CLIN. AND EXPER. HYPERTENSION, 17(1&2), 141-152 (1995)

CENTRAL MONOAMINE SYSTEMS AND NEW ANTIHYPERTENSIVE AGENTS

Geoffrey A. Head Neuropharmacology Laboratory, Baker Medical Research Institute, Prahran, Victoria, 3 18 1, Australia.

Key words: Noradrenaline, Serotonin, Blood pressure, Methyldopa, Clonidine, Rilmenidine, Moxonidine

ABSTRACT This review describes the relationship between central monoamine

pathways and centrally acting antihypertensive agents. By using antagonists with affinity for a2-adrenoceptors and also imidazoline receptors we have found that the first generation agents clonidine and a-methyldopa activate 1x2-adrenoceptors while the newer second generation antihypertensive agents rilmenidine and moxonidine activate imidazoline receptors. Despite the difference in receptors activated, the hypotension produced by central administration of all agents was attenuated after chemical lesioning of the brainstem noradrenergic or serotonergic pathways suggesting a similar dependence on central monoamine pathways. Since the acute 6-hydroxydopamine-induced release of noradrenaline in the brainstem produces hypotension it suggests that these agents normally mimic brainstem noradrenergic function. By contrast the pressor response shortly following 5,6- dihydroxytryptamine suggests serotonergic neurones in the brainstem are pressor and that part of the anti-hypertensive action of centrally acting antihypertensive agents is mediated by inhibition of bulbar serotonergic pathways. We suggest that the similar haemodynamic and baroreflex effects of the two generations of agents can be explained by the a2-adrenoceptors and the imidazoline receptors being in series along the noradrenergic and serotonergic pathways.

141

Copyright 0 1995 by Marcel Dekker, Inc.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

142 HEAD

oductioq

Since the discovery of centrally acting anti-hypertensive agents such as clonidine and a-methyldopa there has been a suggestion that their mechanism of action involves monoamine systems, in particular those containing noradrenaline. a-Methyldopa was originally thought to be a false transmitter for peripheral sympathetic nerves but later it was discovered that it has a predominantly central action. Indeed a-methyldopa is a pro-drug which must first be metabolized to a-methylated metabolites such as a-methylnoradrenaline which is a potent a-adrenoceptor agonist [ 13. Finch and Haeusler first observed that centrally administered 6-hydroxydopamine prevented the hypotension following a-methyldopa and suggested that central noradrenergic neurons were the site of conversion of a-methyldopa to its active metabolites [2]. Clonidine is an imidazoline derivative with affinity for a2-adrenoceptors and like a-methyldopa lowers blood pressure through inhibition of central sympathetic tone [3]. There has always been considerable interest in how clonidine interacts with central noradrenergic pathways. Whether clonidine acts presynaptically on central noradrenergic neurons or postsynaptically on other neurons in the central nervous system has received much attention. While this question is somewhat narrow, being a translation of peripheral pharmacological concepts to the central nervous system, it has focussed our attention on whether such drugs mimic or inhibit endogenous neurotransmitters or whether they promote or prevent their release. In the last decade another dimension has been added to the debate principally with the work of Bousquet and colleagues. They suggest that it is not the a2- adrenoceptors that are important for clonidine’s hypotension but another distinct class of receptors with affinity for imidazoline compounds 141. This has led to the development of a second generation of centrally-acting antihypertensive agents such as rilmenidine and moxonidine which have a better selectivity than clonidine for imidazoline receptors. The important question is whether this greater relative selectivity for imidazoline receptors participates in the hypotensive mechanisms at therapeutically relevant doses. If it does, then it suggests that there are two distinct classes of centrally acting drugs mediating hypotension. Thus, the haemodynamic patterns, the modulation of baroreflex mechanisms produced by the two groups and their interaction with the central monoamine systems may differ markedly.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

CENTRAL MONOAMINES AND NEW ANTIHYPERTENSIVE DRUGS 143

The purpose of the current paper is to compare the first and the second generation centrally-acting antihypertensive agents with respect to the central receptors involved and also the relationship between central monoamine neurons.

Methods All the studies described were performed in male and female crossbred

conscious rabbits bred at the Baker Medical Research Institute that were chronically implanted with a vinyl catheter under halothane anaesthesia as described previously [ 5 ] . We have used conscious rabbits to avoid anaesthesia and central administration to eliminate possible peripheral actions of the drugs. On each experimental day the central ear artery of each rabbit was catheterized and the end of the intracisternal catheter was retrieved from under the skin at the back of the neck under local anaesthesia. Cardiovascular variables were recorded on a Grass (model 7D) polygraph. Mean arterial pressure was measured with a Statham P23 transducer and heart rate determined by a heart rate meter triggered from the arterial pulse.

