Tumour endoproteases: the cutting edge of cancer drug delivery?

Despite progression in anticancer drug development and improvements in the clinical utilization of therapies, current treatment regimes are still dependent upon the use of systemic antiproliferative cytotoxic agents. Although these agents are unquestionably potent, their efficacy is limited by toxicity towards 'normal' cells and a lack of tumour selective targeting, resulting in a therapeutic index which is modest at best. Consequently, the development of more tumour selective cancer treatments, with better discrimination between tumour and normal cells is unequivocally an important goal for cancer drug discovery. One such strategy is to exploit the tumour phenotype as a mechanism for tumour-selective delivery of potent therapeutics. An exciting approach in this area is to develop anticancer therapeutics as prodrugs, which are non-toxic until activated by enzymes localized specifically in the tumour. Enzymes suitable for tumour-activated prodrug development must have increased activity in the tumour relative to non-diseased tissue and an ability to activate the prodrug to its active form. One class of enzyme satisfying these criteria are the tumour endoproteases, particularly the serine- and metallo-proteases. These proteolytic enzymes are essential for tumour angiogenesis, invasion and metastasis, the major defining features of malignancy. This review describes the concept behind development of tumour-endoprotease activated prodrugs and discusses the various studies to date that have demonstrated the huge potential of this approach for improvement of cancer therapy.

Membrane type matrix metalloproteinases (MMPs) show differential expression in non-small cell lung cancer (NSCLC) compared to normal lung: correlation of MMP-14 mRNA expression and proteolytic activity.

Improved understanding of the involvement of matrix metalloproteinases (MMPs), including membrane-type MMPs (MT-MMPs), in human tumours has potential diagnostic, prognostic and therapeutic implications. We assessed the relationship between MT-MMP expression and clinicopathological parameters in human non-small cell lung cancer (NSCLC) and histologically normal lung tissue by quantitative Real Time PCR (qRT-PCR). All MT-MMPs (MMPs 14-17, 24 and 25) were detected by qRT-PCR with significantly higher MMP-14, -15 and -17 expression observed in tumour relative to normal lung specimens. MMP-16 was undetectable in normal lung but expressed in 8% tumours. MMP-15 demonstrated significant overexpression in adenocarcinomas relative to squamous cell carcinomas and normal lung tissue. MMP-14 mRNA expression strongly correlated to MMP-14 proteolytic activity in preclinical tumour models, indicating that qRT-PCR may predict MMP-14 activity levels in NSCLC. These data suggest that MMP-14, -15 and -17 may be good markers of disease, or therapeutic targets for treatment of human NSCLC.

Loss of tumourigenicity of stably ERbeta-transfected MCF-7 breast cancer cells.

Proliferation of breast cancer cells is mediated by estrogen receptors (ER)-ERalpha and ERbeta. At present, contradictory observations complicate the understanding of involvement of ERbeta in breast cancer and functional definition of ERbeta as a prognostic marker. A stable expression of full length ERbeta was established in the ERalpha-positive MCF-7 breast carcinoma cell line to evaluate the role for ERbeta in maintenance of cell viability and estrogenic response, as well as proliferation, morphology and cell cycle progression. In order to verify in vivo tumourigenicity of ERbeta transfectants were transplanted into nude mice. Transfection of ERbeta in MCF-7 resulted in a marginal increase of gelsolin protein expression. Constitutive expression of ERbeta resulted in a significant 30% inhibition of cellular growth compared with transfection of the mock vector alone (p=0.043). This reduction in growth was associated a retardation of transition into S-phase of the cell cycle. The in vitro response to 17beta-estradiol was reversed in cells over-expressing ERbeta (p=0.016). However, no difference in response to the antiestrogens tamoxifen and ICI 182,780 was observed in the presence of ERbeta. Importantly, over-expression of ERbeta prevented establishment and growth of tumours as subcutaneous xenografts in immunodeficient mice in vivo. These observations support the notion that ERbeta is a tumour suppressor and is exploitable in terms of cancer prevention, improving therapeutic response or predicting disease progression.

Negative regulation of G(1)/S transition by the candidate bladder tumour suppressor gene DBCCR1.

