Investigations on the antimalarial activity of alkoxylated and hydroxylated chalcones 3

SECTION SIX MODE OF ANTIMALARIAL ACTIVITY OF CHALCONES 6. MODE OF ANTIMALARIAL ACTIVITY OF CHALCONES 6.1 Introduction The mode of action of chalcones in malaria remains uncertain. As described in the Introduction (Section 1.3.1 ) the antimalarial activity of chalcones came to the fore, in part, due to a database search of compounds that could fit the active site of the malarial cysteine proteases (falcipains). 72 In silico simulations indicated that several antimalarial chalcones have an excellent fit onto the enzyme active site. 72 Subsequent investigations explored this connection but so far, the correlation between antimalarial activity and inhibition of falcipains has not been convincing. 74 In this section, the likelihood of chalcones acting on targets in the digestive vacuole of Plasmodium is investigated. The decision to focus on events involving hemoglobin degradation in the digestive vacuole is prompted largely by earlier findings that chalcones are potential cysteine protease inhibitors. The aspects investigated are the effects of selected chalcones on the enzymatic breakdown of radiolabelled methemoglobin by a crude plasmodial extract, the hydrolysis of a fluorogenic substrate by recombinant plasmodial cysteine protease (falcipain-2) and binding to hematin. The results obtained from these experiments are presented and discussed in the following paragraphs. 6.2 Materials and Methods The following chemicals were purchased from Sigma Chem. Co. (MO, USA): [14C] methemoglobin (30 mCi/g, 0.17 mg/ml), pepstatin A, L-trans-epoxysuccinylleucylamido-(4-guanidino)-butane (E64), benzyloxycarbonyl-Phe-Arg-7-amino-4methylcourmarin (Z-Phe-Arg-AMC), porcine hematin (ferriprotoporphyrin IX hydroxide), chloroquine diphosphate. Other reagents were of analytical grade. 99 6.2.1 Degradation of [14C] methemoglobin by extracts of P. falciparum (K1) P. falciparum (K1)-infected erythrocytes were synchronized at least twice using 5% sorbitol to yield cultures of trophozoites at approximately 20% parasitemia. The infected cells were harvested, washed with PBS, treated with 0.1% (w/v) saponin in PBS, washed × with ice-cold PBS and centrifuged (1000 g, 10 min, 4oC). Water was added to the resulting pellet to give a solution which was subjected to freezethaw cycles, centrifuged at 13,000 g, 10 min, 4oC to give supernatant containing crude parasite extracts (lysate). Following a reported method 125, aliquots of the lysate (25 µl, estimated to contain 1-1.5 µg protein /µl) and test compound in DMSO (10 µl) was added to 50 µl sodium acetate (0.1 M, pH 6.0) and sufficient distilled water to give a final volume of 100 µl. The final concentration of test compound was 100 µM. After h of incubation at 37oC, µl of [14C] methemoglobin (30 mCi/g, 0.17 mg/ml) was added. Incubation was continued for another h after which additions of bovine serum albumin (50 µl, mg/ml) and 50% w/v trichloroacetic acid (100 µl) were made. The samples were incubated for another 30 min, 4oC before centrifugation at 13,000 g, 4oC. Aliquots (100 µl) of the supernatant were taken, added to ml of scintillation fluid and radioactivity was determined with a scintillation counter. Controls consisted of samples processed without test compound ([14C] methemoglobin and lysate) and samples containing only [14C] methemoglobin in the incubation mixture. The same volume of DMSO used to deliver test compound was added in both controls. 