Structure And Function Of The Uterus Biology Essay

http://www.pycomall.com/images/P1/St_Georges_University_of_London_Logo_Vector_Format.jpg

The Putative role of KCNQ-encoded Voltage-gated potassium channels (Kv7)

in isoprenaline-mediated relaxations of the mouse uterus

Joshua A. Fasuyi

Supervisor: Dr Iain Greenwood

Division of Biomedical Sciences

BSc Biomedical Science

Submitted 04/13

Acknowledgements

I would like to firstly extend my deepest thanks to Dr Iain Greenwood who provided me with invaluable advice, support and guidance every step of the way, through which I have been able to complete this dissertation. As well as providing me with academic support, his commitment to developing me as an individual by encouraging me to participate in presentations, has also allowed me to learn and cultivate vital skills which I will be able to apply outside of the lab. His humour, positive energy and words of encouragement have made my time in the lab truly memorable.

I would also like to express my sincerest appreciations to Harry, Tom and Jen for their enduring guidance, without which I would never have been able to carry out my project. Their day-to-day support was invaluable, from providing me with the animal resources required to conduct experiments to answering any questions that I had. Right from the very beginning, Tom and Harry played a huge role in settling me in to the lab and I cannot begin to thank them enough for their unwavering patience and guidance during this time. Despite having their own experiments and assignments to write, they consistently ensured that I was heading in the right direction and provided me with ample support for which I am truly grateful. Furthermore, I would like to personally thank Jen for her help on countless occasions and also for her humour, which almost guaranteed a pleasant day in the lab.

Finally, I would like to say a special thanks to my peers in the lab for making my experience memorable. It has been a pleasure to get to know each of them and I hope the friendship we have cultivated develops all the more.

Table of Contents

Abstract

Background: Voltage-gated potassium channels encoded by the KCNQ gene (Kv7) are pivotal regulators of contractility in a variety of tissue. Early research exhibited expression of Kv7 channels in cardiac and neuronal tissue, however recent studies have identified Kv7 channels in smooth muscle, in the vasculature as well as within other organs where smooth muscle is physiologically relevant. Kv7 channels in the smooth muscle of the uterus are functionally relevant in regulating uterine spontaneous activity. Novel research conducted on the renal arteries of rats, demonstrated Kv7 channels to mediate smooth muscle relaxation through β-adrenoceptor activation, however these findings have not been replicated in any other tissue.

Aims & Objectives: The main objective of this study is to ascertain whether Kv7 channels contribute to β-adrenoceptor-mediated relaxations in the mouse uterus.

Hypothesis: β-adrenoceptor activation by isoprenaline mediates relaxation of uterine contractions through Kv7 channels.

Methods: Female non-pregnant BALB/c mice (aged 6-8 weeks) were killed by cervical dislocation, in accordance with the UK Animals (Scientific Procedures) Act 1986. A cervical smear was performed to assess and record the respective stage of the oestrous cycle of each mouse and uterine tissue was harvested and mounted in a myograph for isometric tension recording. Experiments were conducted using Kv7 activators and blockers to investigate the objectives of the current study.

Results: Kv7 blockade attenuated mean relaxant responses induced by 3nM isoprenaline application in spontaneous and oxytocin-driven contractions. As an additional arm of the study, blockade of KCNH-encoded channels (ERG channels) also attenuated isoprenaline responses. In other studies, activation of Kv7 channels produced relaxation of oxytocin-driven contractions. Furthermore, isoprenaline responses were diminished at the oestrous stage of the cycle.

Conclusion: The current study is the first to suggest that Kv7 channels, and perhaps to a greater extent ERG channels, contribute to β-adrenoceptor stimulated relaxation in the uterus.

Introduction

Structure and function of the uterus

The uterus is a complex and dynamic reproductive organ located deep in the female pelvis between the bladder and rectum. The uterus transforms during the menstrual cycle, in which hormonal changes result in enormous physiological changes which prepare the uterus for pregnancy. Over this period, which last approximately 28 days in humans and 4-5 days in mice, ova are released from the ovaries and are transported through the uterine tubes to the uterus. The walls of the uterus are arranged in three distinct layers surrounding a hollow cavity (Fig. 1).

Endometrium

Fundus

Uterus

Uterine tube

Myometrium

Perimetrium

Vagina

Cervix

Ovary

Ampulla

Fimbria

Ovary

Oviduct

Uterine horn

Lumen

Uterine corpus

Cervix

Vaginal fornix

Vagina

A.

B.

Fig 1. Schematic diagram illustrating the anatomy of the uterus. A. Structure of the human uterus. Three layers can be identified from the innermost to the outermost: the endometrium, myometrium and the perimetrium B. Structure of the murine uterus. The uterus is comprised of two lateral horns that join distally at the uterine corpus. A. Adapted from the ‘Medical terms’ website; B. Obtained from Green (2007)

The endometrium is the innermost layer and comprises of both epithelial and glandular cells, which undergo a great deal of change to support the implantation of potentially fertilised ova (Hamilton-Fairley, 2009). In the event where fertilisation does not occur in humans, this thick layer sheds resulting in bleeding, a process known as menstruation.

The outermost layer of the uterus is known as the perimetrium and is comprised of a serous membrane which supports the uterus (Blackburn, 2013).

The myometrium is an intermediate layer of the uterus consisting of smooth muscle cells (SMC). SMCs line the lumen of a variety of hollow viscera where they are functionally significant in permitting compliance of the walls of the organ when in their relaxed state, while also being able to contract to reduce the diameter of the lumen (Rang et al., 2012). While, it is well established that the myometrium is spontaneously active and SMCs are able to contract independent of neuronal or hormonal input, the identity of cells which constitute pacemaker potential as well as the exact mechanisms behind the myogenic nature of the uterus still remains to be elucidated (Wray et al., 2001; Widmaier et al., 2008). Nevertheless, it has been recently proposed by Berridge (2008) that the excitation of the myometrial SMC membrane is driven by an endogenous membrane oscillator, which is a system comprising of several ion channels, pumps and exchangers which interact to elicit either membrane depolarisation or hyperpolarisation.

Depolarisation occurs when the electrical activity generated by the membrane oscillator reaches threshold potential, resulting in the opening of L-type voltage-dependent calcium channels and a subsequent influx of extracellular calcium (Ca2+) into the cytosol of the SMC (Berridge, 2008). This in turn produces an elevation in the concentration of intracellular Ca2+ ([Ca2+]i), which is a pivotal event in the initiation of contractions within smooth muscle.

As well as an extracellular Ca2+ influx, release of Ca2+ from the sarcoplasmic reticulum (SR) can produce a rise in [Ca2+]i. This can occur by two mechanisms. The first involves the activation of ryanodine receptors (expressed on the SR plasma membrane) by intracellular Ca2+ itself, leading to Ca2+-induced Ca2+ release from intracellular stores (Wray et al., 2001). The second mechanism occurs through the activation and opening of inositol 1,4,5-triphosphate (IP3) receptors, a second Ca2+ release channel, which mediates the sequestration of Ca2+ from the SR into the cytosol. IP3 is generated by the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) by phospholipase Cβ (PLCβ), an enzyme which is activated when ligands such as peptide hormones bind to cell surface G protein-coupled receptors (GPCR), specifically those coupled to Gαq/11 (Walsh, 2011).

While these two mechanisms are essential for producing an elevation of [Ca2+]i, a series of enzymatic steps need to occur in order to stimulate the contractile machinery that permit SMC contraction (Fig. 2). Calmodulin is Ca2+-binding protein which, after forming a complex with the free levels of Ca2+, undergoes a conformational change causing it to target myosin light chain kinase (MLCK) and thereby activate it. Activated MLCK then itself goes onto phosphorylate the two 20-kDa regulatory light chain regions in the neck region of the myosin II motor protein, specifically at serine 19. Phosphorylation of this amino acid promotes cross-bridge formation between the head of the myosin II protein and proximal actin filaments. Thus equilibrium between cross-bridge formation and detachment promotes movement of myosin II along the actin filament, which is catalysed by the energy derived from the hydrolysis of adenosine 5’-triphosphate (ATP). This progressive movement shortens the SMC and consequently results in contraction (Walsh, 2011).

