In the present study, we identified interactions between cannabinoid CB₁Rs and AT_1/2 receptors for Ang II and MasRs for Ang 1–7 in the PVN involved in blood pressure regulation. The experiments were performed on conscious rats and revealed quantitative differences between SHR and WKY. Conscious rats were used since anesthesia reverses BP response to cannabinoids (e.g., THC, AEA or CP55940) relative to prolonged hypotension (for review, see [17,18]). We used the CB_1/2 receptor agonist CP55940 [26] since the effect of this drug microinjected into PVN was reduced by the microinjection of the CB₁R antagonist AM251 but not CB₂R antagonist SR144528 in PVN [19]. We restricted ourselves to pressor responses since the fall in BP elicited by microinjection of CP55940 into PVN was only reduced by AM251 but not by the antagonists of other receptors including AT₁Rs in our previous experiments [20]. CP55940 was routinely examined after a previous i.v. administration of AM251 prevented hypotensive responses to CP55940 [19,20].

Ang II (0.3 nmol/rat), Ang 1–7 (0.3 nmol/rat) and CP55940 (0.1 nmol/rat; after i.v. administration of AM251) microinjected into PVN increased SBP, DBP and MBP in conscious WKY and SHR. The strongest pressor response was elicited by Ang II and the weakest one by Ang 1–7 (by ~20 and ~6 mmHg in WKY, respectively). Sun et al. [11] showed that the same doses of Ang II and Ang 1–7, given in PVN, induced comparable increases in MBP in conscious male Sprague-Dawley rats. CP55940 given in PVN (after i.v. administration of AM251) increased BP both in conscious (current study) and urethane-anaesthetized rats [19,20]. By contrast, as mentioned above, the i.v. administration of cannabinoids mainly increased BP in conscious animals but decreased it in anaesthetized animals [17], suggesting that central mechanisms are responsible for the pressor effects of cannabinoids.

The pressor responses to Ang II and Ang 1–7 result from the activation of AT₁/AT₂ receptors and Mas receptors, since they were almost prevented or strongly inhibited by respective antagonists losartan/PD123319 and A-779 (given into the PVN). AT₁Rs in the PVN are well known to mediate the pressor response to Ang II. By contrast, central AT₂Rs are rather known to counteract the effects of Ang II (for review, see [27]). As in our study, not only losartan but also PD123319 strongly reduced Ang II-induced increases in BP in the PVN of anaesthetized rats [28,29]. Sun et al. [11] showed that losartan inhibited the pressor effects induced by Ang II but not by Ang 1–7, which were reduced by A-779, confirming the involvement of AT₁- and MasRs in the increases in BP elicited by Ang II and Ang 1–7, respectively. As mentioned above, the involvement of CB₁Rs in the pressor effect of CP55940 after its administration into PVN has been proven in our previous study [19].

To the best of our knowledge, we are the first to demonstrate that the pressor responses to Ang II, Ang 1–7 and CP55940 were significantly higher (or tended to be higher) in SHR than in WKY. Higher increases in BP induced by Ang II [10,11] and Ang 1–7 [11,15] have so far been shown in renovascular hypertensive rats only. Increases/decreases in blood pressure in response to a given ligand depend on the value of basal BP. Thus, we cannot exclude the possibility that the higher responses observed in SHR were related to higher basal BP. However, we would like to underline that the higher pressor effects in SHR, when compared to WKY, may result from different densities of the particular receptors in the brain regions responsible for cardiovascular system regulation. The balance of excitatory and inhibitory synaptic inputs depending on glutamatergic and GABAergic transmission is responsible for the final integration of the sympathetic outflow by PVN [1,4,27]. In PVN, facilitatory AT₁Rs and AT₂Rs are expressed primarily on glutamatergic and GABAergic neurons, respectively. In addition, AT₁Rs diminish GABAergic inputs [13,27,30,31]. Presynaptic inhibitory CB₁Rs are localized both on glutamatergic and GABAergic synapses [32]. However, we have provided evidence previously that CP55940 increased BP, acting predominantly at inhibitory presynaptic CB₁Rs on GABAergic neurons [20]. Both an enhancement of sympathoexcitatory and a reduction in sympathoinhibitory inputs might lead to increases in BP.