6.2.2 Inhibition of falcipain-2 Experiments on the inhibition of falcipain-2 were not carried out by the candidate but were tested in Dr Philip Rosenthal’s laboratory in the School of Medicine, University of California, San Francisco, USA. Briefly, soluble parasite 100 extracts containing falcipain-2 were incubated with dithiothreitol (10 mM) and test compound in sodium acetate buffer (0.1 M, pH 5.5) for 30 at room temperature, after which the fluorogenic substrate Z-Phe-Arg-AMC was added to give a final concentration of 50 µM. 126 Cleavage of the substrate caused an increase in fluorescence (due to the free coumarin) which was monitored for 30 at excitation and emission wavelengths of 380 nm and 460 nm respectively. An inhibitor of falcipain will cause fluorescence to decrease. The rates of hydrolysis of the substrate in the presence and absence of test compound were determined. Fluorescence from control cuvettes containing only substrate or compound was also monitored. Test compounds were monitored at different concentrations to give IC 50 values from rates versus concentration plots. Chalcones which inhibited falcipain-2 with IC50 values of ≤ 10 µM were further investigated for the appearance of an abnormal food vacuole in P. falciparum trophozoites. Ring-stage parasites were incubated with the test compound for 24 h, after which Giemsa-stained smears were prepared and evaluated microscopically for abnormal morphology. 6.2.3 Effect on Soret band of hematin The interaction of chalcones with porcine hematin was investigated by monitoring changes in the Soret band of hematin, following a reported method. 25 Appropriate aliquots of hematin (2 mM stock solution in 0.1 M NaOH), test compound (stock solutions of mM or 0.2 mM in methanol) were added to a cuvette (1 ml) containing 43% methanol in sodium acetate buffer (10 mM, pH 5.5), to give final concentrations of 14 µM hematin and 2-128 µM test compound. The solution was vortexed for 15 s and the spectrum was collected from 250 -650 nm. Under these conditions, the Soret band of hematin was observed at 400 nm. Some of the test 101 compounds have strong absorbance in the range of 380-420 nm at concentrations > 100 µM. Thus correction for background absorbance was made using as blank, a solution containing test compound in buffer with no hematin. Chloroquine (2-128 µM) was used as a positive control. The decrease in Soret band absorbance was expressed as a % of the control absorbance obtained in the absence of test compound: % Decrease in absorbance = (Absorbance Control – Absorbance test compound ) / Absorbance Control × 100 6.3 Results 6.3.1 Effects on the enzymatic activity of a crude plasmodial extract catalyzing the breakdown of radiolabelled methemoglobin A crude plasmodial extract was prepared by saponinizing P. falciparum infected erythrocytes to release the intraerythrocytic parasites. The extract has enzymatic activity and catalyzes the breakdown of [14C] methemoglobin, an oxidized and denatured derivative of hemoglobin, into smaller radiolabelled peptide /amino acid fragments that are recovered in the supernatant. Pepstatin A (a specific inhibitor of aspartate protease) and E64 (a specific cysteine protease inhibitor) at 100 µM inhibited the enzymatic activity of the extract to about the same extent (32-38%) A search of the literature indicated that there were no reports of the inhibitory activities of these compounds on a plasmodial extract prepared from P. falciparum for comparison with the present results. However, Pandey and coworkers 127 had investigated the effects of pepstatin A and E64 on the enzymatic activity of a crude P. yoelii extract. They found that the breakdown of radiolabelled