Role of a?? olin r c rs To determine whether imidazoline receptors are important, we compared

the effects of antagonists with and without affinity for imidazoline receptors on the hypotension produced by one of the new second generation centrally-acting antihypertensive agents rilmenidine. Rilmenidine has a 3 fold higher selectivity for imidazoline receptors over a2-adrenoceptors than does clonidine [6]. It should be noted that at present all the antagonists have affinity for a2-adrenoceptors, thus it is important to allow for this effect using “pure” a2-adrenoceptor agonists and antagonists. Thus we compared the reversal of rilmenidine- and a-methyldopa- induced hypotension by idazoxan (a mixed imidazoline receptorlaz-adrenoceptor antagonist) and 2-methoxy-idazoxan (a2-adrenoceptor antagonist). The hypotension after intracisternal rilmenidine (22 pgkg) was reversed by lower doses of idazoxan than was the hypotension produced by intracisternal a-methyldopa (400 pgkg). By contrast, hypotension produced by a-methyldopa was reversed more readily by 2-methoxy-idazoxan [7,8]. The antagonists given by themselves had no effect on blood pressure. Thus the preferential reversal-effect

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

144 HEAD

of idazoxan suggests that rilmenidine acts principally through imidazoline receptors. We did observe that higher doses of 2-methoxy-idazoxan completely reversed rilmenidine-induced hypotension. This finding was surprising since it suggested that blockade of a2-adrenoceptors could reverse the hypotension produced by activation of imidazoline receptors. One explanation for this is that a2-adrenoceptors and imidazoline receptors may be in series along the same autonomic pathway [7,8].

In a further study we also examined the effects of intracisternal clonidine (1 pg/kg) but because of the short time course of action of this agent, were only able to use a single doses of idazoxan (30 pg/kg) and 2-methoxy-idazoxan (1 p g/kg). These doses had similar a2-adrenoceptor blocking effect as determined from earlier experiments with a-methyldopa. The hypotension produced by clonidine and a-methyldopa was similarly reversed by the two antagonists, suggesting that both of these agents act mainly on a2-adrenoceptors. By contrast the rilmenidine-induced fall in blood pressure was not affected by 2-methoxy- idazoxan, but was markedly reduced by idazoxan confirming a preferential effect on imidazoline receptors [9].

It has very recently been suggested that there are two subtypes of imidazoline receptors, 1-1 and 1-2 and that it is the former that are important for mediating hypotensive responses [lo]. Since idazoxan has greater selectivity for 1-2 than 1-1 receptors we also examined the effects of a highly selective 1-1 antagonist efaroxan. In addition we included the second generation agent moxonidine which has been shown from binding studies to have a 100000 fold selectivity for 1-1 compared to 1-2 receptors [lo]. Initially we performed dose response curves to efaroxan in the presence of a-methyldopa (to determine its relative 1x2-adrenoceptor blocking ability) and found that it was 6 times less potent than 2-methoxy-idazoxan. We administered intracisternally to conscious rabbits either clonidine, rilmenidine or moxonidine at a dose that produced 75% of the maximum hypotension (determined from earlier dose response curves). After 15 minutes either efaroxan or 2-methoxy-idazoxan were administered. The results are shown in Figure 1. 2-Methoxy-idazoxan produced a greater reversal of clonidine-induced hypotension than did the I- 1 receptor antagonist efaroxan. By contrast efaroxan produced a greater reversal of the rilmenidine- and moxonidine-

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

CENTRAL MONOAMINES AND NEW ANTIHYPERTENSIVE DRUGS 145

Clonidine Ril menidi ne Moxonidine 10 r r r

-25 L L L after Efaroxan I = a2+1PR block

after 2-M-ldazoxan m- a2 block

Figure 1 : Average change in mean arterial pressure (MAP) in conscious rabbits, after intracisternal clonidine (left panel), rilmenidine (middle panel) and moxonidine (right bars) before (open bar) and after a further injection of 2- methoxy-idaxoxan (striped bar) or the I- 1 imidazoline receptor antagonist efaroxan (black bar).

induced hypotension than did 2-methoxy-idazoxan (Fig 1). These results confirm our earlier findings that the first generation centrally-acting antihypertensive agent clonidine acts through a2-adrenoceptors while the second generation agents rilmenidine and moxonidine lower blood pressure through imidazoline receptors. Since we used efaroxan which has no 1-2 affinity it suggests that principally 1-1 subtype rather than the 1-2 receptors are involved. Thus the effects of idazoxan which were similar to efaroxan were also probably through 1-1 receptors despite it having about a 10 fold higher affinity for the 1-2 subtype. Thus it does appear that the second generation agents activate different receptors in the central nervous system than the first generation agents such as clonidine.

