The Kiss1 gene encodes a 145-amino acid protein that is cleaved into a 54-amino acid peptide known as kisspeptin-54 (Kp-54) as well as shorted peptides of 14, 13 or 10 amino acids ^[1]. These peptides, collectively referred to as kisspeptin, have a wide range of functions in different tissues and physiological processes. Upon its discovery in 1996, Kp-54, also known as metastin, was found to inhibit metastasis of malignant melanoma cells ^[2] which has since been replicated in breast cancer cells treated with Kp-10(Kp-10) ^[3]. This ability to inhibit the migration of cells was further confirmed when the movement of primary trophoblasts, crucial for placental development during pregnancy, was halted by Kp-10 treatment ^[4]. However, the most important role of kisspeptin is currently considered to be its control of reproductive function. Inactivating mutations of the Kp-10 receptor (Kiss1R, also known as GPR54) in humans are associated with failure to progress through puberty and adult infertility (hypogonadotropic hypogonadism) ^[5], ^[6]. Similar defects were found in GPR54 knockout mice ^[5]. Kp-10 stimulates secretion of the gonadotropic hormones luteinising hormone (LH) and follicle-stimulating hormone (FSH) ^[7] via a direct action on the hypothalamic gonadotropin releasing hormone (GnRH) neurones which contain Kiss1R ^[8]. Not only is Kp-10 the most potent activator of gonadotropic hormone secretion, but it plays a key role in regulating GnRH pulse generator frequency a critical component controlling the hypothalamo-pituitary-gonadal axis ^[9].

Recently it has been suggested that members of the kisspeptin family may also possess vasoactive activity. Kp-10 has been shown to inhibit the migration of human umbilical vein endothelial cells and subsequent angiogenesis through inhibition of vascular endothelial growth factor (VEGF) signalling ^[10]. Additionally, Mead et al (2007) identified transcription of both Kiss1 and Kiss1R mRNA in human aorta, umbilical vein and coronary artery but not other vessels ^[11]. It was further suggested that Kp-10, -13 and -54 may be novel vasoconstrictors in vivo as all three peptides resulted in contraction of human vessels in myograph studies with comparable potency to angiotensin II (Ang II), a potent vasoactive peptide ^[11]. Also, endothelial cells were postulated to be an alternative source of kisspeptin, possibly contributing to circulating plasma levels or acting in a local paracrine fashion. To date, studies have been confined to large conductance vessels. By comparison, the smaller vessels of the microvasculature are important for controlling tone and other vasoactive effects, where increases in permeability to plasma proteins lead to plasma extravasation and oedema formation.

Circulating plasma Kp-10 levels in normal individuals are believed to be low. However, the concentration of Kp-10 increases from 1.3fmol/ml pre-pregnancy to 9.6pmol/ml in the third trimester ^[12]. The source of this increase is believed to be the placenta, as Kp-10 levels drop almost immediately after birth ^[12]. The vascular effects of Kp-10 may be important as hypertension and oedema are symptoms of pre-eclampsia prevalent in late-term pregnancies but not after birth ^[13]. It has been shown that trophoblasts from women with pre-eclampsia have significantly higher kiss1 mRNA and Kp-54 peptide levels than trophoblasts during normal pregnancies ^[14]. By comparison, contrasting data has also been published suggesting that serum from women with pre-eclampsia contain less Kp-10 than control individuals ^[15]. At present it is known that circulating levels are sufficient to influence vascular function in physiological situations. However, kisspeptin's effect in either situation where circulating levels are high or where there are high local levels, as observed within the placenta during late pregnancy, may well influence local vasoactive events. With this concept in mind we have utilised established procedures for assessing microvascular blood flow and plasma extravasation in the cutaneous microvasculature of laboratory mice to investigate the ability of Kp-10 to influence vasoactive effects at the microvascular level. We have previously shown in studies of the cutaneous microvasculature and in similar in vivo assays that i) calcitonin gene-related peptide (CGRP) is one of the most potent vasodilators known ^[16] ii) endothelin-1 (ET-1) has the ability to mediate microvascular vasoconstriction ^[17] and iii) alarin (a novel member of the galanin family of peptides) mediates vasoconstrictor activity ^[18].

Here, we investigated the microvascular effects of Kp-10, the shortest biologically active proteolysis product of the Kiss1 gene, in the peripheral microvasculature of mice. The study was controlled through comparison with the known microvascular activities of the neuropeptides substance P, CGRP and endothelin-1.

Kisspeptins have been shown to possess well characterised responses in puberty and reproduction ^[29], but little is known about their cardiovascular effects. It has been suggested that kisspeptins may be potent vasoconstrictor compounds ^[11]. This study presents evidence that kisspeptins have the potential to mediate effects in the peripheral microvasculature.