methemoglobin was reduced by 71% and 26% in the presence of pepstatin and E64 (both at 100 µM) respectively. Noting that P. yoelii is a murine strain, the inhibitory effects of pepstatin 102 A and E64 were investigated using extracts prepared in a similar manner from another murine plasmodia (P. berghei ANKA). This time, the levels of inhibition (21%, 62 % for 100 µM E64, pepstatin A respectively) were comparable to those obtained with P. yoelii. Fifteen of the twenty “actives” were tested for inhibitory activity on the breakdown of methemoglobin at a fixed concentration of 100 µM. Less active chalcones 41 and 125 were also included for comparison (Table 6.1 ). As seen from Table 6.1, varying levels of inhibition are observed among the chalcones. No inhibitory activity was noted for several active members (6, 8, 19, 113) but both the inactives (41, 125 ) inhibited enzymatic activity by 26-38%. On the other hand, maximum inhibition (45 %) was noted for 27 which had the highest in vitro antimalarial activity (IC50 µM) for the present series of chalcones. Overall, no discernible trend is evident from the results. Table 6.1 6.3.2 Effects on falcipain-2 and associated changes in the food vacuole on incubation The same chalcones that were tested for activity on the crude plasmodial extract were investigated for their effects on falcipain-2-catalyzed hydrolysis of the fluorogenic substrate Z-Phe-Arg-AMC. An arbitrary cut-off concentration of 10 µM was used to distinguish between chalcones with inhibitory activity (IC50 < 10 µM) and those without activity. Based on these criteria, chalcones comprising of two inactives (41, 125) and six actives (2, 5, 211, 227, 228, 234) were found to be inhibitory (Table 6.1). Interestingly, the actives are derived from only two classes of ring B substituted chalcones, namely the hydroxylated chalcones (211, 227, 228, 234) and the 103 dimethoxychalcones (2, 5). The 3-quinolinyl chalcone 27 which inhibited methemoglobin breakdown to the greatest extent was not among the falcipain-2 inhibitors. Strongest falcipain-2 inhibitory activity was found in the ethoxychalcone 125 (IC50 1.4 µM) which had weak in vitro antimalarial activity (IC50 39 µM). It is clear that there is no correlation between falcipain-2 inhibition, inhibition of methemoglobin breakdown and in vitro antimalarial activity. Inhibition of falcipain-2 is accompanied by visible changes in the digestive vacuole of Plasmodium, namely the appearance of swollen vacuoles filled with undegraded hemoglobin. These changes were not evident when the chalcones that inhibited falcipain-2 with IC 50 values of < 10 µM were incubated with the ring-stage parasites. A similar finding was reported for some antimalarial phenothiazines that inhibited falcipain-2 but did not cause the expected changes in the food vacuole. 74 The proffered explanation was that the phenothiazines were cytotoxic and had other effects on parasite morphology (cytoplasmic vacuolization) that overshadowed changes in the food vacuole. This may be true for the chalcones as well, but this would be difficult to reconcile with the results of the MTT assay which showed that these chalcones were essentially not cytotoxic at 20 µM (Section 4.3.3). One possibility is that these results reflect the limitations on the transport of these chalcones into the food vacuole that would be necessary before detection of morphological changes becomes evident. That is, the chalcones may be able to inhibit the enzyme when it is isolated (as in an in vitro assay) but may not be able to gain access into the food vacuole readily to cause the anticipated morphological changes. The present series of chalcones are neutral or weakly basic compounds and will not be trapped in the acidic food vacuole in the same way as the more basic aminoquinolines like chloroquine (pKa values = 8.1, 10.2). 