
8. Couderc R
9. Denoyelle F
10. Bodemer C
11. Marlin S
12. Hadj-Rabia S
(2008) A familial case of Keratitis-Ichthyosis-Deafness (KID) syndrome with the GJB2 mutation G45E
European Journal of Medical Genetics 51:35–43.

https://doi.org/10.1016/j.ejmg.2007.09.005
- Google Scholar
(2000) Connexin 26: required for normal auditory function
Brain Research. Brain Research Reviews 32:184–188.

https://doi.org/10.1016/S0165-0173(99)00080-6
- Google Scholar
(2010) A novel mutation in the connexin 26 gene (GJB2) in a child with clinical and histological features of keratitis-ichthyosis-deafness (KID) syndrome
Clinical and Experimental Dermatology 36:142–148.

https://doi.org/10.1111/j.1365-2230.2010.03936.x
- Google Scholar
1. Meigh L
2. Greenhalgh SA
3. Rodgers TL
4. Cann MJ
5. Roper DI
6. Dale N
(2013) CO₂ directly modulates connexin 26 by formation of carbamate bridges between subunits
eLife 2:e01213.

https://doi.org/10.7554/eLife.01213
- Google Scholar
1. Mhaske PV
2. Levit NA
3. Li L
4. Wang HZ
5. Lee JR
6. Shuja Z
7. Brink PR
8. White TW
(2013) The human Cx26-D50A and Cx26-A88V mutations causing keratitis-ichthyosis-deafness syndrome display increased hemichannel activity
American Journal of Physiology. Cell Physiology 304:C1150–C1158.

https://doi.org/10.1152/ajpcell.00374.2012
- Google Scholar
1. Sanchez HA
2. Mese G
3. Srinivas M
4. White TW
5. Verselis VK
(2010) Differentially altered Ca²⁺ regulation and Ca²⁺ permeability in Cx26 hemichannels formed by the A40V and G45E mutations that cause keratitis ichthyosis deafness syndrome
The Journal of General Physiology 136:47–62.

https://doi.org/10.1085/jgp.201010433
- Google Scholar
1. Sbidian E
2. Feldmann D
3. Bengoa J
4. Fraitag S
5. Abadie V
6. de Prost Y
7. Bodemer C
8. Hadj-Rabia S
(2010) Germline mosaicism in keratitis-ichthyosis-deafness syndrome: pre-natal diagnosis in a familial lethal form
Clinical Genetics 77:587–592.

https://doi.org/10.1111/j.1399-0004.2009.01339.x
- Google Scholar
(2004) Connexins: functions without junctions
Current Opinion in Cell Biology 16:507–512.

https://doi.org/10.1016/j.ceb.2004.07.014
- Google Scholar
1. Wang N
2. De Bock M
3. Decrock E
4. Bol M
5. Gadicherla A
6. Vinken M
7. Rogiers V
8. Bukauskas FF
9. Bultynck G
10. Leybaert L
(2013) Paracrine signaling through plasma membrane hemichannels
Biochimica et Biophysica Acta 1828:35–50.

https://doi.org/10.1016/j.bbamem.2012.07.002
- Google Scholar
(2012) Regulation of ventral surface CO₂/H⁺-sensitive neurons by purinergic signalling
The Journal of Physiology 590:2137–2150.

https://doi.org/10.1113/jphysiol.2012.229666
- Google Scholar
1. Xu J
2. Nicholson BJ
(2013) The role of connexins in ear and skin physiology - functional insights from disease-associated mutations
Biochimica et Biophysica Acta 1828:167–178.

https://doi.org/10.1016/j.bbamem.2012.06.024
- Google Scholar

Article and author information

Author details

Louise Meigh

School of Life Sciences, University of Warwick, Coventry, United Kingdom

Contribution
LM, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

Competing interests
The authors declare that no competing interests exist.
Naveed Hussain

Division of Neonatal Pediatrics, Connecticut Children's Medical Center NICU, University of Connecticut Health Center, Farmington, United States

Contribution
NH, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

Competing interests
The authors declare that no competing interests exist.
Daniel K Mulkey

Department of Physiology and Neurobiology, University of Connecticut, Storrs, United States

Contribution
DKM, Conception and design, Analysis and interpretation of data, Drafting or revising the article

Competing interests
The authors declare that no competing interests exist.
Nicholas Dale

School of Life Sciences, University of Warwick, Coventry, United Kingdom

Contribution
ND, Conception and design, Analysis and interpretation of data, Drafting or revising the article

For correspondence
n.e.dale@warwick.ac.uk

Competing interests
The authors declare that no competing interests exist.

Funding

Medical Research Council (G1001259)

Nicholas Dale

The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Ethics

Human subjects: The Institutional Review Board of the Connecticut Children's Medical Center considered this under the category of a case report and thus exempt from formal review.

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.