Comparison of baroreflex effecta

Given the different receptors activated, it would not be unexpected that there would be differences in autonomic effects. We have previously compared intracisternal and intravenous clonidine and a-methyldopa and found relatively little difference in haemodynamic and baroreflex effects at equal hypotensive

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

146 HEAD

doses [ I l l . However since we have now found that they both only activate a2-adrenoceptors this is now not so surprising. More recently we compared the effects of intracisternal rilmenidine with those of a-methyldopa in conscious rabbits with chronically implanted renal nerve electrodes in order to compare the changes to the renal sympathetic baroreflex produced with the two agents [12]. We showed that both drugs shifted the mean arterial pressure-renal sympathetic baroreflex curves to the left (in line with the hypotension) and produced a similar degree of inhibition of the upper and lower reflex sympathetic plateaus, a reduction in baroreflex gain and an inhibition of resting sympathetic tone (Fig. 2). Thus, despite activating different receptors in the central nervous system, the resultant changes to sympathetic control of the kidney were similar. This supports the view stated above that the imidazoline and the a2-adrenoceptors are on the same autonomic pathway [8].

Dependence of centrally-actine ant ihypertensive agents on central monoamine pathways.

There is little doubt that the hypotension and bradycardia produced by a-methyldopa is totally dependent on central noradrenergic neurons [5]. As mentioned above this is most likely due to the need for a-methyldopa to be converted to active metabolites inside the central noradrenergic neurons [ 13. However, we have also found that the chronic effects of intracisternal 6-hydroxydopamine were to reduce the clonidine-induced hypotension by about half [5]. This does not necessarily mean that clonidine is acting presynaptically on noradrenergic nerve endings but it does suggest that the integrity of noradrenergic neurons is important. In further studies we found that one month after intraspinal injection of 6-hydroxydopamine the hypotension or bradycardia to clonidine or a-methyldopa were not affected suggesting that spinal noradrenergic pathways were not involved and that the relevant noradrenergic pathway was in the brainstem [ 131. 6-Hydroxydopamine does not diminish the number of a2-adrenoceptors in the brainstem [14]. One possibility therefore is that the lack of noradrenergic neurons lessened the sensitivity of the “post-synaptic” a2-adrenoceptors. If this receptor was “downstream” from the imidazoline receptors then we would expect that rilmenidine hypotension would also be

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

CENTRAL MONOAMINES AND NEW ANTIHYPERTENSIVE DRUGS 147

Methyldopa Rilmenidine

100

, ,Control

- -1

75 -

50 -

25 -

0 2

O L O L

40 60 80 100 120 140 20 40 60 80 100 120

Mean Arterial Pressure (mm Hg)

Figure 2 : Renal sympathetic nerve activity (RSNA)-mean arterial pressure baroreflex curves produced by ramp changes blood pressure in conscious rabbits before (dashed line, control) and 3 hours after intracisternal a-methyldopa (400 pgkg, left panel, solid line) or 15-30 minutes after rilmenidine (22 pgkg, right panel, solid line). Symbols on curves indicate resting blood pressure and heart rate. Data adapted fiom [12].

attenuated after 6-hydroxydopamine. Indeed we have shown this to be so for rilmenidine [15] and others have shown this to be so for moxonidine [16]. In the latter study responses to clonidine was examined and found to be normal 4 days after intracisternal injection of 6-hydroxydopamine in rabbits. This suggests that at this early time point in the degenerative process produced by 6-hydroxydopamineY the relevant noradrenergic pathway was not hct ional and hence the “upstream” imidazoline receptor response was blocked but the post- synaptic a2-adrenoceptors were still operating normally.

We have also considered a possible role of serotonergic pathways in the actions of central acting hypotensive drugs and found that intracisternal injection of 5,6-dihydroxytryptamine reduced the hypotensive and bradycardic response to a-methyldopa, rilmenidine and clonidine [5,15]. Spinal injections of 5,6-dihydroxytryptamine did not alter the responses to either clonidine or a-methyldopa suggesting that spinal serotonergic pathways were not involved.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

