CHAPTER 4 CHAPTER 4 by gyvwpsjkko

VIEWS: 462 PAGES: 10

									Inhibitors of HIV reverse transcriptase (RT) are important drugs for the treatment of acquired
immuno-deficiency syndrome (AIDS). One approach to identify novel inhibitors of HIV-1-RT is the
screening of natural compounds - these represent a valuable source of potential anti HIV compounds.
In this study, several crude plant extracts (Table 11) were assessed for their ability to inhibit
recombinant HIV-1 reverse transcriptase in an ELISA-based technique. Following this initial
inspection of RT inhibition by crude extracts, Vernonia oligocephala and Gunnera perpensa were
further investigated.


V. oligocephala is a herbaceous perennial plant that is widespread in the grassland regions of South
Africa (Van Wyk et al., 1997). Roots and twigs are commonly used for medicinal treatment. Infusions
are taken to treat abdominal pain and colic, rheumatism, dystentery and diabetes (Watt & Breyer-
Brandwijk, 1962; Pujol, 1990; Hutchings, 1996). The family Asteraceae (Compositae), to which V.
oligocephala belongs, is known for the presence of terpenes especially sesquiterpene lactones (Koul et
al., 2003). Other types of natural products also present include flavones, triterpenes and sterols,
acetylenes, vernolic acid and similar compounds (Bohlmann et al., 1978). The chemistry and
medicinal attributes of Vernonieae has been reviewed by Harborne & Williams (1977), Hedberg et al.
(1982), Bohlmann & Jakupovic (1990), and Johri & Singh (1997). Phytochemical studies by
Bohlmann et al. (1984) resulted in the isolation of triterpenes, such as lupeol, and sesquiterpene
lactones, such as 17,18-epoxyvernonatolide, from V. oligocephala.
                                                                                         CH2


      17,18-Epoxyvernonatolide                                       Lupeol       H3C


                         O
                                                     O
                                       O                                    CH3    CH3
                                                     CH3                                          CH3

                                 CH3       O
                                                                                         CH3
                                               OAc
                   H3C       O                                 HO
                                   O
                                                                    H3C    CH3
                                       O



Crude V. oligocephala extracts showed an inhibition of 28% with the initial screen. Fractionation of
crude V. oligocephala leafs-and-branches extract delivered fractions with no inhibition of HIV RT
(section 3.3). Plant-derived terpenic RT-inhibitors that have been reported are fulvoplumierin (Tan et
al., 1991), 1-β-hydroxyaleuritolic acid-3-ρ-hydroxybenzoate (Pengsuparp et al., 1995), and nigranoic
acid (Sun et al., 1996). Even though Asteraceae family seem to be rich in terpenes, V. oligocephala
does not seem to contain RT inhibitors – all be it terpene or non-terpene related. Lupeol and ursolic
acid are triterpenoids previously isolated from members of the Asteraceae family (Oliveira & Salatino,
2000; Chaturvedula et al., 2002; Li et al., 2003; Chaturvedula et al., 2004). These two compounds
have been reported to inhibit HIV in vitro, but targeted the protease enzyme (Ma et al., 1998; Mengoni
et al., 2002). It is therefore still possible that V. oligocephala could contain compounds with the ability
to inhibit HIV replication, but by targeting steps in the virus’s replicative cycle other than reverse
transcription.


G. perpensa is a vigorous perennial herb that grows in marshy areas and along stream banks in the
southern, eastern and northern parts of South Africa and northwards into tropical Africa (Van Wyk,
1997). Infusions or decoctions of the thick, fleshy rhizomes is used to induce or augment labour, as an
antenatal medication to tone the uterus, to assist in the expulsion of the placenta, to treat stomach
trouble, rheumatic fever, swellings, menstrual pain and stomach bleeding (Watt & Breyer-Brandwijk,
1962; Mendes, 1978; Pujol, 1990; Kaido et al., 1994; Hutchings, 1996; Kaido et al., 1997). It is also
applied externally for the dressing of wounds and sporiasis (Watt & Breyer-Brandwijk, 1962). The
major constituent of the G. perpensa rhizomes appears to be Z-venusol, a phenylpropanoid glucoside
(Khan et al., 2004). Minor components isolated include pyrogallol, succinic acid, lactic acid, and the
trimethyl ether of ellagic acid glucoside.

