β-Adrenergic Activation Initiates Chamber Dilatation in Concentric Hypertrophy
It is uncertain whether chronic β-adrenoreceptor (β-AR)–activation in hypertension could initiate the progression from compensated left ventricular (LV) hypertrophy to pump dysfunction. It is also uncertain if this effect is through adverse LV remodeling (chamber dilatation with wall thinning and pump dysfunction) or intrinsic myocardial contractile dysfunction. We evaluated the effect of 5 months of isoprenaline (0.02 mg · kg−1 · d−1) on hemodynamics, LV wall thickness, cavity size, and interstitial characteristics in spontaneously hypertensive rats (SHR) with compensated LV hypertrophy. In the absence of myocyte necrosis, changes in volume preload, pressure afterload, and heart rate or decreases in baseline systolic myocardial elastance (load independent measure of intrinsic myocardial contractility), ISO produced a right shift in LV diastolic pressure–volume (P-V) relations (chamber dilatation), a decrease in LV wall thickness despite a further increase in LV weight in SHR, LV pump dysfunction (right shift in LV systolic P-V relations), and deleterious interstitial remodeling (increments in total and noncrosslinked myocardial collagen concentrations). The isoprenaline-induced LV geometric, chamber performance, and interstitial changes were similar to alterations noted during decompensation in older SHR. In summary, in the absence of tissue necrosis and baseline intrinsic myocardial contractile dysfunction, chronic β-AR activation induces interstitial and chamber remodeling and, hence, pump dysfunction. These data suggest that chronic sympathetic activation initiates the progression from compensated concentric LV hypertrophy in hypertension to cardiac dysfunction primarily through deleterious cardiac remodeling rather than intrinsic myocardial contractile dysfunction.
Left ventricular (LV) hypertrophy (LVH) is an independent risk factor for the development of heart failure.1 Although LVH in hypertension can be considered an adaptive response to reduce wall stress,2 LVH precedes the development of chronic heart failure.3–5 Indeed, present guidelines have been adapted to underscore the evolution and progression from hypertension via LVH to overt heart failure with dilatation and systolic dysfunction.6 Despite the acknowledged importance of LVH as a risk factor for heart failure, the fundamental mechanisms involved in contributing to the progression from compensatory LVH in hypertension to heart failure are largely undefined.
It is now well recognized that adrenergic activation in heart failure7 contributes to progressive cardiac dysfunction.8,9 Although plasma catecholamine concentrations are elevated in hypertensive patients with LVH,10 and LV mass is correlated with cardiac noradrenaline release,11 there is no evidence to indicate that in patients with LVH, excessive catecholamine production contributes to progressive cardiac dysfunction. The mechanisms that underlie cardiac dysfunction are decreased intrinsic myocardial contractility, adverse chamber remodeling, or a combination of these. In patients with heart failure, it has been postulated that adrenergic activation contributes to progressive cardiac dysfunction through intrinsic myocardial contractile dysfunction and, subsequently, chamber dilatation.12 However, recently it has been suggested that deleterious cardiac remodeling, rather than decreased intrinsic myocardial contractility, may be a precursor of heart failure.13,14 We hypothesized that excessive adrenergic activation could initiate the progression from compensated LVH in hypertension to cardiac dysfunction, and that this effect is primarily through adverse LV remodeling (dilatation). To examine these hypotheses, we evaluated whether chronic β-adrenoreceptor (β-AR) activation, after daily low-dose isoproterenol administration, could induce premature progression to LV pump dysfunction through either intrinsic myocardial contractile dysfunction or LV dilatation in spontaneously hypertensive rats (SHR) with compensated concentric LVH. The response to β-AR antagonists was not assessed, as their antihypertensive action would not be distinguishable from myocardial effects.
The present study was approved by the Animal Ethics Screening Committee (AESC) of the University of the Witwatersrand (AESC approval number 99/01/2b).