Fig 2. Schematic illustrating the Gαq mediated G protein-coupled receptor (GPCR) signal transduction pathway leading to smooth muscle cell contraction. GPCRs are coupled to trimeric G proteins which are comprised of α, β and γ subunits. The Gαq subunit is a subfamily of Gα proteins which activates phospholipase Cβ (PLCβ) located in the cell membrane. Once activated this enzyme catalyses the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) into inositol 1,4,5-triphosphate (IP3). The interaction of this ligand with IP3 receptors on the sarcoplasmic reticulum (SR) membrane leads to a sequestration of calcium (Ca2+) into the cytosol, which rapidly binds to calmodulin (CaM). The Ca2+-CaM complex then in turn activates myosin light chain kinase (MLCK), which itself permits cross-bridge formation between myosin II and actin by phosphorylation of serine at position 19 to form phosphoserine (p-S19). Hydrolysis of adenosine 5’-triphosphate (ATP) provides the energy for movement of myosin along the actin filament to produce contraction. Obtained from Walsh (2011)

With an increase in [Ca2+]i resulting in the contraction of uterine SMCs, it is therefore logical to see why a reduction in [Ca2+]i, by various calcium extrusion mechanisms (such as the Ca2+-ATPase pump) as well as inactivation mechanisms targeting myosin II (mediated by myosin phosphatase (MLCP)), lead to periods of intermittent relaxation (Rang et al., 2012; Blackburn, 2013). It is also worth noting that SMC relaxation can also occur downstream of GPCR activation through pathways coupled to Gαs, however these mechanisms will be discussed in greater detail later.

Like the endometrium, the myometrium undergoes rapid structural modifications over the course of the menstrual cycle, but is also capable of vast expansion during pregnancy to support the growing foetus as well as initiating strong contractions to expel the baby during labour (Blackburn, 2013).

An overview of Potassium channels

As illustrated previously in Fig. 2, Ca2+ influx is central to contractility of smooth muscle. K+ channels regulate membrane potential and therefore control Ca2+ entry into the cell. In the uterus, contractility is largely governed by the activity of voltage-gated K+ channels, a subgroup of K+ channels containing 6 transmembrane domain subunits.

The nomenclature of voltage-gated K+ channels

Functional voltage-gated K+ channels (Kv) are formed from 4 α-subunits (which assemble to make a tetramer) each containing 6 transmembrane domains (S1-S6) with a single pore loop located between S5 and S6. 12 members have been identified in this subfamily (Kv1-12). Central to the function of these channels is the aqueous pore which is comprised of highly conserved signature motifs that selectively permit the flow of K+ ions (Shieh et al., 2000). Another key feature of Kv channels is that they are in a constant state of flux between their active and inactive state which is controlled by changes in membrane potential detected by a voltage sensor region in the S4 transmembrane domain. These channels are activated by membrane depolarisation which allows the flow of K+ out of the cell therefore maintaining negative membrane potential and transiently inhibiting excitability. Kv channel tetramers can be formed by the multimerisation of homogeneous α-subunits to form homotetramers or a product of the assembly of heterogeneous α-subunits within one family to form heterotetramers. Additionally, the association of α-subunits with auxiliary subunits (β-subunits) generates further diversity and also plays a significant role in the modulation of Kv channel activity (Shieh et al., 2000; Gutman et al. 2005).

An Introduction to KCNQ-encoded K+channels

Kv channels encoded by the KCNQ genes (Kv7) are an outward-rectifying subfamily of Kv channel, comprising of 5 members (Kv7.1-7.5). Kv7 channels are responsible for controlling membrane excitability in various cell types where mutation of the KCNQ genes has been found to underlie several pathologies (Lerche et al. 2001; Robbins, 2001; Greenwood & Ohya, 2009).

KCNQ1

The first of the Kv7 channels to be identified was the protein encoded by the KCNQ1 gene (Kv7.1). Kv7.1 channels are formed when KCNQ1 encoded α-subunits homomultimerise, however it is well established that these α-subunits can also co-assemble with KCNE-encoded auxiliary subunits to generate heteromeric channels (Fig. 3). The KCNE1-5 gene encodes auxiliary subunits which interact with KCNQ1 encoded subunits to modulate the behaviour and pharmacology of the functional channel (Shieh et al., 2000). This is evident in the human heart, where KCNQ1 gene products have been found to be co-expressed with KCNE1-3 encoded subunits with the various combinations boasting different electrophysiological characteristics (Lundby et al., 2010). Furthermore, research conducted by Bendahhou et al. (2005) on KCNQ1 and KNCE gene family interactions has also suggested the involvement of KCNE4 and KCNE5 encoded subunits in altering the activity of Kv7.1 in the heart, again emphasising the importance of these auxiliary proteins.

Co-expression with KCNE1 subunit in the heart forms functional Kv7.1 channels which are slower to activate and also exhibit higher current amplitude compared to the homomeric Kv7.1 channel (Robbins, 2001). This current is known as the slow delayed rectifier current (IKs) and is functionally significant in controlling repolarisation of the ventricles of the heart (Wulff et al., 2009). As a result, defects in either KCNQ1 or KCNE1-encoded subunits due to mutation can result in a delayed repolarisation of the heart and prolonged cardiac action potential, presenting clinically as ventricular arrhythmia in a condition known as Long QT syndrome (Wang et al., 1996; Ashcroft, 2000; Shieh et al., 2000; Robbins, 2001).

In addition to their role in the heart, Kv7.1 channels have also been found to be major players in physiological processes such as gastric acid secretion and sodium ion (Na+) absorption across renal and intestinal epithelium (Robbins, 2001; Wulff et al., 2009).

KCNQ1 encoded subunit

KCNE encoded subunit

Full-size image (73 K)

Fig 3. Illustration of the structure of KCNQ1 and KCNE encoded subunits. KCNQ1 α-subunits contain 6 transmembrane domains (S1-S6) where S1-S4 encode the voltage sensing domain (VSD) and S5-S6 encode the pore domain (PD). While Kv7.1 α-subunits can homomultimerise to form functional channels they can also interact with membrane spanning β-subunits encoded by KCNE genes (KCNE1-5) which can modulate the activity of the channel. Adapted from Lundby et al. (2010)

KCNQ2/KCNQ3 heterotetramer

KCNQ2 and KCNQ3-encoded subunits homomultimerise to form Kv7.2 and Kv7.3 channels respectively. However, as demonstrated by Yang et al. (1998), KCNQ2 and KCNQ3 subunits can also heteromutimerise to form a functional tetrameric channel, capable of producing a K+ current 10 fold greater than that produced by its homomeric constituents. These heteromeric channels were first identified in peripheral sympathetic neurones as well as the central nervous system where they were demonstrated to modulate synaptic plasticity and the excitability of neurones, through the generation of a non-inactivating neuronal K+ current referred to as the M-current (IKM) (Brown & Adams, 1980).

KCNQ2/KCNQ3 encoded heteromeric channels (M-channels) supress neuronal excitability by hyperpolarising the neuronal cell membrane thereby reducing the frequency of action potential firing, a mechanism of which can be counteracted by Gαq coupled muscarinic GPCR activation by agonists such as acetylcholine (Brown & Adams, 1980; Ashcroft, 2000; Wulff et al., 2009). M-channels require PIP2 to open, hence degradation of PIP2 by downstream activation of PLCβ increases neuronal excitability.

In conditions where the function of M-channels is compromised, for example as a result of mutation, neuronal hyper-excitability can ensue. Characterised by brief frequent seizures that present within the first 3 days of life and cease by the third month post-partum, Benign Familial Neonatal Convulsion (BFNC) is the hallmark of point mutations in both the KCNQ2 and KCNQ3 genes (Ashcroft, 2000; Maljevic et al., 2008). While the exact mechanism behind BFNC remains to be elucidated, the role of M-channels in its pathogenesis is clear.

KCNQ4

KCNQ4 encoded channels (Kv7.4) are mainly expressed in hair cells in both the inner (type 1) and outer ear, as well as in other areas of the vestibulocochlear system (Kharkovets et al., 2000; Shieh et al., 2000). The physiological action of these homomeric channels is to maintain an outward rectifying K+ current, which may be essential in maintaining the electrophysiology of hair cells (Kharkovets et al., 2000; Robbins, 2000; Trussell, 2000).