                                      HO

                                              O              Z-Venusol
                                 HO
                                                      O

                                      HO       O
                                                     O                    OH




Crude extracts from leaves, stems, and rhizomes of G. perpensa significantly inhibited the enzyme by
up to 97% (section 3.2). The fractionation of crude G. perpensa rhizome extract lead to the isolation of
a significantly active fraction (section 3.4). No conclusion could be made about the structure and
identity of the active principle in the crude extracts of G. perpensa. It is difficult to speculate which
type of compound might be responsible for the RT inhibitory activity of G. perpensa extracts. Cos et
al. (2004) indicated that flavonoids might be potential HIV RT inhibitors. These compounds also have
the ability to inhibit viral entry, integration of the proviral genome, and its consequent transcription.
The activity of various flavonoids, against several viruses in cell culture and in animal models, has
been demonstrated by Musci and Prágai (1985).


Baicalein (Animashaum et al., 1993), quercetin (Patel, et al., 1993), myricetin (Nakashima et al.,
1996) and quercetagenin (Kashman et al., 1992) have been shown to inhibit RTs of certain
retroviruses, including Rauscher murine leukaemia virus and HIV, in vitro (Ono et al., 1989; Ono et
al, 1990). Comparative studies with other flavonoids revealed that the presence of both the double

                                                                                      OH

                   Baicalein                                       Quercetin                OH

             HO              O
                                                           HO             O


             HO
                                                                                 OH
                      OH     O
                                                                    OH    O


                                        OH                                             OH
                   Myricetin                  OH
                                                                 Quercetagetin              OH



             HO             O                               HO             O
                                              OH


                                   OH                       HO                    OH

                      OH    O                                        OH    O




bond between positions 2 and 3 of the flavonoid pyrone ring, and the three hydroxyl groups introduced
on positions 5, 6 and 7 (i.e. baicalein) were a prerequisite for the inhibition of RT activity.


The removal of the 6-hydroxy group of baicalein required the introduction of three additional hydroxyl
groups at positions 3, 3’ and 4’ (quercetin). Quercetagetin contains the structures of both baicalein and
quercetin, whereas myricetin has the structure of quercetin with an additional hydroxyl group on the 5’
position. The inhibition of RT by baicalin is highly specific, while quercetin and quercetagetin are
strong inhibitors of DNA polymerase β and DNA polymerase I respectively. Myricetin was a potent
inhibitor of both DNA polymerase α and DNA polymerase I (Vlietinck et al., 1998).


Toxicity studies of G. perpensa crude extracts performed on a lymphoid cell line (CEM.NKR.CCR5)
and peripheral blood mononuclear cells (PBMCs) revealed that, among the three different crude
extract concentrations tested, a concentration of 5 µg/ml was the least toxic to the cells (Tables 14 and
15, section 3.5.1.). Since these two cell types proved to possess sufficient amounts of the correct co-
receptor (CCR5) for successful HIV-1 subtype C entry (Figures 37 and 38, section 3.5), they were
used for anti-HIV studies: extracts were introduced to chronically and acutely HIV-1 infected cells at
5 µg/ml to assess their ability to inhibit HIV-1 replication in vitro. AZT was included as a control drug
in all cases at 10 µM.
Cell viability was assessed, using the XTT tetrazolium dye, as an indirect indication of viral
replication: the assumption being that viral replication would lead to a decrease in cellular viability;
whereas inhibition of viral replication would be reflected by an increase in cellular viability when
compared to untreated infected cells. A comparison of two tetrazolium dyes lead to a preference for
XTT over MTT since the former dye did not require a solubilization step, which in turn could lead to
complications (such as incomplete dissolving of the formazan crystals) that could result in ambiguous
results (Scudiero et al., 1988; Nargi & Yang, 1993). Also, overall absorbance readings using XTT are
higher than with MTT, without any loss in sensitivity. Detecting the HIV-1 core-protein, p24, in
culture supernatants, directly monitored virus replication: a decrease in p24 levels would indicate that
viral replication has been restrained.