148 HEAD

mdiovascular role o f central noradre- and wotone rpic - ' svstem

Considering the dependence of the centrally-acting antihypertensive agents on brainstem noradrenergic and serotonergic pathways it is relevant to discuss the normal function of these pathways. Do the drugs mimic or inhibit their activity ? In order to examine this question we have used specific neurotoxic drugs 6-hydroxydopamine and 5,6-dihydroxytryptamine administered to conscious rabbits. These drugs are selectivity taken up into the neuron by way of the neuronal amine uptake pump and destroy specific monoamine neurons [17]. 6-Hydroxydopamine is taken up mostly into noradrenergic neurons while 5,6-dihydroxytryptamine selectively targets serotonergic neurons. Within the first few hours after iiitracisternal injection of the neurotoxins, cardiovascular changes were observed which were due to an early neurotransmitter release phase (noradrenaline by 6-hydroxydopamine and serotonin by 5,6-dihydroxytryptamine), presumably as part of the initial degenerative process [ 1 8,191. These responses were blocked by pretreatment with specific antagonists (phentolamine for 6-hydroxydopamine and methysergide for 5,6-dihydroxytryptamine) and were prevented by prior destruction of the respective neurotransmitter system.

We examined the role of the brainstem monoamines in pontine decerebrate animals in order to eliminate the influence of suprabulbar monoamine pathways. Intracisternal injection of 6-hydroxydopamine produced a fall in blood pressme which was maximal at 30 minutes and which was of similar magnitude to that produced by clonidine. In addition, an increase in the gain of the baroreceptor heart rate reflex was also observed which was similar to that produced by clonidine [20]. Thus, the centrally-acting antihypertensive agents appear to mimic the function of the brainstem noradrenergic pathways. By contrast intracisternal administration of 5,6-dihydroxytryptamine in decerebrate rabbits produced the opposite effects, namely a hypertensive response and an inhibition of the baroreflex gain. The dependence on and opposite action to the serotonergic pathways of the clonidine-like drugs suggests that they normally inhibit brainstem serotonergic activity.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

CENTRAL MONOAMINES AND NEW ANTIHYPERTENSIVE DRUGS 149

Moxonidine

2-Methoxyidazoxan di &? Figure 3 : Schema showing the relationship between the receptors activated by the centrally-acting antihypertensive agents and the central monoamine pathways. For further explanation see conclusion. Abbreviations : NA, Noradrenergic ; 5- HT, serotonin ; IR, imidazoline receptors ; BP, blood pressure ; SNA, sympathetic nerve activity; downward arrow indicates a decrease.

Conclusion

Our studies suggest an important interaction of both the first and second generation centrally-acting antihypertensive agents with principally brainstem noradrenergic and serotonergic pathways. A schema representing our current hypothesis is shown in Figure 3. The predominant actions of the noradrenergic neurons in the brainstem are to inhibit sympathetic vasomotor tone. Clonidine directly activates a2-adrenoceptors (mimicking noradrenaline) while a-Methyldopa is first taken up into noradrenergic neurons (NA) to lower sympathetic vasomotor tone and hence blood pressure. The a2-adrenoceptors are blocked by all the antagonists (2-methoxy-idazoxan, idazoxan and efaroxan). By contrast, rilmenidine and moxonidine activate imidazoline receptors of the I- 1 subtype “up stream” from the noradrenergic neurons. These receptors are only blocked by idazoxan and efaroxan. With this arrangement it can be seen that high doses of a2-adrenoceptor blockers will abolish the responses of all the centrally- acting antihypertensive agents as will destruction of the brainstem noradrenergic

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

150 HEAD

neurons. In addition the haemodynamic patterns and baroreflex responses will be virtually identical with all the agents. Thus imidazoline receptor agonist drugs have the advantage of the same cardiovascular profile as the first generation centrally-acting antihypertensive agents but without some of the side effects such as sedation which appears to be due to a2-adrenoceptor stimulation at the level of the locus coeruleus [21]. The serotonergic neurones in the brainstem normally increase blood pressure and are inhibited by a2-adrenoceptor activation. This serotonergic pathway pressor neuron must lie “downstream” from the noradrenergic pathway since destruction of these neurons attenuated hypotension produced by all agents so far examined. It is not absolutely clear at the present moment where each of the relevant monoamine pathways lies in the brainstem.

Acknowledgmen&

These studies were supported by a block Institute grant from the National Health and Medical Research Council of Australia and by a grant from Institut de Recherches Internationales SERVIER & Compagnie Developpement. The technical assistance of Shirley Godwin is greatly appreciated.

Reference3

1.

2.

3.

4.

5.