6.3.3 Binding to heme 104 Degradation of hemoglobin results in the formation of toxic heme, the disposal of which has been the target of several antimalarial drugs. In the food vacuole, heme is formatted to non-toxic hemozoin. Compounds that interfere with this process are characterized by binding to heme. In this investigation, the interaction of the chalcones with hematin was investigated in an acetate buffer with an apparent pH 5.5 to mimic the acidic pH of the plasmodial food vacuole. 25 The buffer contained a high proportion of methanol (43%) to keep heme in solution and in the monomeric state. Under these conditions, hematin displays a Soret band at 400 nm with a shoulder at 360 nm. When a test compound binds to heme, a decrease in the Soret band absorbance is observed. This is illustrated with chloroquine which was used in this investigation as a positive control. The incubation of hematin with chloroquine (2-128 µM) caused a concentration-dependent fall in the Soret band absorbance (Figure 6.1 ), similar to that reported by other investigators. 25 At the highest concentration (128 µM) of chloroquine, the observed absorbance was approximately 48% of the control Soret band absorbance. That is, chloroquine has reduced Soret band absorbance by 52% at this concentration. 36 chalcones were screened for changes in Soret band absorbance. These included 19 active chalcones (except 234) in Table 6.1 as well as 16 other less active members. Table 6.2 lists the % decrease in the absorbance of the Soret band in the presence of 128 µM test compound. A compound that binds to hematin should cause a large drop in absorbance. Only compounds that decreased absorbance by more than 10% were considered to have an effect on the binding interaction with hematin. 14 such compounds were identified, but only five of these compounds are actives (7, 113, 207, 211, 228 ). The greatest reduction in the Soret band absorbance (64.5%) was observed for the dihydroxychalcone 205 . A concentration-dependent effect was also 105 evident for these compounds, that is increasing concentrations resulted in a greater reduction in the Soret band absorbance. The results show a clear structural trend among chalcones that bind to hematin. Binding is observed mostly among the hydroxylated chalcones, in particular, the 2’hydroxychalcones. There is also a strong preference for naphthalene and pyridine rings among the hematin-binding chalcones. In contrast, quinoline and the usual benzenoid ring A are conspicuously under-represented. Since the structural features that predispose towards binding to hematin are not the same as those associated with good antimalarial activity, it is probable that heme binding does not contribute significantly to antimalarial activity. However, heme binding may account to some extent for the activity of the less active chalcones like the 2’-hydroxychalcones. Figure 6.1 Table 6.2 106 Table 6.1 Effect of antimalarial chalcones on the enzymatic activity of a crude P. falciparum K1 extract using [14C] methemoglobin as substrate, in vitro falcipain inhibition and cell viability. O R' No. A B Substitution on Ring A R KB3-1 cell viability (%) c (SD) 88.8 (17.5) 80.8 (15.0) 77.8 (23.5) 126.1 (19.7) % inhibition of [14C] MetHb breakdown at 100 µM (SD) 10.8 (4.0) 45.1 (16.0) 25.8 (5.3) IC50 (µM) Falcipain inhibition 27 36 2,3,4-trimethoxy 