According to Marozsan et al. (2004), endogenous RT activity, not p24 content, is the best surrogate
measure of infectious HIV-1 titer in cell-free supernatant. Thus, RT activity would have been a more
accurate reflection of the effects of crude extracts on HIV replication in vitro. Unfortunately no RT
activity could be detected in the culture supernatants – not even in the infected control samples. The
levels of RT in the supernatants could have been below the detection limit of the kit used. As
discussed earlier, anti-HIV drug-screening protocols have a range of possible incubation times. In this
study, cells were successfully infected in 4 days as indicated by detecable p24 levels (>10 pg/ml) and a
50% decrease in cell viability. Infection with a higher concentration of virus could have lead to
increased amounts of viral progeny, and hence increased p24 levels and possibly detectable RT.
Incubation over an extended period of time would have the time for viral replication, leading to the
detection of higher amounts of p24 and RT.


Another possibility could be the presence of PEG in the RT reaction solution: PEG offers a convenient
way of pelleting HIV particles from culture supernatants. The manufacturer clearly stated that, after
pelleting the virus from supernatant with PEG, the utmost care should be taken to remove all visible
traces of PEG from the pellet. Although considerable care was taken to remove all visible trances of
PEG from the viral pellet, small amounts might still have been present and which in turn could have
inhibited the RT reaction. PEG could have been eliminated from the protocol by pelleting viral
particles with an ultracentrifuge, instead of with the aid of PEG.


It was expected that, since the crude extracts from G. perpensa could inhibit a recombinant HIV-1
reverse transcriptase in a direct enzyme assay, these crude extracts would also inhibit HIV-1
replication in vitro to some degree. Unfortunately, none of the G. perpensa crude extracts seemed to
significantly inhibit HIV-1 replication in either chronically- or acutely-infected cells in vitro (Figures
40-45, section 3.5.2.). There could be some possible explanations for this. Firstly, the concentration of
the active principle(s) at 5 µg/ml of crude extract could have been below any effective concentration,
thereby not affecting HIV replication to a noticeable degree. Unfortunately, the concentration of crude
extracts could not be increased since higher concentrations proved to be toxic to the cells (Tables 14
and 15, section 3.5.1.). Secondly, an in vitro cell-system is far more complex than a direct enzyme
assay. The active principle(s) would have to face a multitude of metabolic barriers before reaching its
target, the reverse transcriptase enzyme. By that time the active principle could have been metabolised
by cellular enzymes to a lesser- or non-active compound. Thirdly, in the direct RT assay, the enzyme
came into close contact with concentrated extracts. In an in vitro system the virus is still intact, with
the RT enzyme located in the core of the virus. The enzyme would only be exposed to a diluted
amount of extract, and only after the virus has penetrated and the enzyme is released from the viral
core.


The control drug, AZT, managed to significantly inhibit viral replication (Figure 40, section 3.5.2.1.)
in acutely infected T-lymphoid cells (CEM.NKR.CCR5). The effect of AZT on PBMCs (acutely- or
chronically infected) was marginal (Figures 42-45, sections 3.5.2.3. and 3.5.2.4.). This could be
because of the presence of more than one type of cell in the PBMC mixture – any inhibitory effect that
the AZT could have had on viral replication in one cell type would be overshadowed by the unscathed
viral replication in the other cell types.