Seven-month-old SHR (n=32; OLAC) and Wistar Kyoto controls (WKY, n=34; Kleinterfarm Madorin Ltd, Germany) were assigned to a group of rats (11 SHR and 8 WKY) who received isoprenaline (ISO, Imuprel, Adcock Ingram) at a dose of 0.02 mg · kg−1 · d−1 subcutaneously (≈0.1 mL) for 5 months,15 to a group of rats (11 SHR and 14 WKY) that received the same volume of the vehicle of ISO (0.1 mL of 0.9% saline) for 5 months, and to a group of rats (10 SHR and 12 WKY) that received no treatment until hemodynamic and interstitial changes were evaluated at 21 to 22 months of age. Rats at 21 to 22 months of age were included in the study in order to assess whether the effect of ISO on LV geometry, hemodynamics, and interstitial characteristics in SHR has similarities consistent with the LV changes noted in SHR during the period when decompensation and dilatation occurs.4,16,17 None of the rats died during ISO or vehicle administration, and 1 untreated SHR died of an indeterminate cause at 16 months of age (not included in the above sample numbers). Noninvasive systolic blood pressures (SBPs) were measured as previously described18 at regular periods throughout the study.
Chronic ISO administration may produce an enhanced LV filling volume through β2-AR–mediated vasodilation and/or sympathetic effects on the kidney and, consequently, mediate LV remodeling through indirect mechanisms (preload-induced changes). To assess this hypothesis, we also evaluated whether a relatively short period of ISO (4 weeks) administered to Sprague-Dawley (SD, Harlan-Olac, Bicester, UK) rats (9 ISO-treated and 8 controls) produced alterations in LV filling dimensions as determined in intact animals. SHR could not be used for this study, as the small LV cavity volumes of untreated 7-month-old SHR prevented us from accurately identifying the leading edge of the posterior wall endocardial surface (which was obscured by papillary muscle images) from echocardiographic images in this group.
LV Cavity Size and Geometry Determined In Vivo
LV dimensions and geometry were determined in anesthetized, ventilated, open-chest rats. The surgery, instrumentation, experimental techniques, and calculations used in the present study have previously been described and validated.4,16,19 Briefly, LV end diastolic (LVED) short-axis external diameters were measured using piezoelectric ultrasonic transducers, and LVED pressures (LVEDP) were measured using a fluid-filled catheter (with an amplitude-frequency response, as determined when coupled with the pressure transducer, uniform to 10 Hz)17 inserted through the apex of the LV. Measurements of LVED external diameters and LVEDP were obtained over a range of LVEDP values by manipulating blood volume using an iso-oncotic solution, as well as through inferior vena cava occlusion. LVED radius (r) and wall thickness (h) was determined from previously described formulae.16 Relative wall thickness was determined from LVEDh/LVEDr. LV remodeling was also assessed from the LVEDr intercept of the LVEDP-LVEDr relation (LVEDr0). An appropriate range of LVEDP and LVED external diameter measurements were successfully obtained in all rats surviving the duration of the study.
In SD rats used to assess the effects of ISO on preload, a day before assessing LV geometry and dimensions using piezoelectric transducers, echocardiography was performed ≈1 hour after the last dose of ISO and after 4 weeks of daily injections of ISO. The anesthetic and the methodological approach used have previously been described.20 In the present study, all data were acquired using a model 2500 Hewlett Packard echocardiograph with a 7.5-MHz transducer.
Isolated Perfused Heart Preparations
After the collection of LVEDP and LVED external diameter data in vivo, hearts were excised and LV systolic function and LV remodeling were assessed in vitro in isolated perfused heart preparations as previously described.15 Briefly, hearts were perfused retrogradely at a constant flow (12 mL/g wet heart weight) with 37°C physiological saline solution, and paced at 360 bpm with platinum electrodes attached to the left atrium and the apex of the heart. LV systolic and diastolic pressures were determined by use of a water-filled balloon-tipped catheter over a range of LV volumes. LV pressures were determined at as many multiple small increments in volume as were practically possible to improve on the accuracy of curve fitting during later analysis.
LV systolic chamber performance (a measure of systolic pump function) was determined in vitro from the slope (E) of the linear portion of the LV peak systolic P-V relation. Intrinsic myocardial systolic performance (a load-independent measure of intrinsic myocardial contractility) was assessed in vitro from the slope (En) of the systolic stress (ς)-strain relation.14 Systolic ς and strain were calculated using previously described formulae,21 assuming a thick-walled spherical model of LV geometry.