Autosomal dominant nonsyndromic deafness (DFNA2) is associated with a variety of point mutations that exert a negative effect on the function of the Kv7.4 channel, resulting in hereditary deafness. These mutations can either cause deletions or substitutions in the KCNQ4 gene sequence which can alter pore structure, the level of channel expression or the overall structure of the channel, rendering it dysfunctional (Shieh et al., 2000; Nie, 2008). Despite the clear association between KCNQ4 mutation and DFNA2, the pathogenesis is largely still unknown.

KCNQ5

The most recently discovered member of the KCNQ encoded K+ channel family is Kv7.5. Encoded by the KCNQ5 gene, studies revealed that these channels are widely expressed in skeletal muscle and sub regions in the brain (Lerche, 2000). KCNQ5 encoded subunits are capable of both homo- and heteromultimerisation. Assembly of KCNQ5 and KCNQ3 gene products forms a heteromeric channel which exhibits 4-5 times higher amplitude currents than the homomeric configuration of the KCNQ5 encoded channel alone and it is also thought that these heteromeric products may form an M-channel similar to the KCNQ2/KCNQ3 heteromultimer (Lerche, 2000; Robbins, 2000). To date, no mutations in the KCNQ5 gene have been associated with any hereditary disorder.

KCNQ encoded K+ channels in smooth muscle

A considerable amount of research has been conducted on Kv7 channel expression in cardiac (Kv7.1) and neuronal tissue (Kv7.2-7.5), however the recent identification of Kv7 channels in smooth muscle provides a new and exciting field of investigation.

Research conducted by Ohya et al. (2003) was the first to reveal the expression of the KCNQ gene in vascular SMCs, where the KCNQ1 encoded channel was shown to be predominantly expressed in the murine portal vein. As in cardiac tissue, KCNQ1 gene products in the murine portal vein are similarly thought to interact with KCNE encoded β-subunits. The question of the function of KCNQ encoded channels was also soon investigated, where a study on the effect of KCNQ modulators in altering the electrophysiology provided evidence that supported the idea that Kv7 channels contribute to controlling the excitability of mouse portal vein SMCs (Yeung & Greenwood, 2005).

Further studies by Yeung et al. (2007) examined whether there was KCNQ gene expression in SMCs within murine arteries. Results from these experiments revealed KCNQ expression in SMCs in the thoracic aorta, carotid artery, femoral artery and mesenteric artery, with KCNQ1 and KCNQ4 being predominantly expressed. Similar experiments carried out in rat pulmonary and mesenteric arteries were also consistent with these findings (Mackie et al., 2008; Joshi et al., 2009).

Most recently, a study conducted by Ng et al. (2011) has ventured outside of the rodent vasculature and has provided pioneering evidence for the expression of Kv7 channels in both human visceral adipose and mesenteric arteries.

As well as exhibiting functionality in the vasculature, Kv7 channels have been found to be physiologically relevant in the smooth muscle of a variety of viscera including the airways, stomach, colon, bladder and the uterus (Jepps et al., 2013).

KCNQ transcripts have been shown to exist in the airways of both guinea pigs and humans where they have been proposed to be involved in the regulation of airway diameter. The expression profile of these channels however differs significantly between these two species. KCNQ1 expression is most abundant in humans and KCNQ2 expression is negligible, whereas the reverse of this applies in guinea pig airway SMC. Another significant finding of this study was that activation of Kv7 channels by modulators such as retigabine (Kv7.2-7.5 activator) inhibited narrowing of the airways, therefore presenting Kv7 channels as a therapeutic target for inflammatory airway diseases such as asthma (Brueggemann et al., 2012).

Determining the role of KCNQ encoded K+ channels within the digestive system has also recently become an area of interest amongst researchers. Regional variances have been observed within the rat stomach, where KCNQ1 expression (and assembly with KCNE1 subunits) is high in the antrum (distal stomach) but negligible in the fundus (proximal stomach) (Ohya et al. 2002). Conversely, high levels of Kv7.4 and Kv7.5 expression have been seen in SMCs in the rat fundus (Ipavec et al., 2011). This is very much the same for SMCs in the muscular layer of the colon in mice, where a similar expression profile exists (Jepps et al., 2009).

Numerous studies have highlighted Kv7 channels as an attractive pharmacological target for the treatment of bladder over-activity. Research by Rode et al. (2010) has shown that bladder over activity is reduced when Kv7 channels are activated by retigabine in animal models. It has also been proposed that K+ currents generated by Kv7 channels inhibit pacemaker cell electrical activity in the bladder, again highlighting their role in regulating resting membrane potential (Anderson et al., 2009).

Studies conducted by McCallum et al. (2009) in the mouse uterus have provided evidence for KCNQ encoded isoforms in the myometrium. Furthermore, investigation into the different stages of the reproductive cycle in mice revealed changes in the expression profile of KCNQ and KCNE genes suggesting a functional contribution to the changes that are observed in the structure and function of the uterus over this cycle. Although the KCNQ1 transcript was observed to be the most abundant in the myometrium throughout the reproductive cycle of non-pregnant mice, later studies demonstrated a switch in the most abundant transcript to KCNQ3 in the early stages of pregnancy (McCallum et al., 2011). This isoform is likely to form a channel which exhibits a different K+ current which is functionally essential to the early stages of gestation. In the same paper, the first in vitro investigations conducted on human myometrium interestingly revealed KCNQ1-4 gene expression. Activators of Kv7 channels (such as retigabine) were observed to attenuate spontaneous uterine contractions in both humans and mice, suggesting that Kv7 channels may be important for regulating uterine contractility. Additionally, the effects of the activators were also shown to be reversed by application of Kv7 channel blockers such as XE-991 highlighting the potential clinical use of Kv7 channel modulators in inducing and prolonging labour.

KCNQ encoded K+ channels and G-protein coupled receptor interactions

G protein-coupled receptors (GPCR) are a family of membrane-spanning metabotropic proteins which interact with intracellular proteins, called G-proteins, to stimulate effector systems and therefore elicit various cellular responses (Rang et al., 2012).

Adrenoceptors are a family of GPCRs that were first identified to respond to adrenaline and noradrenaline (a systemic hormone and neurotransmitter respectively) but are now known to be activated by a range of chemical mediators. It is now generally accepted that adrenoceptors fall into two categories: α- or β-adrenoceptors, which are subdivided into 2 α-receptor subtypes (α1 and α2) and 3 β-receptor subtypes (β1, β2 and β3).

SMCs express β-adrenoceptors in a variety of organs and are responsible for causing relaxation of these cells via several signalling events. Activation of these receptors results in the formation of the active Gαs subunit which leads to the stimulation and increased activity of adenylate cyclase (AC). AC is a cytoplasmic enzyme which catalyses the conversion of ATP into cyclic adenosine monophosphate (cAMP), an important secondary messenger that can activate a family of proteins called protein kinases. In SMCs, cAMP-dependent protein kinase A stimulation results in relaxation due to inactivation of enzymes responsible for contraction and activation of proteins required for relaxation (Rang et al., 2012).

While increasing amounts of evidence have provided support for the idea that Kv7 channels regulate SMC excitability and thus contractility, it has not yet been discerned whether Kv7 channels are involved in the physiological mechanisms, such as β-adrenoceptor stimulation, that are known to produce relaxation.

Recently, studies by Chadha et al. (2012a) have demonstrated that Kv7 channels contribute to β-adrenoceptor-mediated relaxation in the renal arteries of normotensive (NT) rats. Application of isoprenaline - a synthetic, non-selective β-adrenoceptor agonist - to precontracted NT renal arteries produced a concentration-dependent relaxation which correlated with increases in Kv7 currents. Furthermore, it was found that isoprenaline-induced vasorelaxations were significantly reduced in KCNQ4 knockdown mice where approximately 60% of Kv7.4 channel expression was inhibited. Experiments conducted on the renal arteries of spontaneous hypertensive rats (SHR) also revealed a reduced expression of KCNQ4 and failed to respond to increasing isoprenaline concentrations (Chadha et al., 2012a). These findings appear increasingly consistent with the idea that Kv7 channels contribute to β-adrenoceptor-mediated relaxations in NT rat renal arteries and also suggest a potential causative link between diminished Kv7 responses and the pathogenesis behind essential hypertension. Despite these interesting findings, it is yet to be investigated if Kv7 channels are involved in β-adrenoceptor-mediated relaxations of smooth muscle in other tissue.