Since Traore and Meyer (2002) found that cells shown as viable by tetrazolium dyes may already be
navigated towards some form of cell death, it is therefore advantageous to confirm or disprove
viability data with a similar but more accurate technique such as flow cytometry. The effect of the
crude extracts on cell viability or death of infected T-lymphoid cells (CEM.NKR.CCR5) treated with
crude extracts was analysed with Annexin-V antibodies (for detecting apoptosis) and propidium iodide
(for detecting necrosis) utilizing a flow cytometer. Results showed that extract-treated and untreated
infected cells were inherently in the late apoptotic phase of cell death after 4 days in culture (Figure
41, section 3.5.2.2.). This was expected since apoptosis is a means by which HIV-infection can lead to
T-cell depletion (explained fully in Section 1.1.9.). At 10 µM, AZT managed to rescue infected T-cells
from apoptosis and return them to a viable state. None of the crude extracts reflected this ability at 5
µg/ml, probably due to reasons discussed previously.




An initial screen of several crude plant extracts for inhibition of HIV RT in a direct enzyme assay
indicated that extracts from V. oligocephala and G. perpensa had a good potential for the isolation of a
natural product(s) with anti-RT activity. Crude extract from V. oligocephala was fractionated, but
yielded no active fraction. The fractionation of crude G. perpensa rhizome extract lead to the
collection of an active fraction. The identity and structure of the active principle could not be
concluded due to the presence of an epimeric mixture of glucose in the active fraction, as indicated by
nuclear magnetic resonance spectroscopy. An optimised isolation procedure for obtaining purified
active principle would have allowed for its identification and structure elucidation. This might be
achieved by chromatography over hydroxyl-form anion exchange surfaces prepared from anion
exchange resins at relatively low hydroxyl concentrations. When a strong base anion exchange resin,
e.g. a chloride-form strong base anion exchange resin, is conditioned with a low concentration of
hydroxyl (e.g.1-10 mM NaOH solution), the conditioned resin separates a number of sugars and sugar
alcohols from one another, while still allowing ready desorption from the resin.


Crude extracts from G. perpensa, at a non-toxic concentration of 5 µg/ml, did not prove to inhibit HIV
replication in vitro. The control drug (AZT) did manage to inhibit HIV replication but only in a T
lymphoid cell line. Since crude extracts did not prove to repress HIV replication, anti-HIV screens in
vitro on the pure compound might have proven to oppose the findings of screens with the crude
extracts: a possibility that increased amounts of the active principle (contradictory to the possibly low
concentrations present in the crude extracts) might inhibit HIV replication in vitro, has to be
investigated. Infection protocols should have been optimised as to allow for the detection of HIV RT
in the culture supernatants, giving a better reflection of the effect of crude extracts on viral replication
in vitro.




All crude extracts that were initially screened for HIV RT inhibition might have other possible
microbial targets. Qualitative antibacterial disc-diffusion assays of these extracts, performed on Gram-
positive (Enterococcus faecalis) and Gram-negative (Pseudomonas aeruginosa) bacterial strains,
indicated that some extracts had the ability to inhibit the growth of these bacteria (see Figures 46 and
47). Extracts 36B and 36F showed a slight inhibition on the growth of Enterococcus faecalis. Extracts
21 and 33A showed strong inhibition of Pseudomonas aeruginosa. Extracts 3, 22A, 25B, 26A, and
35B showed inhibition of Pseudomonas aeruginosa to a lesser degree. Please refer to Table 11
(section 2.2.1.) for the identity of the extracts.