2,4-dichloro 4-trifluoromethyl 3-quinolinyl a 4-fluoro IC 50 (µM) in vitro antimalarial activity b 5.4 3.0 2.0 9.5 29 2,4-dimethoxy 4-trifluoromethyl 2,4-difluoro 2,4-dimethoxy 4-ethyl 3-quinolinyl a 5.9 6.2 2.1 2.4 2.2 99.6 (18.3) 107.8(7.5) 84.0 (20.5) 85.4 (8.2) 82.6 (17.1) 27.2 (12.4) 32.1 (1.9) 30.8 (16.2) ND 2.3 7.8 >10 >10 ND 19 31 38 113 4-methoxy 4-hydroxy 3-quinolinyl 4-fluoro 2,4-dimethoxy 7.0 4.8 14.4 6.4 78.9(16.7) 82.9(15.2) ND 121.6(13.3) ND ND >10 ND ND >10 125 127 4-ethoxy 4-nitro 4-cyano 39.0 540.0 ND ND 25.6 (12.8) ND 1.4 >10 202 203 207 208 211 2,4-dihydroxy 3-quinolinyl a 2,4-difluoro 2-pyridinyl a 2-naphthalenyl a 4-chloro 16.1 16.0 19.7 20.0 12.3 89.3(28.5)* 88.9(22.1)* 96.4(19.3)* ND 100.7(32.0)* ND ND 9.4 (3.7) ND ND ND >10 ND 9.8 214 227 228 4-hydroxy 2-pyridinyl a 3,4-dichloro 4-dimethylamino 16.3 18.4 17.7 112.3 (18.2)* 90.6 (23.3)* 84.8 (32.2)* ND 40.7 (9.7) 39.0 (14.3) ND 2.4 9.7 234 2-hydroxy 4-chloro 12.9 ND 14.7 (8.6) 2.3 0.27 4.0 80.0 104.1 (17.5) ND ND ND 38.2 d (16.3) 31.5 d (5.3) ND ND ND Ring B Chloroquine E64 Pepstatin A >10 >10 >10 >10 ND = not done. SD is given in parentheses. a. Phenyl ring A is substituted by heterocyclic or naphthalene ring. b. Inhibition of [3H] hypoxanthine uptake into P. falciparum K1 infected erythrocytes. c. Mean of or more determinations. Compounds are tested at 20 µM except for chloroquine (80 µM) and those marked with * (40 µM). d. % inhibition of extracts prepared from P. berghei (ANKA) infected erythrocytes for E64 and pepstatin A were 24.3 (12.9) and 62.2 (18.7) respectively. 107 C: calcd, 61.64 found, 61.46 H: calcd, 3.45 found, 3.46 Cl: calcd, 23.95 found, 24.33 229 67-75.8 (C) 10.0 230 174.2-175.0 (A) 175-177f 16.1 C: calcd, 80.33 found, 80.58 H: calcd, 5.40 found, 5.54 231 176-178 (A) 10.0 C, H, Cl: calcd, 23.95 found, 24.33 232 69-71 (A) 25.0 C, H, N: calcd, 5.24 found, 5.39 233 196-198 (A) 32.6 234 149-150 (A) 48.1 C, H, N: calcd, 5.09 found, 5.18 C: calcd, 69.76 found, 70.04 H: calcd, 4.30 found, 4.75 Cl: calcd, 13.55 found, 13.28 235 114-115 (A) 120-121f 44.6 C, H 236 89-91 (A) 91-92 f 35.4 C, H 237 111-113 (A) 25.0 238 [...]... DISUBSTITUTION d 3 4 6 11 12 13 27 28 35 36 40 128 129 130 131 132 133 134 3. 3776 3. 3466 3. 9 830 3. 10 73 3. 430 1 3. 84 93 3. 030 0 3. 0749 3. 0168 3. 47 13 3.2579 3. 8691 4.0709 3. 4016 4.1201 3. 9210 3. 17 13 3.4658 3. 85 2.59 3. 31 2. 43 2.92 3. 45 2 .31 2 .31 2 .34 2.57 4 .31 2.71 2.17 3. 73 3.14 3. 14 3. 14 2.42 9.74 10.05 9.27 9.99 9.22 9.68 10. 23 10. 23 9 .37 8.77 11.27 8.79 9 .37 9.74 9.25 9.25 9.25 8.76 0 .38 7 0 .39 2 0 .37 6 0 .39 5 0 .38 9... 035 93 -.178 088 93 370 b 000 93 361 b 000 93 1.000 93 b Pearson correlation is significant at the 0.01 HOMO 052 647 80 0 13 905 93 391 b 000 93 386 b 000 93 305 b 0 03 93 336 b 001 93 0 53 615 93 1.000 93 LUMO -.0 13 911 80 065 533 93 1 23 241 93 261 a 012 93 091 38 7 93 125 232 93 -.002 988 93 7 13 b 000 93 1.000 93 DISUBSTITUTION -.054 635 80 31 5 b 002 93 125 234 93 321 b 002 93 202 0 53 93 179 086 93. .. 3. 2568 3. 4794 3. 2051 3. 2 532 3. 2618 3. 6945 4 .36 89 2.4142 3. 8824 3. 7707 3. 833 1 4.75 4.60 4.06 3. 46 3. 18 4.20 3. 06 3. 06 3. 67 3. 09 3. 34 3. 32 3. 89 4.04 4. 03 4.48 2.92 4 .35 2.49 3. 44 3. 07 3. 04 3. 04 3. 30 4.58 4.04 3. 16 3. 65 2.90 3. 32 10.15 9.12 8.65 8.17 9 .37 9.07 9.62 9.62 8.60 8.76 9.44 8.15 8. 63 8.92 8.65 9.12 8.75 9. 83 7.68 7.55 8.14 9.00 9.00 7.54 8.51 8. 03 8.76 7.99 8. 13 8.82 0 .38 9 0 .38 7 0 .37 5 0 .39 0 0 .39 5... 