Head, G.A., de Jong, W. Central versus peripheral adrenergic regulation of blood pressure: contributions to the actions of alpha-methyldopa. Adrenergic Blood Pressure Regulation, (ed.) W. Birkenhager, F. B. Struyker-Boudier, & H. Struyker-Boudier, Excerpta Medica, Amsterdam,

Finch, L., Haeusler, G. Further evidence for a central hypotensive action of a-methyldopa in both rat and cat. Br J Pharmacol. 1973; 47: 217-228. Kobinger, W. Central alpha-adrenergic systems as targets for hypotensive drugs. Rev Physiol Biochem Pharmacol. 1978; 81: 40-100. Bousquet, P., Feldman, J., Schwartz, J. Central cardiovascular effects of alpha adrenergic drugs: Differences between catecholamines and imidazolines. J Pharmacol Exp Ther. 1984; 230: 232-236. Head, G.A., Korner, P.I., Lewis, S.L., Badoer, E. Contribution of noradrenergic and serotonergic neurons to the circulatory effects of centrally acting clonidine and alpha-methyldopa in rabbits. J Cardiovasc Pharmacol.

1985,103-113.

1983; 5: 945-953.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

CENTRAL MONOAMINES AND NEW ANTIHYPERTENSIVE DRUGS 151

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Bricca, G., Dontenwill, M., Molines, A., Feldman, J., Tibirica, E., Belcourt, A., Bousquet, P. Rilmenidine selectivity for imidazoline receptors in human brain. Eur J Pharmacol. 1989; 163: 373-377. Sannajust, F., Head, G.A. Rilmenidine induced hypotension involves imidazoline preferring receptors in conscious rabbits. J Cardiovasc Pharmacol. 1994; 23: 42-50. Head, G.A., Sannajust, F. Importance of imidazoline receptors in the cardiovascular responses to clonidine and rilmenidine in conscious rabbits. Fundam Clin Pharmacol. 1992; 6 (Suppl 1): 3 1 s-43s. Head, G.A., Godwin, S.J., Sannajust, F. Differential receptors involved in the cardiovascular effects of clonidine and rilmenidine in conscious rabbits. J Hypertens. 1993; 11: S322-S323. Ernsberger, P.R., Westbrooks, K.L., Christen, M.O., Schafer, S.G. A second generation of centrally acting antihypertensive agents act on putative I( 1)- imidazoline receptors. J Cardiovasc Pharmacol. 1992; 20: sl-sl 0. Badoer, E., Head, G.A., Korner, P.I. Effects of intracisternal and intravenous alpha-methyldopa and clonidine on haemodynamics and baroreceptor-heart rate reflex properties in conscious rabbits. J Cardiovasc Pharmacol. 1983; 5:

Sannajust, F., Head, G.A. Effect of rilmenidine on the baroreflex control of renal nerve activity. J Hypertens. 1993; 1 1 : S328-S329. Elghozi, J.L., Head, G.A., Wolf, W.A., Anderson, C.R., Korner, P.I. Importance of spinal noradrenergic pathways in cardiovascular reflexes and central actions of clonidine and alpha-methyldopa in the rabbit. Brain Res.

U'Prichard, D.C., Bechtel, W.D., Rouot, B.M., Snyder, S.H. Multiple apparent alpha-noradrenergic receptor binding sites in rat brain: effect of 6-hydroxydopamine. Mol Pharmacol. 1979; 16: 47-60. Head, G.A., Burke, S.B. Importance of central noradrenergic and serotonergic pathways in the cardiovascular actions of rilmenidine and clonidine. J Cardiovasc Pharmacol. 1991 ; 18: 8 19-826. Schlicker, E., Armah, B.I., Gothert, M. Central presynaptic alpha 2- autoreceptors are involved in the blood pressure-lowering effect of moxonidine. J Cardiovasc Pharmacol. 1990; 16: 15-22. Jonsson, G. Chemical neurotoxins as denervation tools in neurobiology. Ann RevNeurosci. 1980; 3: 169-187. Head, G.A., Korner, P.I. Mechanisms of acute hypertension and bradycardia following intracisternal 6-hydroxydopamine in conscious rabbits. Eur J Pharmacol. 1980; 66: 111-115.

760-767.

1989; 499: 39-52.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.

152 HEAD

19. Head, G.A., Komer, P.I. Cardiovascular functions of brain serotonergic neurons in the rabbit as analysed from the acute and chronic effects of 5,6- dihydroxytryptamine. J Cardiovasc Pharmacol. 1982; 4: 398-408. Head, G.A. Central monoamine neurons and cardiovascular control. Kidney Int. 1992; 41 (Suppl. 37): S8-Sl3. Scheinin, M., Schwinn, D.A. The locus coeruleus - site of hypnotic actions of alphd- adrenoceptor agonists. Anesthesiology. 1992; 76: 873-875.

20.

2 1.

Clin

Exp

Hyp

erte

ns D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

onas

h U

nive

rsity

on

02/1

4/13

For

pers

onal

use

onl

y.