Crude extracts were also screened for anti-malarial activity. Malaria is a major health problem in many
tropical and subtropical countries, with an estimated 300 to 500 million cases and 1.5 to 2.7 million
deaths annually (Burton & Engelkirk, 2000). About 90% of all malaria cases occur in Africa, where
approximately 1 million children die from malaria each year. Atamna and Ginsburg (1998) observed
that GSH interacts with haemin in a complex way. Haemin-binding compounds can inhibit this
interaction, and also disrupt haemin polymerisation and show antimalarial properties. Although the
nature of GSH-haemin interaction remains obscure, Steele et al. (2002) utilized the characteristic
spectral changes to develop a simple test for haemin-binding compounds. This assay is likely to
predict antimalarial activities. All crude extracts were screened with the multiwell plate GSH-haemin
interaction assay. Two antimalarial drugs, artemisinin and amethopterin, were included as positive
controls. The haemin-binding effect was evaluated as the decrease in absorbance (A360nm) compared
with the control absorbance. Several crude extracts showed to have inhibitory action in the GSH-
haemin binding assay (see Figure 48). These included extracts 4, 12, and 17B. Please refer to Table 11
(section 2.2.1.) for the identity of the extracts.


Further investigation into the active principles present in these extracts might lead to the identification
of compounds with anti-bacterial activity and antimalarial activity.
Figure 46: Antibacterial disc-diffusion assay of crude plant extracts for activity against a Gram-
              positive bacterium, Enterococcus faecalis.
A hundred and fifty micrograms of crude plant extracts were applied onto paper discs (5 mm diameter) , and the solvents allowed to
evaporate. A 100 ml of warm nutrient agar was poured into a bioassay dish (245 mm x 245 mm x 25 mm) and allowed to set.
Twentyfive millilitres of tryptone soy broth was inoculated with one millilitre of frozen bacterial stock solution, and incubated for 24 hours
at 37oC with agitation. One and a half millilitres of this overnight culture was mixed with 100 ml of warm nutrient agar, and then over-layed
on top of the solidified agar in the bioassay dish, and allowed to set. The paper discs were then applied onto the agar. Ampicillin (1 µg/µl)
was included as a control (at 1 µg, 2 µg, and 3 µg) as well as the solvent (DMSO) After incubation at 4oC for 1 hour, the dish was inverted
and incubated at 37oC overnight. The following day, the agar was inspected for bacterial growth. Clear rings indicate inhibition of bacterial
growth.
Figure 47: Antibacterial disc-diffusion assay of crude plant extracts for activity against a Gram-
              negative bacterium, Pseudomonas aeruginosa.
A hundred and fifty micrograms of crude plant extracts were applied onto paper discs (5 mm diameter) , and the solvents allowed to
evaporate. A 100 ml of warm nutrient agar was poured into a bioassay dish (245 mm x 245 mm x 25 mm) and allowed to set.
Twentyfive millilitres of tryptone soy broth was inoculated with one millilitre of frozen bacterial stock solution, and incubated for 24 hours
at 37oC with agitation. One and a half millilitres of this overnight culture was mixed with 100 ml of warm nutrient agar, and then over-layed
on top of the solidified agar in the bioassay dish, and allowed to set. The paper discs were then applied onto the agar. Ampicillin (1 µg/µl)
was included as a control (at 1 µg, 2 µg, and 3 µg) as well as the solvent (DMSO) After incubation at 4oC for 1 hour, the dish was inverted
and incubated at 37oC overnight. The following day, the agar was inspected for bacterial growth. Clear rings indicate inhibition of bacterial
growth.
                                    80


                                    60
    A362nm(t30min)-A362nm(t0min)




                                    40


                                    20


                                     0


                                   -20


                                   -40


                                   -60


                                   -80


                                   -100




                                                                                                         Extracts

Figure 48: GSH-haemin binding assay that displays possible antimalarial activities in crude extracts.
Crude extracts were subjected to a GSH-hemin binding assay (Steele et al., 2002) at a final concentration of 500 µg/ml. Two control drugs, nl. artemicinin and amethopterin, were included at 10 and 20 µM. Solvent
controls were also included. Good inhibition is indicated by a small difference between the A362nm (t=30min.) and the A362 (t=0 min.) values.

Error bars indicate standard deviation

								
To top