Aims of the present study

Summary

As previously mentioned, there is evidence for the expression of KCNQ encoded K+ channels in the myometrium (McCallum et al., 2009) as well as evidence for the involvement of Kv7 channels in isoprenaline-induced relaxation in vascular smooth muscle (Chadha et al., 2012a).

Hypothesis

Fig. 4 illustrates the hypothesis that Kv7 channels contribute to isoprenaline-mediated relaxations of the mouse uterus through cAMP-dependent processes. If this theory holds true then it can be expected that application of Kv7 channel modulators should alter uterine responses to isoprenaline.

Aims & objectives

The main objective of this study is to ascertain whether Kv7 channels contribute to β-adrenoceptor-mediated relaxations in the mouse uterus. The effects of blockers and activators of Kv7 channels on isoprenaline-induced relaxations will be investigated in both spontaneous and oxytocin-driven uterine contractions. Oxytocin is a hormone which increases the amplitude and frequency of uterine contractions through GPCR pathways linked to Gαq (Rang et al., 2012). The uterus is increasingly sensitive to oxytocin prior to labour so application of this drug may reveal changes in the isoprenaline-induced relaxations under these conditions. Furthermore, the effects of selective Kv7 modulators will also be studied to determine the impact of specific Kv7 channels in the uterus.

Finally, following research by Greenwood et al. (2009) which identified isoforms of murine KCNH (Kv11) gene transcripts in the myometrium, the effects of modulators of the channels encoded by this gene (ERG channels) on isoprenaline-mediated relaxations will also be examined.

Fig 4. Schematic illustrating the proposed role of Kv7 channels in isoprenaline-mediated relaxations in the uterus. Isoprenaline non-selectively activates β-adrenoceptors which leads to the production of cAMP by adenylate cyclase (AC). cAMP-dependent PKA, or cAMP directly, activates Kv7 channels resulting in an efflux of K+ which hyperpolarises the cell membrane therefore reducing the probability of voltage-gated Ca2+ channels opening. Consequently, intracellular [Ca2+] does not rise sufficiently to cause contraction therefore shifting the balance towards SMC relaxation

Materials and Methods

Tissue Collection and Preparation

Female non-pregnant BALB/c mice (aged 6-8 weeks) were killed by cervical dislocation, in accordance with the UK Animals (Scientific Procedures) Act 1986. A cervical smear was performed after the mouse had been killed to determine the stage of the oestrous cycle by gently pipetting 150µl of Krebs solution in and out of the vagina to form a suspension containing cells from the cervical mucus. A volume of 50µl of this suspension was applied to a slide and observed by light microscopy (Motic AE21) at a magnification of x400. The presence and abundance of cornified squamous epithelial cells was then noted to determine the specific stage of the reproductive (oestrus) cycle, as exemplified in Fig. 5.

G:\tileshop.fcgi.jpg

Fig 5. Photomicrograph of unstained vaginal secretion from mice in a. Pro-oestrus b. oestrus c. metoestrus d. Dioestrus stage of the cycle. Three main cell types are present including nucleated epithelial cells (N), leukocytes (L) and cornified squamous epithelial cells (C). Obtained from Caligioni (2009)

After performing the cervical smear, the uterus was dissected out of the mouse. Surrounding fat and blood vessels were removed and 2 segments from the upper part of each uterine horn were cut transversely at an approximate length of 2mm. Segments were specifically taken from the upper parts of the uterus due to evidence of this region producing higher myogenic activity compared to lower regions (Griffiths et al., 2006).

Isometric tension studies

The uterine segments were mounted in a myograph (Danish Myo Technology, Aarhus, Denmark) as small ring preparations on micro-pins for isometric tension recording (Fig. 6). This is an ex vivo technique which allows the changes in tension of a tissue to be measured (millinewtons, mN). This tension is detected by a force transducer which relays this data to be processed on a computer.

Force transducer

Micro-pins

Micrometer

Fig 6. Schematic diagram illustrating the myograph system used in the investigation. To generate spontaneous uterine contractility, the micro-pins at each side are moved apart using the micrometer. Tension generated by the contractile activity of the tissue pulls at the micro-pin connected to the force transducer allowing it to detect changes in tension. A constant oxygen supply to the tissues ensures that they are optimally aerated. Adapted from the Danish Myo Technology website.

A small amount of tension was applied to stretch the tissue to its approximate in situ length to encourage spontaneous contractions. The tissue preparations were maintained in a 5ml bath containing Krebs solution (Table 2) aerated with 95% O2/ 5% CO2 at a temperature of 37°C and at a pH of 7.4 to provide optimal physiological conditions. All preparations were left to equilibrate (~1-2 hours) until regular spontaneous contractions had developed.

The majority of the experiments were carried out in preparations that were spontaneously contracting however in some experiments the tissue preparations received oxytocin (100nM) in order to determine the effect of agents in oxytocin-driven uterine contractions.

Experimental Protocols

After allowing ~ 1 hour for regular spontaneous contractions to develop (Fig. 7), isoprenaline was administered at a concentration of 1, 10 and 100nM to three of the four uterine preparations, with the fourth preparation receiving distilled water as a control for isoprenaline.

Similar experiments were performed in uterine preparations where contractions were driven with application of oxytocin. Following observation of rhythmic myogenic contractions, 100nM oxytocin was administered to all four tissue preparations. After a period of ~ 5 minutes, 1nM, 10nM and 100nM isoprenaline was administered to three of the four uterine preparations, with the fourth preparation again receiving distilled water as a control.

Other experiments were carried out to determine the effect of isoprenaline on both spontaneous and oxytocin-driven uterine contractions using various selective and non-selective Kv7 channel blockers (Table 1). After regular spontaneous contractions had developed, a single concentration of 3nM isoprenaline was applied to segments of the uterus, which had been previously treated with either 10µM of a non-selective Kv7 channel blocker (linopirdine or XE991) or dimethyl sulfoxide (DMSO) for 10 minutes. Vehicle control preparations received distilled water as a control for isoprenaline. The same procedure was carried out on oxytocin-driven contractions.

The effect of 3nM isoprenaline on both myogenic and oxytocin-driven uterine contractions was observed in the presence of selective Kv7.1 channel blockers JNJ-538 or chromanol-239b (10µM and 3µM respectively) as well as a KCNH2 encoded (ERG) channel blocker, E4031 (10µM).

Following a period of regular myogenic contractions, 1, 3, 10 and 30µM RL-3 were applied to uterine segments in a concentration-dependent manner to observe the effects of the Kv7.1 channel activator on oxytocin-driven (100nM oxytocin) contractions. A similar protocol was used in a separate investigation to observe the effects of the same concentrations of S-1 (Kv7.2-7.5 activator) on oxytocin-driven uterine contractions. Additional experiments were conducted in the presence of 10µM XE991, where the relaxant effects of 10µM RL-3 and S-1 were observed.

Control experiments for all Kv7 and ERG channel modulators were conducted in the presence of an equivalent volume of DMSO.

After 1 hour +

Equilibriation period

Start of experiment

Fig 7. Schematic illustrating the nature of the contractions in the equilibration period prior to the start of experiments. A period of approximately 1 hour is allowed for contractions to become regular so as to ensure optimal contractions before beginning each protocol

Materials and solutions

Table 1. List of materials showing their respective mechanisms of action, concentrations used in the investigation and the manufacturing company from which they were obtained. *Materials not publically available

Table 2. List of solutions showing their chemical composition and purpose within the investigation

Data and Statistical Analysis

All upward deflections signify contractions and the tension generated by contractions were recorded in millinewtons (mN).In order to accurately analyse the contractile activity generated by the uterine preparations, the maximum amplitude of contraction, duration of contraction and the mean integral of tension (MIT) from the minimum were measured (Fig. 8). This was achieved using the lab chart 7 reader software (ADInstruments LTD, Oxford, UK). Statistical analysis of the data was conducted to obtain graphs using GraphPad Prism 6.0 (GraphPad Software Inc, San Diego, USA). Unpaired t-tests and two way analysis of variance (ANOVA) tests (followed by Bonferri post-hoc correlation) were used in the analysis of the data collected and a p value of less than 5% (p <0.05) was considered statistically significant. Data values are expressed as the mean ± standard error of mean (S.E.M) for the corresponding number (n) of mice from which uterine segments were obtained.