Compared to WKY, the PVN of SHR has higher levels of AT₁ [12,13] and CB₁ (current study) receptors. AT₁ and CB₁ receptor expressions are also higher in SHR than WKY in other brain regions, that are crucial for cardiovascular regulation, such as in RVLM and NTS, in which similarly to PVN, glutamate and GABAs interact to modulate BP (for AT₁Rs—[12,33]; for CB₁Rs—current study). A higher level of CB₁Rs but not of endocannabinoids in the RVLM of SHR has been described previously by Wang et al. [34]. To summarize, enhanced AT₁R (stimulation of glutamatergic neurons) and CB₁R (inhibition of GABAergic neurons) densities in the PVN of SHR might explain higher BP response to Ang II and CP55940 in SHR than in WKY (Figure 8).

With respect to AT₂ and Mas receptors, Western blotting analysis showed significantly higher expression in WKY than SHR in each of the three brain regions: PVN, RVLM and NTS. AT₂ and Mas receptors are known to counteract the pressor effect of AT₁R stimulation by the enhancement of GABAergic input directly by a facilitatory effect on GABAergic neurotransmission (AT₂Rs are located predominantly on GABAergic neurons; [27]) or indirectly via nitric oxide stimulating GABA release determined in the PVN both for AT₂ [35] and Mas [36] receptors. Thus, the decrease in AT₂ and Mas receptor expression and of the sympathoinhibitory input observed in our study might contribute to the higher level of BP in SHR. Of course, one should keep in mind that a change in receptor expression does not always correlate to functionality, and our results require confirmation by studies dedicated to the downstream pathways regarding specific receptor activation. Interestingly, the density of MasRs in the PVN was higher in renal hypertensive rats (2K1C) compared to their normotensive controls [15]. However, the following question arises: how can we explain the pressor effects of Ang 1–7 and the involvement of AT₂Rs in the increase in BP induced by Ang II? The enhancement of BP elicited by Ang 1–7 has so far been explained by the production of reactive oxygen species (ROS; [16]) and the pressor response induced by AT₂Rs is probably mediated via vasopressinergic neurons in the PVN [37], since the increases in BP and HR induced by vasopressin microinjected into the PVN of anaesthetized rats were partially attenuated by the AT₁R antagonist losartan and strongly reduced by the AT₂R antagonist PD123319 [37].

The following facts confirm the existence of an interaction between cannabinoid CB₁Rs, AT₁ and AT₂ receptors for Ang II and MasRs for Ang 1–7 in the PVN. Firstly, the CB₁R antagonist AM251 given into the PVN reduced or tended to diminish the pressor response to Ang II both in WKY and SHR and even reversed increases in BP elicited by Ang 1–7 to hypotensive responses. An inhibitory effect of AM251 on the pressor response to Ang II (both given into the PVN) was previously shown by Gyombolai et al. [21]. Secondly, the pressor response to CP55940 was strongly reduced by previous topical applications of AT₁, AT₂ and Mas receptor antagonists losartan, PD123319 and A-779, respectively. In our previous papers, losartan i.v. did not modify the cardiovascular response to AEA i.v. (including its pressor phase; [38]) and the hypotensive and bradycardic responses to CP55940 microinjected into PVN. On the other hand, it even reversed the increase in BP and HR that was induced by this latter compound injected in PVN [20]. In the current study, losartan did not reverse the pressor response to CP55940 in BP probably because it was given locally to PVN in contrast to i.v. injection in the previous paper [20]. Thirdly, topically administered A-779 decreased BP by itself to a comparable degree both in WKY and SHR but only after the previous i.v. administration of AM251. By contrast, losartan and PD123319 failed to modify BP both in the absence or presence of AM251 i.v. Thus, it seems that endocannabinoids acting on CB₁ receptors in PVN interact with the endogenous pressor tone elicited by Ang 1–7 but not Ang II. Similarly, it has been shown previously that A-779 (3 nmol/rat) given in PVN reduced BP, renal sympathetic nerve activity and plasma noradrenaline levels in anaesthetized high salt-induced or renal hypertensive rats but not in the respective control animals [11,16]. We are the first to demonstrate an interaction of Ang 1–7 and its MasRs with cannabinoid CB₁Rs in the central nervous system. Thus far, the existence of such an interaction was suggested only in the rat uterus in which local Ang 1–7 infusion enhanced CB₁ and CB₂ receptor expression [22].