LV remodeling was assessed in isolated perfused heart preparations from the volume intercept (V0) of the LV diastolic P-V relation.15
Myocardial Collagen and Histology
Samples of LV tissue were weighed and stored at −70°C for tissue analysis. Myocardial hydroxyproline concentrations were assessed using the method of Stegemann and Stalder after acid (HCl) hydrolysis23 Myocardial collagen was also extracted and digested with cyanogen bromide.15 The cyanogen bromide–digested collagen sample was subjected to acid hydrolysis and hydroxyproline concentrations determination. The amounts of noncrosslinked (soluble) and crosslinked (insoluble) collagen in the myocardium were determined as previously described.15
Before cardiac tissue was stored for biochemical assessment, a longitudinal slice of the LV from the apex to the base through both the anterior and the posterior LV walls was stored in formalin for subsequent histology. LV tissue was processed, sectioned, stained, and assigned a pathological score as previously described.15,16 We have previously validated the usefulness of this approach in detecting myocyte necrosis in SHR with LV dilatation.16 Briefly, pathological score was determined from histological evidence of the extent of patchy diffuse contiguous subendocardial or transmural fibrosis.16
Regression analysis was used to determine the lines of best fit for the cardiac function relations. LV systolic P-V and LV systolic ς-strain relations were found to best fit a linear function. Differences in LV geometry, LV chamber and myocardial performance, hemodynamics, pathological score, and myocardial collagen biochemical analysis between groups were assessed by a 1-factor ANOVA followed by a Tukey post hoc test. All values in the text are represented as mean±SEM.
Blood Pressure and Heart Rate
As determined in unanesthetized, restrained rats SHR had an increased SBP, but no differences in heart rate, compared with WKY controls throughout the study (SBP a week before assessing LV geometry and function: SHR, 185±8 mm Hg; WKY, 122±6 mm Hg; P<0.001). Similarly, throughout the duration of the study, ISO administration failed to influence either SBP (SHR receiving ISO: determined 2 weeks after initiating ISO injections, 180±9 mm Hg; determined a week before LV geometry and function measurements, and 182±8 mm Hg) or heart rate (SHR not receiving ISO, 462 bpm; SHR receiving ISO for 2 weeks, 472 bpm; and SHR receiving ISO, 470 bpm as determined a week before LV geometry and function measurements and 15 to 45 minutes after ISO injection) in either SHR or WKY. The lack of effect of ISO on SBP and heart rate in unanesthetized, restrained rats was noted throughout the study, irrespective of whether measurements were taken within 30 minutes, 2 hours, or 12 hours of ISO injection (data not shown). Moreover, ISO failed to influence carotid SBP, diastolic BP, or pulse pressure (SHR, 44 mm Hg; ISO-treated SHR, 48 mm Hg) as determined in anesthetized rats.
LV and Body Weight
SHR at all ages had an increase in LV weight compared with age-matched WKY controls (Table 1). ISO administration from 7-to-12 months of age further increased LV weight in SHR to values not significantly different from SHR at 21 to 22 months of age (Table 1). Although ISO administration to WKY tended to reduce body weight and to increase LV weight, these effects were not statistically significant (Table 1). Body weights were lower in the SHR compared with age-matched WKY (Table 1).
LV Cavity Dimensions
Consistent with the development of LV remodeling (LV dilatation), at 21 to 22 months of age, SHR developed an increased LVEDr0 and LV volume (LVV0) (Figure 1). Chronic ISO administration to SHR produced a marked increase in LVED internal radius and LVEDr0 to values comparable with those noted in SHR at 21 to 22 months of age (Figure 1). Similarly, chronic ISO administration produced a right shift in the LV diastolic P-V relations and an increase in LVV0 in SHR, to values similar to those noted in 21 to 22 months old SHR (Figure 1). Although there was a trend for ISO administration to increase LV cavity size in WKY, this effect was statistically insignificant (Figure 1).
As determined using echocardiography in intact anesthetized rats, ISO administration to SD rats failed to increase LVED internal diameters (ISO, 0.799±0.018 cm; control, 0.807±0.014 cm). Although filling dimensions were not increased in intact SD rats after 1 month of ISO treatment, obvious LV remodeling was noted. Both V0 (ISO, 0.28±0.02 mL; control, 0.23±0.01 mL, P<0.01) and LVEDr0 (ISO, 0.36±0.01 cm; control, 0.30±0.02cm; P<0.01) were increased in ISO-treated SD rats.