A

B

C

Fig 8. Schematic diagram illustrating the parameters that were measured in this investigation. A.Maximum amplitude of contraction (mN) B. Duration of contraction at 10% maximum amplitude of contraction (seconds) C. Mean integral of tension from minimum (mN), which measures the area under the curve shaded in blue

Results

Spontaneous uterine contractions

It is important to appreciate that the myogenic nature of the uterus brings with it a great deal of variability. In the present study, 116 segments from 29 mice were mounted for isometric tension recording and were found to exhibit spontaneous contractions that displayed different contractile patterns. Out of these segments, 43 displayed transient smooth type contractile patterns, as seen in Fig. 9A, whereas 70 segments produced contractions with a spiked appearance, as demonstrated in Fig. 9B. Furthermore, 3 segments produced contractions that exhibited both contractile patterns.

A.

B.

Fig 9. A representive trace demonstrating myogenic uterine contractions with A. a transient smooth appearance and B. a characteristic spiked appearance

Effect of oxytocin on murine spontaneous uterine contractions

Oxytocin (OxyT) is a peptide hormone released from the posterior pituitary gland (Rang et al., 2012). Application of 100nM OxyT produced a 212.0% (n=3) increase in contractile activity (MIT), as indicated by an increase in the contractile frequency and magnitude of individual contractions as well as a rise in the baseline of contractions (Fig. 10).

Fig 10. A representive trace illustrating the effect on myogenic uterine contractions after application of 100nM oxytocin

Effect of isoprenaline on spontaneous and oxytocin-driven contractions

Application of isoprenaline (1nM to 100nM) to spontaneous and oxytocin-driven contractions resulted in a concentration-dependent relaxation (Fig. 11A). The relaxant response displayed an initial and latter phase characterised by an initial inhibition of contractions which waned in the continued presence of the drug at lower concentrations. Hence, 3 minutes after 10nM isoprenaline application, there was still contractile activity. Administration of 1nM isoprenaline (IsoP) produced a greater relaxation of uterine contractions driven by OxyT with the mean % relaxation (MIT) reaching 39.7 ± 11.4 (n=10) compared to 6.6 ± 14.4% (n=7) in spontaneous conditions (Fig. 11B). The concentration producing half the maximal % relaxation (EC50) in OxyT-driven contractions was 1.9nM.

A.

B.

Fig 11. Effect of isoprenaline on spontaneous and 100nM oxytocin-driven contractions A. Trace illustrating the relative effect of 1-100nM isoprenaline on spontaneous contractions B. Concentration-effect curve for isoprenaline illustrating the % relaxation (MIT) on spontaneous (n=5-7) and 100nM oxytocin-driven contractions (n=8-10). Data expressed as mean ± S.E.M

Effect of selective and non-selective Kv7 and ERG channel blockers on spontaneous uterine contractions

Various selective and non-selective Kv7 channel and ERG channel blockers were administered to uterine segments to firstly evaluate their effects on spontaneously active uterine contractions. As demonstrated in Fig. 12, incubation with 10μM XE-991, a non-selective Kv7 channel blocker, demonstrated an increase in uterine contractility by 36.7 ±13.4% compared 0.6 ± 5.7% (n=12) in DMSO controls. This was due to an increase in the frequency of contractions rather than an increase in the individual magnitude of contractions. Conversely, treatment with 3μM chromanol 293-b, a selective Kv7.1 blocker, reduced MIT by 5.4 ± 6.0%.

As an additional arm of the investigation the effect of 10μM E4031, an ERG channel blocker, was observed. Greenwood et al. (2009) recently demonstrated that ERG channel inhibitors produce huge stimulatory effects in the non-pregnant myometrium compared to vehicle-treated preparations. Moreover in the present study, it was found that application of E4031 produced an increase in contractility by 49.4 ± 24.2%, which was due to a change in the frequency of contractions rather than an increase in the magnitude of individual contractions (Fig. 12).

E4031 demonstrated additional effects on spontaneous uterine contractions as seen by the generation of much broader contractions as well as a corresponding increase in the duration of contraction, measured in seconds (s) (Fig. 13A). On average, the duration of contractions increased by 10.1 ± 2.8 seconds following application of 10μM E4031 whereas in control-treated uterine segments the duration of contraction reduced by a mean average of 2.6 ± 1.2 seconds (Fig. 13B). Furthermore compared to control responses (Fig. 13C), application of 100nM OxyT onto uterine segments pre-incubated with 10μM E4031 produced a sustained contraction which eventually recovered to continue generating broad contractions (Fig. 13D).

Fig 12. Effect of Kv7 and ERG channel blockers on spontaneous uterine contractions. Application of XE-991 (n=15) and E4031 (n=9) produced a significant (p<0.05) increase in uterine contractility whereas chromanol-239b (n=9) resulted in an inhibition of contractions compared to control values. Values were obtained by measurement of MIT and the degree of significance corresponds to the number of asterisks, represented as the following: p<0.05 (*); p<0.01 (**); p<0.001 (***). Data expressed as mean ± S.E.M

A.

B.

C.

D.

Fig 13. Effects of E4031 A. Representive trace illustrating the characteristic broadening effect of 10μM E4031 on spontaneous uterine contractions B. Histogram comparing the effects of 10μM E4031 (n=9) on the duration of contraction to control (n=8), as indicated by measurement in seconds (s) (p<0.01).Measurements were taken 1 minute after application of the respective agent, for a period of 3 minutes. Data expressed as mean ± S.E.M. C. Trace illustrating the response to 100nM Oxytocin in uterine segments pre-incubated with DMSO D. Trace illustrating the sustained contraction associated with application of 100nM oxytocin to 10μM E4031 pre-incubated uterine segments

Effect of Kv7 and ERG channel blockers on 100nM oxytocin responses

Experiments were carried out to investigate whether the OxyT response itself was altered in the presence of these modulators (Fig. 14). A stimulatory response to 100nM OxyT was seen in DMSO-treated segments, increasing the MIT by 148.7 ± 31.0% (n=9). Pre-incubation with 10μM XE-991 and 3μM chromanol-293b potentiated the OxyT response, increasing contractility by 246.8 ± 75.3% and 240.3 ± 127.9% respectively. Moreover, application of 10μM E4031 fused contractions to produce a sustained contraction which consequently provoked a statistically significant (p<0.01) attenuation of the stimulatory OxyT response, exhibiting a 41.4 ± 19.2% increase in the MIT.

Fig 14. Comparative histogram illustrating the stimulatory effect of 100nM oxytocin on spontaneous contractions in the presence of 10μM XE-991 (n=10) and E4031 (n=9) and 3μM chromanol-293b (n=5). Application of XE-991 and chromanol-293b potentiated the OxyT response compared to controls. Pre-incubation with 10μM E4031 however significantly attenuated the OxyT response (p<0.01). Data values expressed as mean ± S.E.M

Effect of selective and non-selective Kv7 and ERG channel blockers on 3nM isoprenaline-mediated relaxations of spontaneous and oxytocin uterine contractions

Having established the independent effects of Kv7 and ERG channel blockers, further experiments were carried out to investigate whether application of these modulators altered the response to 3nM isoprenaline. Responses to 3nM isoprenaline applied in the presence of 10μM XE-991 and 10μM E4031 were attenuated relative to tissues pre-incubated with DMSO, although the effect did not reach significance (p<0.05) (Fig. 15A). The mean relaxant response (MIT) in the presence of 10μM XE-991 was 19.8 ± 4.4% (n=8) compared to mean control responses at 22.8 ± 3.8% (n=7). Pre-incubation with 10μM E4031 however, suggested a more profound attenuation producing a mean % relaxation of 13.5 ± 5.7 (n=8).

In order to determine the function of Kv7.1 in isoprenaline-mediated relaxations, 3μM chromanol-293b was administered to uterine preparations preceding application of 3nM isoprenaline and was found to produce a mean % inhibition of contraction of 20.3 ± 7.7 (n=7). Notably, application of these channel modulators had no effect on the relaxant responses of high concentrations of isoprenaline.