How can we explain the interaction between CB₁ receptors with AT₁, AT₂ and Mas receptors (Figure 8)? The paracrine transactivation of CB₁ receptors by endocannabinoids released after activation of AT₁Rs by Ang II in cells expressing both AT₁ and CB₁ receptors diminished the response to Ang II in various cultured cells [39] and vessels (for review, see [40]) and in the heart [41]. On the basis of the latter data, one might assume that AT₁R activation in PVN leads to the release of endocannabinoids, which in turn activate presynaptic inhibitory CB₁Rs mainly on GABAergic neurons (for detailed explanation, see above). Thus, the CB₁R antagonist AM251 inhibits the pressor response to Ang II since it antagonizes presynaptic CB₁Rs responsible for the reduction in GABA release. Consequently, the enhanced inhibitory influence on glutamatergic input reduces the pressor response to Ang II. On the other hand, in our hands, losartan inhibited the pressor response to CP55940, which mainly results from the reduction in inhibitory GABAergic tones via activating presynaptic CB₁Rs by their agonist (see above). Since the final BP response in PVN is determined by the balance between GABAergic and glutamatergic inputs, we suppose that losartan blocking AT₁Rs on glutamatergic neurons might reduce the pressor response to CP55940. We are not able to comment on the mechanisms related to AT₂ and Mas receptors because only very limited data regarding their pressor effects are available so far. One should keep in mind that impaired GABAergic and/or enhanced glutamatergic inputs result in elevated sympathetic outflow and hypertension (including SHR) [4]. In our hands, AM251 reduced the pressor response to Ang II in SHR more strongly than in WKY, confirming the above suggestion that it mainly inhibits presynaptic inhibitory CB₁ receptors located mainly on GABAergic neurons.

Interestingly, CB₁R antagonists enhanced vasoconstrictor responses to Ang II in various isolated arteries (for review, see [40]) but decreased the increase in BP induced by the microinjection of Ang II into the PVN (current paper), confirming that CB₁R activation in the peripheral vascular bed and in the PVN is mainly responsible for hypotensive and hypertensive effects of endocannabinoids, respectively (for review, see [18]).

In this context, it is also of interest to comment on the effect of the CB₁R antagonist AM251 when given alone. AM251 administered i.v. and into PVN increased BP by about 15–20 mmHg and 6–8 mmHg, respectively, both in WKY and SHR. Its effect following i.v. injection probably results from the activation of peripheral presynaptic inhibitory CB₁Rs located on sympathetic nerve endings innervating resistance vessels [42]. Accordingly, the fall in BP in response to CP55940 in anaesthetized rats was reversed to hypertension by the peripheral CB₁R antagonist AM6545 [19]. On the other hand, the weaker enhancement of BP after its application in PVN might be caused by the blockade of CB₁Rs responsible for the fall in BP by endocannabinoids released in the PVN (first phase of the response to CP55940; [19,20]. An increase in BP by AM251 given in PVN was also observed in the urethane-anaesthetized [21] (0.3 nmol/rat) rats but not by our group [19]. In our previous study, AM251 given i.v. also failed to affect BP in urethane-anaesthetized rats; they have a lower sympathetic tone (lower basal BP than in the present paper; [19,20]) and probably also a lower activity with respect to the endocannabinoid system.