LV Relative Wall Thickness
Consistent with concentric LVH, 12-month-old untreated SHR had increases in LVED relative wall thickness (h/r) as determined at controlled filling pressures (Figure 2). In contrast, consonant with deleterious LV remodeling, at 21 to 22 months of age LVED h/r was diminished in SHR (Figure 2), despite the marked increase in LV weight noted at this time (Table 1). Similarly, although ISO administration to SHR from 7 to 12 months of age produced a further increase in LV weight to values comparable with those obtained in 21- to 22-month-old SHR (Table 1), ISO-treated SHR developed a reduced LVED h/r in comparison to untreated age-matched SHR (Figure 2). Despite the extent of the increment in LV weight noted in ISO-treated SHR and in 21- to 22-month-old SHR in comparison to the WKY control groups (Table 1), LVED h/r values in ISO-treated SHR and 21- to 22-month-old SHR were only comparable with those of WKY controls (Figure 2). ISO administration to WKY rats failed to influence LVED h/r (Figure 2).
LV Systolic Chamber Performance and Intrinsic Myocardial Contractility
Untreated SHR at 12 months of age had an increased LV systolic chamber performance (E) but similar intrinsic myocardial contractility (En) compared with untreated WKY controls (Figure 3). In contrast, 21- to 22-month-old SHR and ISO-treated SHR developed a reduced LV systolic chamber function (E) compared with untreated SHR, but no change in intrinsic myocardial contractility (En) (Figure 3). ISO administration to WKY also tended to reduce systolic chamber performance, but not intrinsic myocardial contractile function; however, this effect failed to reach statistical significance (Figure 3).
Pathological Score and Collagen
Neither untreated nor ISO-treated SHR at 12 months of age had evidence of significant myocyte necrosis (Table 2). However, 21- to 22-month-old SHR had evidence of myocyte necrosis (Table 2). Untreated 12-month-old SHR had an increase in myocardial hydroxyproline concentrations, an effect that was considerably enhanced in 21- to 22-month-old SHR and by the administration of ISO to SHR (Table 2). ISO administration to WKY produced no significant change in either hydroxyproline concentrations or pathological score (Table 2).
As a consequence of a decrease in the percentage of myocardial collagen soluble to cyanogen bromide digestion (SHR, 24±2%; WKY, 36±4%), untreated SHR at 12 months of age had an increase in insoluble (crosslinked), but not soluble (noncrosslinked), collagen concentrations (Table 2). In contrast, at 21 to 22 months of age, SHR had marked increases in both insoluble and soluble collagen concentrations (Table 2). Moreover, following ISO-induced increases in myocardial collagen concentrations, both the insoluble and the soluble myocardial collagen concentrations were increased in SHR (Table 2). ISO administration to WKY produced no significant effect on either soluble or insoluble myocardial collagen concentrations (Table 2).
The principal finding of the present study is that chronic administration of ISO to SHR, without mediating tissue necrosis or intrinsic myocardial contractile dysfunction, promotes the progression from compensated concentric LVH to LV dilatation and pump dysfunction in hypertension. Chronic β-AR activation in SHR produced a further increment in LV weight, which was accompanied by an increased chamber volume, wall thinning, increases in myocardial collagen concentrations, and decrements in chamber performance but not in intrinsic myocardial contractility. This deleterious LV geometric remodeling mediated by chronic β-AR activation, was compatible with changes noted during LV decompensation in old SHR (22 months).
LV dilatation is perceived to be secondary to a number of sympathetic-mediated intrinsic myocardial functional changes (alterations in cell signaling and calcium handling, and effects mediated through necrosis and apoptosis), all of which are thought to contribute to initiating pump dysfunction and, subsequently, producing chamber remodeling.12 In contrast, in the present study, although baseline intrinsic myocardial contractility (load independent) was maintained, pump dysfunction was noted. The reduction in systolic pump function was attributed to adverse chamber remodeling. The present study therefore suggests that chronic β-AR activation in hypertensive LVH can mediate pump dysfunction through primary rather than secondary effects on chamber remodeling. Hence, our data support the concept originally proposed by Cohn23—and subsequently substantiated by data obtained in human13 and animal14 studies—that cardiac dilatation is a precursor of LV dysfunction. Although previous data have indicated that nonnecrotic doses of ISO in normotensive rats are able to produce LV dilatation and pump dysfunction, no distinction was made in this study15 between myocardial and geometric effects on pump abnormalities.