Similar to spontaneous contractions, application of 3nM isoprenaline to OxyT –driven contractions pre-incubated with their respective modulators produced a small degree of attenuation of the isoprenaline response; however it is worth noting that % relaxation was considerably greater than those experiments conducted on spontaneously active contractions. Mean responses in preparations treated with XE-991, E4031 and chromanol-293b were 40.4 ± 7.0 (n=10), 10.6 ± 21.8 (n=10) and 47.7 ± 6.3% (n=10) respectively, compared to 57.9 ± 7.9% as observed in control segments (n=10) (Fig. 15B).

A.

B.

Fig 15. Histogram comparing the effect of various selective (chromanol-293b) and non-selective (XE-991) Kv7 and ERG channel (E4031) blockers on 3nM isoprenaline-induced relaxations of the uterus in experiments conducted on A. spontaneous contractions and B. 100nM oxytocin-driven contractions. Data expressed as mean ± S.E.M

Effect of 1mM 4-aminopyridine on spontaneous contractions

Application of 1mM 4-AP (a non-selective Kv channel blocker) to spontaneous uterine contractions provoked a 154.6 ± 48.2% rise in the MIT compared to incubation with DMSO, which increased MIT by only 4.7 ± 5.1% (Fig. 16A). Conversely, application of 1mM 4-AP had no effect on increasing the maximum amplitude of individual contractions, producing only a slight increase in the magnitude of mean contractions by 3.3 ± 2.1%. This was not significantly different from control experiments, which similarly exhibited only a 3.3 ± 2.7% increase in the amplitude of the mean contraction (Fig. 16B).

A.

B.

Fig 16. Effect of 4-AP on spontaneous contractions. A. Application of 1mM 4-AP (n=6) produced an increase in the frequency and amplitude of contractions that are significantly greater than DMSO controls (n=4), as indicated by measurement of MIT (p<0.05) B. Pre-incubation with 1mM 4-AP (n=5) suggests no significant difference in the mean maximum amplitude of contraction when compared to control segments (n=4), as indicated by measurement of the mean amplitude of contraction (p>0.05). Data expressed as mean ± S.E.M

Isoprenaline response in the presence of 1mM 4-aminopyridine

Observations on the effects of pre-incubation with 4-aminopyridine (4-AP) on 3nM isoprenaline-mediated relaxations of 100nM OxyT-driven contractions found that pre-incubation with 1mM 4-AP produced a 53.9 ± 12.3% relaxation (MIT). This reduction in contractility was not significantly different from mean control responses which exhibited a 51.4 ± 10.8% inhibition of contraction (Fig. 17). The purpose of this investigation was to discern whether the suggested trends towards attenuation of the isoprenaline response observed in the presence of the blockers, particularly those targeting Kv7 channels, was due to the specific blockade of Kv7 channels and not because of an increase in contractility. It may simply be the case that the isoprenaline response is attenuated because of increased uterine contractility rather than by blockade of channels that may play a vital role in the isoprenaline response. Hence, the rationale behind these investigations was to see whether application of 4-AP (a non-selective Kv channel blocker) at 1mM (a concentration which is considered to block Kv but not Kv7 channels) would present similar effects to those observed with application of the Kv7 and ERG channel blockers.

Fig 17. Comparative histogram demonstrating the difference between application of 1mM 4-AP (n=6) and DMSO (n=5) on 3nM isoprenaline-induced relaxations of 100nM oxytocin-driven contractions. Relaxations induced in the presence of 4-AP showed no significant difference to control (p>0.05). Data values expressed as the mean ± S.E.M

Effect of selective Kv7.1 activation on uterine contractions

Studies by McCallum et al. (2009) have found KCNQ1 to be the most expressed isoform in the myometrium of female C57/BL6 mice. As well as using chromanol-293b, a selective Kv7.1 blocker, as a means of ascertaining the impact of Kv7.1 blockade on isoprenaline-mediated relaxation, experiments to see whether selective Kv7.1 activation could induce relaxations similar to those seen on application of isoprenaline were also conducted. Hence, the attention of the investigation rapidly turned to the impact of activators of these channels in particular RL-3, a selective Kv7.1 channel activator.

Stepwise application of 1 to 30μM RL-3 produced a concentration-dependent relaxation of 100nM OxyT-driven contractions with an EC50 value of 4.1μM (Fig. 18). Application of 30μM RL-3 exhibited a mean % relaxation (MIT) of 72.4 ± 10.6 whereas treatment with DMSO produced a 14.5 ± 13.0% relaxation.

Fig 18. Concentration-effect curve for RL-3 illustrating % relaxation (MIT) of 100nM OxyT-driven contractions compared to DMSO. Increases in relaxant effects correspond to increases in the concentration of RL-3. A significant relaxant effect was observed at 30μM (n=4) compared to controls (n=4) (p<0.05). Data values expressed as mean ± S.E.M

Effect of selective Kv7.1 activation on uterine contractions, in the presence of 10μM XE-991

It is well known that RL-3 functions by specifically activating Kv7.1 channels. However, to physiologically provide evidence of RL-3 producing relaxant effects directly through Kv7 channels, its effects were investigated in the presence of XE-991. Incubation with 10μM XE-991 attenuated 10μM RL-3 induced relaxations of 100nM OxyT-driven contractions compared to DMSO controls, where the mean relaxant response was 5.8 ± 7.2% and 24.9 ± 10.4% respectively (Fig. 19). These values however failed to reach statistical significance (p>0.05).

Fig 19. Comparative histogram comparing the relaxant effects of 10μM RL-3 in the presence of 10μM XE-991 (n=6) on 100nM OxyT-driven contractions compared to DMSO-treated controls (n=6). Data values expressed as the mean ± S.E.M

Effect of Kv7.2-7.5 activation on uterine contractions

Based on the rationale surrounding experiments conducted with RL-3, similar investigations were carried out using S-1, a specific Kv7.2-7.5 channel activator.

Application of increasing concentrations of S-1 from 1 to 30μM produced a sigmoidal concentration-effect curve exhibiting an EC50 of 5.2μM (Fig. 20). The mean relaxant effects of 10μM and 30μM RL-3 on 100nM OxyT-driven contractions were significantly more pronounced compared to incubation with DMSO, producing a % relaxation of 71.8 ± 7.0 and 87.0 ± 2.3 respectively. Moreover, control responses for 10μM and 30μM RL-3 application were 17.3 ± 15.8 (n=4) and 18.1 ± 22.4% (n=4) respectively.

Fig 20. Concentration-effect curve for S-1 illustrating % relaxation (MIT) of 100nM OxyT-driven contractions compared to DMSO. Increases in relaxant effects correspond to the stepwise increases in the concentration of S-1. A statistically significant difference was observed at 10μM (n=5) (p<0.01) and 30μM (n=5) (p<0.001). Data values expressed as mean ± S.E.M

Effect of Kv7.2-7.5 activation on uterine contractions, in the presence of 10μM XE-991

Expanding upon previous experiments examining the effects of Kv7.2-7.5 activation on uterine contractions, the effect of 10μM S-1 on 100nM OxyT-driven contractions was also observed in the presence of 10μM XE-991 to physiologically show that S-1 targets Kv7 channels in its mechanism of action (Fig. 21). Similar to experiments conducted with RL-3, the mean % relaxation induced by 10μM S-1 was slightly attenuated in the presence of XE-991, producing a 14.6 ± 9.4% relaxation compared to 25.1 ± 13.1% when segments were administered DMSO. This however did not reach statistical significance (p>0.05).

Fig 21. Histogram comparing the relaxant effects of 10μM S-1 on 100nM OxyT-driven contractions in the presence of 10μM XE-991 (n=6) compared to DMSO-treated controls (n=5). Data values expressed as the mean ± S.E.M

Effect of Kv7 blockade on 10μM S-1-induced relaxations of spontaneous uterine contractions

Measurement of MIT 4 minutes after application of 10μM S-1 to spontaneous contractions found contractility to be reduced by 73.3% (n=1) (MIT). The spontaneous activity of the uterus did not recover despite allowing an extended period of time for the tissue to regain contractility. However, approximately 23 minutes after administration of S-1, 10μM XE-991 was applied which appeared to recover contractions producing activity that was 1196.745% of the initial contraction prior to addition of XE-991 (Fig. 22).