0 .37 8 0 .38 1 0 .38 4 0 .38 5 0 .38 0 0 .37 6 0 .38 2 0 .38 1 0 .36 9 0 .38 9 0 .38 3 0 .38 5 0 .38 5 0 .37 8 0 .38 1 0 .38 7 0 .38 6 0 .39 0 0 .39 5 0 .38 8 0 .37 5 0 .38 3 0 .38 7 0 .38 3 0 .38 2 0 .38 7 0 .38 2 0 .38 0 0 .37 7 0 .37 6 2.4 532 2.4176 2.4611 2.4901 2.5066 2.5051 2.4574 2.4945 2.4971 2.5260 2.4774 2.4842 2.5169 2.48 03 2.4500 2.4482 2.4586 2.41 03 2.48 03 2.4279 2.4479 2 .39 11 2.4645 2 .38 71 2.4570 2.4 231 2.4247 2.4506 2 .37 77 2.4488 2.4176 2 .38 12... DISUBSTITUTION d 3. 98 3. 52 3. 52 3. 52 2. 73 3 .31 3. 31 4.89 3. 69 2.55 2.95 4.11 2.97 4. 23 2.81 8.75 7.55 7.55 7.55 7.68 7.52 7.52 8.91 7.57 7.67 7.07 8.04 8 .35 8.04 7.06 0 .38 1 0 .38 1 0 .38 1 0 .38 1 0 .38 5 0 .38 3 0 .38 1 0 .38 3 0 .36 9 0 .38 5 0 .38 2 0 .38 1 0 .38 0 0 .38 1 0 .38 1 2.4566 2.4145 2.4128 2.4144 2. 431 9 2.4176 2.41 73 2.4977 2.4401 2.4278 2 .39 41 2. 439 5 2.4624 2.4400 2 .38 52 2.4274 2 .37 84 2 .37 91 2 .37 84 2 .39 53 2 .37 99 2 .37 98... 10. 53 found, 10.67 31 6.1 132 (C18H17O4F = 31 6.1111) 37 4.15 03 (C24H22O4 = 37 4.1518) 33 4.1017 (C18H16O4F2 = 33 4.1017) 34 3.1066 (C18H17O6N = 34 3.1055) 36 6.0427 (C18H16O4Cl2 = 36 6.0425) 33 2.0817 (C18H17O4Cl = 33 2.0815) 33 2.0808 (C18H17O4Cl = 33 2.0815) 33 2.0816 (C18H17O4Cl = 33 2.0815) 298.1199 (C18H18O4 = 298.1025) 34 0.1680 (C21H24O4 = 34 0.1675) 33 6. 034 6 (C17H14O3Cl2 = 33 6. 032 0) 16 63. 30 (υC=O) 16 63. 30 (υC=O)... 0 .37 5 0 .39 0 0 .39 5 0 .38 9 0 .38 6 0 .38 2 0 .38 9 0 .39 1 0 .39 2 0 .38 8 0 .38 7 0 .38 6 0 .38 8 0 .38 7 0 .39 2 0 .38 8 0 .38 5 0 .38 4 0 .38 5 0 .38 0 0 .37 6 0 .38 2 0 .38 1 0 .36 9 0 .38 9 0 .38 3 0 .38 5 0 .38 5 2.5687 2.5060 2.5052 2.47 23 2. 532 8 2.5084 2.51 53 2.5062 2.4850 2.4965 2.5228 2.46 53 2.4826 2.4918 2.4887 2.5047 2.4 936 2.5151 2. 431 7 2. 432 5 2.4607 2.4785 2.4715 2.4258 2.4687 2.4684 2.4988 2.44 73 2.4569 2.4892 2.5456 2.48 83 2.4809 2.4414... 0 0 0 1 0 1 0 0 0 A- 13 Compound LOGKW a CLOGP b MR b CHARGES b,c LOGA b LOGV b TDM b HOMO b LUMO b DISUBSTITUTION d 117 135 25 26 33 34 39 121 122 1 23 124 125 126 127 136 201 202 2 03 204 205 206 207 208 209 210 211 212 2 13 214 215 3. 2 539 2.59 3. 15 3. 97 3. 60 3. 57 3. 57 3. 83 5.11 4.57 3. 69 4.18 3. 43 3.85 3. 12 3. 68 4.19 2.65 3. 05 2.77 3. 94 3. 65 1.27 3. 94 1.27 2.65 3. 48 3. 98 2.69 1 .31 2.69 8.00 7.52 8.02... account for the antimalarial activity of the chalcones Finally, there is little evidence of significant binding to hematin among the active chalcones Decreases in the Soret band of hematin were observed mainly for 2’-hydroxychalcones that have naphthalene and pyridine rings, and to a lesser extent among alkoxylated chalcones and chalcones that have other types of Ring A The planarity of naphthalene and pyridine... -1 .32 9 1.054 0.629 0 .31 7 -0.294 -0.482 -0. 231 -0. 139 -0.462 -3. 095 0.000 Compound LOGKW a CLOGP b MR b CHARGES b,c LOGA b LOGV b TDM b HOMO b LUMO b DISUBSTITUTION d 41 1 2 5 7 8 29 30 101 102 1 03 104 105 106 107 108 109 110 19 22 23 31 32 38 111 112 1 13 114 115 116 4.6529 2.8421 3. 135 1 3. 54 13 3.29 83 3.4605 3. 25 03 3.2924 3. 8564 3. 4471 3. 7266 5.2297 3. 6592 3. 1729 4.0246 4.5598 3. 54 83 4.2297 3. 2470 3. 2568 . alkoxylated chalcones and chalcones that have other types of Ring A. The planarity of naphthalene and pyridine rings may favor π-π interactions between the electron clouds of these rings and the porphyrin. substituted chalcones, namely the hydroxylated chalcones (211, 227, 228, 234 ) and the 104 dimethoxychalcones (2, 5). The 3- quinolinyl chalcone 27 which inhibited methemoglobin breakdown to the greatest. µ M (Section 4 .3. 3). One possibility is that these results reflect the limitations on the transport of these chalcones into the food vacuole that would be necessary before detection of morphological