Increases in LV cavity size with a proportionately greater increase in LV weight, and hence an enhanced relative wall thickness (as previously shown to follow chronic administration of nonnecrotic doses of ISO to rats),15 may not represent LV remodeling consistent with advanced heart failure. In advanced heart failure, LV cavity size increases out of proportion to the growth of the LV wall, and a resultant decrease in absolute or relative wall thickness is thought to contribute to pump dysfunction.23–25 In the present study, the β-agonist–mediated reduction in wall thickness, despite mediating further increases in LV weight in SHR, represents the first evidence to indicate that chronic β-AR activation in LVH produces LV dilatation together with wall thinning without inducing myocyte necrosis.
If, as the data from the present study indicate, LV remodeling is not secondary to intrinsic myocardial contractile dysfunction, what are the potential mechanisms responsible for geometric changes? Chronic β-AR activation may mediate LV remodeling indirectly through alterations in hemodynamic loads. Chronic β2-AR activation may induce sustained increments in LV preload through vasodilatation. However, LV filling dimensions remained unchanged in intact normotensive rats after a relatively short period (1 month) of ISO administration, despite early evidence of LV remodeling (an increased V0 and LVEDr0) in these animals. Moreover, chronic vasodilator therapy with the potent nonspecific vasodilator hydralazine has previously been shown to prevent rather than encourage the development of LV dilatation and wall thinning in SHR.16 Chronic β1-AR activation could also lead to LV remodeling through an enhanced afterload secondary to increases in contractility and heart rate. However, we were unable to detect alterations in SBP, diastolic BP, pulse pressure, or heart rate after chronic ISO administration to SHR.
Irrespective of whether β-AR–mediated LV dilatation occurs as a consequence of direct myocardial effects, or load-induced effects, chamber remodeling may follow either cardiomyocyte lengthening (mainly through the addition of sarcomeres26) or changes in the interstitium (alterations in the quantity and/or quality of myocardial collagen.15,25,27–29 In the present study, chronic ISO administration resulted in marked increases in myocardial collagen concentrations and consequent increments in both soluble (noncrosslinked) and insoluble (crosslinked) myocardial collagen concentrations. This is in contrast to the myocardial collagen change noted in age-matched untreated SHR, in which although myocardial collagen concentrations were increased, the enhanced crosslinked properties of collagen (decreased solubility) resulted in increments in the crosslinked, but not in the noncrosslinked portion of myocardial collagen. The ISO-mediated increase in noncrosslinked myocardial collagen was synonymous with interstitial changes noted in older SHR and may have contributed to progressive LV dilatation associated with reductions in wall thickness. Noncrosslinked myocardial collagen is susceptible to degradation by collagenases, thus contributing to side-to-side slippage.15,28,29 Although in the present study, we have been able to show that some interstitial changes may contribute to ISO-mediated LV remodeling, we have not examined whether alterations in cardiomyocyte length,26 the activity and expression of matrix metalloproteinases and their tissue inhibitors,28,29 or apoptosis12 contribute to the remodeling process.
Despite marked ISO-induced effects on LV remodeling in SHR, the same dose of ISO produced only a trend for LV remodeling in control WKY. Although this finding does not affect the interpretation of our data (evaluations of the actions of ISO on WKY was not a primary goal of the present study), these data suggest that animals with LVH are more sensitive to the detrimental effects of chronic β-AR activation. As LV tissue noradrenaline and angiotensin II concentrations are reported to be increased in SHR compared with WKY,30 it is likely that SHR are predisposed to cardiac remodeling. However, without more extensive studies with a longer duration of ISO administration, this hypothesis remains speculative.
In conclusion, we have been able to show that chronic administration of a β-AR agonist to rats with compensated concentric LVH promotes the development of pump dysfunction through a mechanism which is independent of intrinsic myocardial contractility changes and necrosis, but associated with interstitial and chamber remodeling. These data suggest that adrenergic activation contributes to the progression from compensated concentric LVH to pump dysfunction in hypertension through a novel mechanism (primary effects on cardiac remodeling).