Fig 22. Trace illustrating 10μM S-1-induced relaxation of spontaneous contractions and the stimulatory effect of application of 10μM XE-991 approximately 23 minutes after S-1-induced relaxation

Supplementary experiments

A series of additional experiments were conducted to see whether the stage of the oestrous cycle had an effect on the responses to the drugs that were investigated in the present study. Despite the interesting observations obtained from these supplementary experiments, they should be considered cautiously due to low n values. Hence, any trends observed may be purely speculative.

Impact of the stage of the oestrous cycle on 3nM isoprenaline-mediated relaxations of spontaneous and 100nM oxytocin-driven uterine contractions

Recent evidence has shown variations in KCNQ expression over the course of the oestrous cycle (McCallum et al., 2009). As a supplementary part of this investigation, experiments were conducted to see if responses to 3nM isoprenaline on spontaneous and OxyT-driven contractions were also altered at different stages of the oestrous cycle as this could potentially correlate to changes in Kv7 channel expression (Fig. 23). Application of 3nM isoprenaline to spontaneous contractions of segments from mice in the pro-oestrous part of the cycle produced a 25.6 ± 6.8% (n=3) relaxation (MIT), whereas the mean % relaxation observed in oestrous mice was 17.9 ± 5.9% (n=3) (Fig. 23A). A single experiment (n=1) conducted in a mouse at the metoestrous part of the cycle produced a 29.2% reduction in contractions (MIT).

Furthermore, experiments conducted on 100nM OxyT-driven contractions exhibited a 73.9 ± 6.4 (n=6), 33.3 ± 9.2 (n=3) and 36.4% (n=1) relaxation (MIT) when 3nM isoprenaline was applied to segments from pro-oestrous, oestrous and metoestrous mice respectively (Fig. 23B).

A.

B.

Fig 23. Comparative histogram illustrating the relaxant effects (MIT) of 3nM isoprenaline on uterine contractions at different stages of the oestrous cycle. Experiments were conducted on A. spontaneous and B. 100nM OxyT-driven contractions. Notably, the difference in the mean % relaxation (MIT) observed in segments from mice in the oestrous (n=3) compared to pro-oestrous (n=6) part of the cycle in OxyT-driven experiments was statistically significant (p<0.01). Error bars expressed as mean ± S.E.M

Impact of the stage of the oestrous cycle on the response to Kv7 activation in 100nM oxytocin-driven contractions

Expanding upon previous experiments investigating the relaxant effects of 1μM to 30μM S-1 and RL-3 on 100nM OxyT-driven contractions, supplementary analysis allowed the results of these initial experiments to be categorised into the specific stages of the oestrous cycle to see if the responses to these drugs were altered (Table 3; Table 4)

The relaxant responses of RL-3 on 100nM Oxy-T driven contractions suggest a slight stage-dependent response between the oestrous and metoestrous part of the cycle (Fig. 24A). Similarly, a stage-dependent trend was also indicated in the S-1 concentration-effect curve (Fig. 24B).

Table 3. Mean data for the RL-3 concentration-effect curve in 100nM OxyT-driven contractions at different stages of the oestrous cycle. Data values expressed as mean ± S.E.M

Table 4. Mean data for the S-1 concentration-effect curve in 100nM OxyT-driven contractions at different stages of the oestrous cycle Data values expressed as mean ± S.E.M where appropriate

A.

B.

Fig 24. Histogram illustrating the % relaxation (MIT) of 100nM OxyT-driven contractions for A. RL-3 and B. S-1 concencentration-effect curves. Data values expressed as mean ± S.E.M where appropriate

Discussion

This present study aimed to determine whether Kv7 channels are involved in isoprenaline-mediated relaxations of the mouse uterus through the use of various Kv7 modulators, namely chromanol-293b (Kv7.1 blocker), XE-991 (Kv7.1-7.5 blocker), RL-3 (Kv7.1 activator) and S-1 (Kv7.2-7.5 activator). Additionally, supplementary experiments using E4031 (ERG channel blocker) were conducted to investigate whether these channels play a role in generating the relaxant responses mediated by isoprenaline. Data obtained from these experiments were the first to suggest that Kv7 channels and perhaps to a greater extent, ERG channels, mediate the relaxant response to isoprenaline in spontaneous and OxyT-driven uterine contractions.

The project also investigated whether the physiological changes that the uterus undergoes during the oestrous cycle correlate to changes in its response to isoprenaline, or indeed its response to the Kv7 channel modulators. Data from the present study yielded some observations, for instance a reduced isoprenaline response at the oestrous stage of the cycle, but these are purely speculative due to the low number of experiments conducted.

Contribution of Kv7 channels to isoprenaline-mediated relaxations of the mouse uterus

β-adrenoceptors are fundamental regulators of contractility within smooth muscle. It is well established that activation of β-adrenoceptors stimulates an increase in the production of cAMP, an important secondary messenger which activates protein kinase A (PKA). PKA subsequently phosphorylates proteins that are involved in regulating contractility to eventually hyperpolarise the membrane and thereby produce relaxation of the smooth muscle by inhibition of the processes leading to contraction.

Several studies have demonstrated various K+ channel subclasses as targets for these cAMP and PKA-dependent processes, thus illustrating their role in mediating smooth muscle cell relaxation. One candidate that has been recurrently investigated is the large-conductance Ca2+-activated K+ channel (BKCa), which has been found to mediate smooth muscle relaxation via cAMP-dependent mechanisms in the coronary arteries of pigs (Scornik et al., 1993) and in the murine aorta (Sadoshima et al., 1988). Studies conducted more recently have demonstrated a similar role of these channels in β-adrenoceptor mediated relaxation of non-vascular smooth muscle, including the human bladder (Takemoto et al., 2008) and uterus (Chanrachakul et al., 2004). Notably, other K+ channel subtypes including ATP-sensitive K+ channels have been shown to be involved in the β-adrenoceptor signalling pathway (Shi et al., 2007). In light of these findings, it is possible that other K+ channels, such as Kv7 channels may be downstream effectors of β-adrenoceptor signalling. There is already evidence of GPCR and Kv7 channel interaction as seen by the negative effect of neuronal muscarinic receptor activation on M-channel currents (Brown & Adams, 1980). More recently it has been established that the relaxant effects of isoprenaline on rat renal myocytes are mediated through the activation of Kv7 channels (Chadha et al., 2012a).

Similarly, the current study proposes the idea that cAMP/ PKA dependent mechanisms also occur within the uterus as β-adrenoceptor activation by isoprenaline was shown to provoke a concentration-dependent relaxation of spontaneous and OxyT-driven contractions (Fig. 11). Furthermore, experiments conducted in the presence of XE-991, and to a lesser extent with chromanol-293b, suggested an attenuation of the mean relaxant response to isoprenaline in both spontaneous and OxyT-driven contractions (Fig. 15), indicating that Kv7 channels may be involved in β-adrenoceptor-mediated relaxation of SMCs in the uterus. These findings appear to be consistent with those by Chadha et al. (2012a) which similarly exhibited an attenuation of the isoprenaline response as well as diminished Kv currents generated by isoprenaline as a result of pre-incubation with linopirdine (Kv7 channel blocker) in rat renal arteries.

Interestingly, the present study also observed that application of chromanol-293b, a Kv7.1 channel blocker, elicited a slight relaxation of spontaneous contractions (Fig. 12), despite enhancing the OxyT response (Fig. 14). Neither chromanol-293b nor any other Kv7.1 blocker has been shown to mimic the stimulatory effects of XE-991 on spontaneous contractions despite KCNQ1 being the most abundantly expressed KCNQ gene in the non-pregnant uterus (McCallum et al., 2009). This is similar to the situation in all arteries studied (Chadha et al., 2012b). While no clear explanation can be drawn from this it is possible that this response may be due to the effects of chromanol-293b inhibition on channels, other than Kv7.1 channels, that are involved in regulating contractility. Research by Bachmann et al. (2001) has provided evidence that chromanol-293b inhibits chloride (Cl-) channels at higher concentrations (19μM). Perhaps then the concentration used in the current study inhibits these channels to a small extent, consequently preventing the efflux of Cl- out of the SMC resulting in hyperpolarisation of the membrane and thus producing a paradoxical inhibition of contraction (Jackson, 2000; Bachmann et al., 2001). Despite the considerable uncertainty surrounding the role of Kv7.1 in the uterus, it is widely thought that Kv7.1 channels do not play much of a role in regulating contractility of smooth muscle (Greenwood & Ohya, 2009; McCallum et al., 2011; Chadha et al., 2012b).