A novel insight into Kp-10 biology gained during this study was the ability of Kp-10 to increase plasma extravasation, measured by both the spread of the plasma extravasation, and through an indication of intensity of staining, determined following Evans Blue accumulation. This was an acute response, visible at 7.5 minutes post-treatment, increasing in a dose-dependent manner, but requiring approximately a 30 times greater dose than that for substance P to elicit a similar plasma extravasation area. The plasma extravasation induced by Kp-10 was found to be histamine mediated, as co-administration of mepyramine, a histamine H₁ receptor antagonist ^[26] , abolished plasma extravasation. An obvious explanation for this phenomenon is Kp-10 induced mast cell degranulation. This is a characteristic shared by substance P in some tissues, whereby a hydrophobic amino acid sequence and terminal basic residue are required to stimulate mast cells, leading to release of histamine and other inflammatory mediators ^[30]. Accordingly, Kp-10 possesses characteristics similar to these structural elements. Substance P can also cause plasma extravasation via a histamine-independent mechanism, as observed here, providing a positive control in this study ^[31]. The use of a Kiss1R peptide antagonist p234, was attempted. However this possessed an increased hydrophobocity due to the addition of an N-terminal acetyl group ^[27], and was also shown to cause plasma extravasation. The lack of pallor ring at the injection site suggests that p234 is unable to stimulate vasoconstriction, most likely due to Kiss1R antagonism. This potentially identifies it as a functioning antagonist, but was not suitable for use locally in skin. Non peptide-based Kiss1R antagonists have been developed which may be more beneficial for future studies ^[32].

Kp-10 was shown to decrease peripheral blood flow as assessed by ^99mTc clearance from injection sites. However, it is noted that in the murine cutaneous microvasculature endothelin-1 is at least 100 times more potent. Similarly, Kp-10 induced plasma extravasation presented with a pallor ring of inhibition at its centre. This is consistent with previous findings whereby Kiss1R activation leads to increases in intracellular Ca²⁺ concentration ^[33] necessary for contraction of vascular smooth muscle cells ^[34]. However, inhibition of prostaglandin and thromboxane production by treatment with the cyclooxygenase antagonist indomethacin partially reduced Kp-10 mediated vasoconstriction. Kisspeptins have previously been shown to cause arachidonic acid liberation, the substrate for cyclooxygenases, in CHO cells ^[33]. The most likely cyclooxygenase-derived mediator of Kp-10-induced decreases in peripheral blood flow is the potent vasoconstrictor thromboxane A₂^[35]. The vasoconstriction reported in this study supports previous work suggesting kisspeptins cause constriction of several, but not all, human blood vessels ^[11], but we were unable to assess the contribution of Kiss1R activation to vasoconstriction due to the unsuitability of p234 as an antagonist.

An assessment of Kiss1 and Kiss1R mRNA expression levels in the heart, aorta and kidney was also performed. In agreement with previous studies in humans tissues Kp-10 expression levels were negligible ^[12], ^[36], even after treatment with a well known hypertensive agent (angiotensin II) ^[37]. In contrast, Kiss1R expression was present in all three tissues, but did not increase during hypertension either. This provides evidence that up-regulation of vascular-derived Kp-10 does not play a primary role in the onset of hypertension, although receptors are present in vascular tissues to respond to circulating or locally produced kisspeptin. In support of this concept, a recent publication was not able to provide a positive link between circulating Kp-10 levels and blood pressure changes ^[38].

Kiss1R has previously been shown to be expressed in the developing embryonic kidney and at birth in mice ^[39]. However, studies in human tissues have suggested that expression of both Kiss1 and Kiss1R mRNA is high in the brain and placenta, but low in both heart and kidney tissue ^[33], ^[36]. Pre-eclampsia, of which both increases and decreases in Kp-10 levels have been observed ^[14], ^[15], is a major condition which threatens the health of both mother and child during pregnancy. Expression of Kp-10 protein in syncytiotrophoblasts ^[4], found at the exchange site of maternal and foetal blood supplies, could influence maternal vascular tone as observed in this study. Additionally, this present data provides evidence for Kp-10 mediated stimulation of mast cell degranulation. Mast cell degranulation has been shown to lead to potent vasoconstriction in isolated placental vascular beds ^[40]. Furthermore, decreases in placental blood flow has been associated with increases in pre-eclampsia, decreased foetal development and other disorders. Whilst mast cell sub-types are likely to differ between skin and placenta, both are known to be equally stimulated by substance P ^[41]. Therefore, it is possible that kisspeptins may also be able to activate these mast cells if present in sufficient amounts locally in the placenta. Alternatively, increases in local perivascular production of kisspeptins could contribute to hypertension and plasma extravasation during pregnancy, but also other disorders in which expression of kisspeptins and Kiss1R increase. It is difficult to compare the doses used here with the plasma concentrations observed during pregnancy, but it is possible that the chronic exposure of mothers and foetus to Kp-10 during pregnancy as opposed to the acute, local treatment performed here may elicit the same vascular effects demonstrated.

In conclusion, this is the first study to describe the peripheral microvascular vasoactive properties of kisspeptins. The doses required indicate that these effects may not be of physiological relevance. On the other hand, these effects may become important in conditions where local tissue levels are high, such as in the placenta under certain conditions. Furthermore, knowledge of cardiovascular “off-target” effects could inform the design and application of kisspeptin-based therapeutics, including both antagonists and analogs with improved potency ^[42].

Competing Interests: The authors have declared that no competing interests exist.

Funding: These studies were supported by the British Heart Foundation and a Capacity Building Award in Integrative Mammalian Biology funded by the Biotechnology and Biological Sciences Research Council, British Pharmacological Society, Higher Education Funding Council for England, Knowledge Transfer Network, Medical Research Council, and Scottish Funding Council. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

We thank Aisah Aubdool for administrative assistance and Anthony Davenport for advice on kisspeptin expression data.