The present study indicates that excessive adrenergic activation in hypertension could initiate the progression from concentric cardiac hypertrophy to pump dysfunction. Furthermore, the results of the present study indicate that the effect of β-AR activation on pump dysfunction is primarily through chamber dilatation rather than decreased intrinsic myocardial contractility. These results contribute to our understanding of the mechanisms involved in the progression from compensatory hypertrophy to heart failure in hypertension and therefore bear relevance to future clinical management strategies to prevent or reverse cardiac remodeling.
This research was supported by the Medical Research Council of South Africa, the University Research Council of the University of the Witwatersrand, and the H.E. Griffin Charitable Trust. We are grateful to Adcock Ingram for the generous donation of Imuprel for this study.
- Received August 6, 2002.
- Revision received August 26, 2002.
- Accepted January 3, 2003.
Grossman W, Jones D, McLaurin KP. Wall stress and patterns of hypertrophy in the human left ventricle. J Clin Invest. 1975; 58: 56–64.
Spann JF Jr, Buccino RA, Sonnenblick EH, Braunwald E. Contractile state of cardiac muscle obtained from cats with experimentally produced ventricular hypertrophy and heart failure. Circ Res. 1967; 21: 341–343.
Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feldman AM, Francis GS, Ganiats TG, Goldstein S, Gregoratos G, Jessup ML, Noble RJ, Packer M, Silver MA, Stevenson LW. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure. Circulation. 2001; 104: 2996–3007.
Hasking GJ, Esler MD, Jennings GL, Burton D, Korner PT. Norepinephrine spillioner to plasma in patients with congestive heart failure: evidence of increased overall and cardiorenal sympathetic nervous activity. Circulation. 1986; 73: 615–621.
Sabbah HN. The cellular and physiologic effects of β blockers in heart failure. Clin Cardiol. 1999; 22: 16–20.
Woodiwiss AJ, Tsotetsi OJ, Sprott S, Lancaster EJ, Mela T, Chung ES, Meyer TE, Norton GR. Reduction in myocardial collagen cross-linking parallels left ventricular dilatation in rat models of systolic chamber dysfunction. Circulation. 2001; 103: 155–160.
Tsotetsi OJ, Woodiwiss AJ, Netjhardt M, Qubu D, Brooksbank R, Norton GR. Attenuation of cardiac failure, dilatation, damage, and detrimental interstitial remodeling without regression of hypertrophy in hypertensive rats. Hypertension. 2001; 38: 846–851.
Norton GR, Tsotetsi J, Trifunovic B, Hartford C, Candy GP, Woodiwiss AJ. Myocardial stiffness is attributed to alterations in cross-linked collagen rather than total collagen or phenotypes in spontaneously hypertensive rats. Circulation. 1997; 96: 1991–1998.
Norton GR, Tsotetsi OJ, Woodiwiss AJ. Hydrochlorothirazide improves ventricular compliance and thus performance without reducing hypertrophy in renal artery stenosis in rats. Hypertension. 1993; 21: 638–645.
Woodiwiss AJ, Norton GR. Exercise-induced cardiac hypertrophy is associated with an increased myocardial compliance. J Appl Physiol. 1995; 78: 1303–1311.
Weber KT, Janicki JS, Shroff SG, Pisk R, Chen RM, Bashey RI. Collagen remodeling of the pressure-overloaded, hypertrophied non-human primate myocardium. Circ Res. 1988; 62: 757–765.
Cohn JN. Structural basis for heart failure: ventricular remodeling and its pharmacological inhibition. Circulation. 1995; 91: 2504–2507.
Mann DL. Mechanisms and models in heart failure: a combinational approach. Circulation. 1999; 100: 999–1008.
Li YY, Feng Y, McTiernan C, Pei W, Moravec CS, Wang P, Rosenblum W, Kormos RL, Feldman AM. Down-regulation of matrix metalloproteinases and reduction in collagen damage in the failing heart after support with a left ventricular assist support device. Circulation. 2001; 104: 1147–1152.
Mann DL, Spinale FG. Activation of matrix metalloproteinases in the failing human heart: breaking the tie that binds. Circulation. 1998; 98: 1699–1702.