Another remarkable finding of the current study was that the response of the uterus to isoprenaline in the presence of the Kv7 channel blockers appeared to be more pronounced in OxyT-driven contractions compared to spontaneous contractions (Fig. 15). This observation may have been due to physiological antagonism, where an increase in the contractility of the uterus prevents isoprenaline from producing a maximal response, rather than blockade of Kv7 channels that are potentially crucial components of β-adrenoceptor-mediated relaxation. Moreover, experiments demonstrated OxyT to independently increase the frequency and magnitude of contractions by greater than 100 fold (Fig. 14) and blockade of Kv7 channels by XE-991 were also observed to potentiate the OxyT response (Fig. 14) as well as enhance spontaneous contractility (Fig. 12), as shown in previous studies (Yeung et al., 2007). To address whether this effect was due to physiological antagonism, investigations into the effect of isoprenaline on OxyT-driven contractions were observed in the presence of 1mM 4-AP, which at this concentration is considered to block most non-KCNQ Kv channels. Results demonstrated that pre-incubation with 1mM 4-AP had no effect on the response to isoprenaline despite 4-AP increasing the frequency and magnitude of contractions, therefore contesting any claims of physiological antagonism. Reinforcing this argument is the observation that lower concentrations of isoprenaline had a greater effect on OxyT-driven contractions compared to spontaneous contractions, despite contractility being significantly higher (Fig. 11). These findings therefore suggest a direct involvement of Kv7 channels in isoprenaline-mediated relaxations in the uterus.

The use of Kv7 channel activators provided an alternative approach to investigating the objectives of the present study, demonstrating that activation of Kv7 channels produces a relaxation of uterine contractions similar to that observed upon application of isoprenaline. Administration of RL-3, which selectively activates Kv7.1 channels, provoked relaxation of OxyT-driven contractions where mean responses were attenuated in the presence of XE-991 (Fig. 19). Experiments were also conducting using S-1 which is known to specifically activate Kv7.2-7.5 channels and has been shown to be a potent vasorelaxant (Chadha et al., 2012a). Similarly, the present data found S-1 to produce a relaxation of OxyT-driven contractions, which was notably less potent than those induced by RL-3 application. It is possible that this finding is a reflection of the observation that KCNQ1 is the most abundant isoform in the non-pregnant murine uterus (McCallum et al., 2009). Moreover, contrary to findings by Chadha et al. (2012a) where pre-incubation with linopirdine completely abrogated S-1 relaxations, the current study demonstrated S-1 responses to be only slightly attenuated in the presence of XE-991 (Fig. 21). Another key observation was that XE-991 reversed S-1-induced relaxations (Fig. 22), falling in line with results by McCallum et al. (2011) which used retigabine (Kv7.2-7.5 channel activator) in its place.

The findings from the present study are the first to suggest that Kv7 channels contribute to β-adrenoceptor mediated relaxations of the mouse uterus. Furthermore, observations also indicate a role of Kv7.2-7.5 channels in regulating the mechanical activity of the uterus but, in conjunction with other studies, refute that Kv7.1 channels has an effect on governing contractility. Despite this, selective activation of Kv7.1 channels produces a dose-dependent relaxation of contractions and activators targeting these channels are very potent at producing such responses. In light of this, it may be that instead of being major players in regulating contractility, Kv7.1 channels are more so key components of physiological dilator mechanisms.

Expanding upon the findings of this present study, the use of specific inhibitors and analogues of components involved downstream of the β-adrenoceptor signalling cascade may provide a greater insight into the mechanisms behind how these mediators modulate Kv7 activity to produce relaxation. Suppression of cAMP production by MDL-12, 330A (an adenylate cyclase inhibitor) as well as inhibition of cAMP-dependent PKA by KT5720 may shed light on the relative contribution of these mediators to β-adrenoceptor-mediated relaxation of the uterus. Furthermore, experiments to identify potential phosphorylation sites for cAMP/ PKA on Kv7 channels may consolidate the idea that these channels are coupled to cAMP/PKA dependent mechanisms. In relation to the role of Kv7 channels in regulating uterine contractility, the development of more selective modulators is likely to provide a greater insight into the relative importance of each isoform in modulating contractions.

Contribution of ERG channels to isoprenaline-mediated relaxations of the mouse uterus

There has been a growing amount of interest into the function of ERG channels in smooth muscle, where these channels have been shown to be expressed in the bovine epididymis as well as in the jejunum, portal vein and more recently, within the uterus of mice (Greenwood et al., 2009). In the present study blockade of ERG channels by E4031, significantly prolonged the duration of individual contractions (Fig. 13B) and also markedly increased ex vivo spontaneous contractions (Fig. 12), as demonstrated by Greenwood et al. (2009). This advocates the notion that ERG channels are functionally relevant components of uterine contractility. Furthermore, the effects of ERG channel blockade were observed in OxyT-mediated contractions which exhibited a fusion of contractions to produce a tonic contraction with an elevated baseline (Fig. 13D), replicating observations by Greenwood et al. (2009). However, contrary to findings by Greenwood et al. (2009) the MIT was attenuated compared to control values (Fig. 14), whereas it increased MIT in the literature. The dissimilarity of these results may be due to the fact that the concentration used in the current study (10μM) was greater than that used in the literature (1μM). Clearly, more experiments are required to elucidate these findings.

Experiments conducted to observe the effects of isoprenaline on spontaneous and OxyT-mediated contractions suggested diminished responses in the presence of E4031 (Fig. 15), indicating for the first time that ERG channel may be involved in β-adrenoceptor mediated relaxation. Notably, the attenuation of the response was greater than Kv7 channel blockade, suggesting that ERG channels may be relatively more important than Kv7 channels in producing the relaxant effects associated with isoprenaline application. This possibility itself opens a whole new opportunity for study.

Effect of the oestrous cycle on the responsiveness to isoprenaline and Kv7 channel activation

Experiments were conducted to observe whether the responsiveness to isoprenaline was altered by the stage of the oestrous cycle. Most of the mice used in the current study were in the pro-oestrous or the oestrous stage, which may presumably be a result of entrainment into the oestrous cycle by males in adjacent cages to where the mice were stored. Results from the present study found that the relaxant responses to isoprenaline were diminished at the oestrous part of the cycle compared to the pro-oestrous and metoestrous stages. Interestingly, this observation was statistically significant in OxyT-mediated contractions (Fig. 23B). Oestrogen levels are at their peak at the transition between the pro-oestrous and oestrous stages (Walmer et al., 1992). Oestrogen is thought to increase uterine contractility by increasing OxyT production and receptor expression which may explain the reduced response to isoprenaline in the current study (Griffiths et al., 2006). However, these findings may also be due to a decreased expression of Kv7 channels associated with the oestrous stage of the cycle, again reinforcing the idea of Kv7 channels contributing to relaxations of the uterus through β-adrenoceptor signalling.

Reinforcing this stage-dependent physiological role of Kv7 channels is the observation by McCallum et al. (2009) which demonstrates an upregulation of Kv7.1 channels in the transition from oestrous to the metoestous. Results from the current study appear to support this where activation of Kv7.1 channels by RL-3 produced greater relaxant responses at the metoestous compared to oestrous stage (Fig. 24A). Again, these findings consolidate the idea that Kv7 channel activity is a key component of the physiological processes that take place during the oestrous cycle. Despite these remarkable findings, it should be noted that the results generated from these experiments may be purely speculative and consequently require further study.

Conclusion

The current study is the first to suggest a role of Kv7 channels in β-adrenoceptor mediated relaxation of the uterus. Despite the interesting findings of the study, a few unaddressed disparities remain. The first is the characterisation of the subtype/s of β-adrenoreceptor that are involved in relaxation mediated through Kv7 channels. The second is the importance of cAMP/PKA dependent processes over other relaxant mechanisms (such as those mediated by guanylate cyclase) in producing relaxation in the uterus. Finally, development of more specific modulators may possibly clarify the relative physiological relevance of each Kv7 channel in regulating uterine mechanical activity. Given their role in governing uterine contractility, targeting Kv7 channels provides the potential for the development of novel therapies for the treatment of prolonged and pre-term labour. Moreover, with the widespread expression of Kv7 channels, their potential as targets to treat disease in other tissue appears endless.