Deletion of all or part of chromosome 9q is the most common genetic alteration in all stages and grades of bladder cancer. DBCCR1 (deleted in bladder cancer chromosome region candidate 1) maps to the chromosome region 9q32-33, a candidate tumour suppressor locus for bladder cancer. Although no mutations of DBCCR1 have been detected in bladder tumours, expression of DBCCR1 is silenced by promoter hypermethylation in 50% of bladder cancer cell lines analysed. Here we sought to provide functional evidence to authenticate DBCCR1 as a tumour suppressor using gene-transfer methods. Exogenous expression of DBCCR1 protein or an HA epitope-tagged fusion protein, HA-DBCCR1 in NIH3T3 cells and human bladder tumour cell lines resulted in suppression of proliferation. Cell cycle analyses in NIH3T3 cells revealed that DBCCR1-mediated growth inhibition was due to an increase in the number of cells in the G(1) phase of the cell cycle. The levels of apoptosis were not altered. These results demonstrate a role for DBCCR1 in cell cycle control, thereby supporting the hypothesis that this is the tumour suppressor gene targeted by 9q32-33 deletion in bladder cancer.

The rodent non-genotoxic hepatocarcinogen Nafenopin and EGF alter the mitosis/apoptosis balance promoting hepatoma cell clonal growth.

The responses of a series of rat hepatoma cell lines (FaO, HTC, RH1) to the rodent non-genotoxic hepatocarcinogen and per-oxisome proliferator (PP) Nafenopin were studied to determine if this PP acts with EGF, a naturally occurring liver growth regulator, to perturb the balance between mitosis and apoptosis. EGF enhanced the growth of FaO cells (well differentiated) and HTC cells (intermediate differentiation) but not of the poorly differentiated RH1 cell line. Nafenopin also increased FaO cell growth but, surprisingly, retarded the growth of both HTC and RH1 cells. Since population expansion kinetics result from mitosis and death, replicative DNA synthesis (RDS) and apoptotic cell death were measured in HTC cells. As expected, EGF elevated RDS and suppressed cell death. However, Nafenopin depressed HTC net population expansion via a suppression of cell death coupled to a more marked inhibition of RDS. This apparent paradox was investigated using soft agar cloning. This revealed sub-populations with differing growth kinetics suggesting selective clonal expansion via an alteration in the balance between mitosis and apoptotic cell death. At later stages, cells are refractory to EGF and Nafenopin, suggesting that genetic changes may have superseded such factor-dependence.

Treatment of bilateral corneal ulceration in a Peregrine Falcon (Falco peregrinus) using 360 degree conjunctival flaps.

A wild Peregrine Falcon (Falco peregrinus) was presented with extensive bilateral fluorescein positive corneal damage. Local therapy and bilateral tarsorrhaphies resulted in slow improvement over 5 weeks. When bilateral 360 degree conjunctival flaps were used subsequently, healing proceeded more rapidly over the next 8 weeks. Although bulbar conjunctival flaps have been reported as difficult in birds due to their small size and relatively immobile bulbar conjunctiva, 360 degree conjunctival flaps made from palpebral rather than bulbar conjunctiva were found to be technically feasible in a larger bird species such as the Peregrine Falcon.

Binding of [3H]benzimidazole carbamates to mammalian brain tubulin and the mechanism of selective toxicity of the benzimidazole anthelmintics.

The binding of tritiated benzimidazole carbamates ([3H]BZCs) to mammalian brain tubulin was examined to investigate the kinetics of the BZC-tubulin interaction and to establish the mechanism of the selective toxicity of the BZC based anthelmintics. [3H]BZC binding to tubulin was markedly greater at 4 degrees than at 37 degrees for all ligands. The association constant (Ka) and maximum amount of [3H]BZC bound (Bmax) were temperature dependent for [3H]mebendazole ([3H]MBZ), [3H]oxibendazole ([3H]-OBZ) and [3H]oxfendazole ([3H]OFZ). The Ka and Bmax values obtained for [3H]MBZ, [3H]OBZ and [3H]OFZ, and the comparatively weak binding of [3H]carbendazim, reflected the known in vitro potency of these compounds as microtubule inhibitors. Dissociation of the [3H]MBZ-tubulin complex was also temperature dependent, the first order dissociation rate constant being reduced by two orders of magnitude at 4 degrees compared with that observed for 37 degrees. These results indicate that the binding of BZCs to mammalian brain tubulin is temperature dependent and suggest that temperature induced conformational changes in the tubulin dimer influence the ability of the BZCs to form a stable BZC-tubulin complex. The temperature dependence of BZC binding and the affinity of the BZCs for mammalian tubulin are therefore unlike the BZC-tubulin interaction observed for parasitic nematodes, where optimum BZC binding occurs at 37 degrees and results in the formation of a pseudo-irreversible complex.

In vitro activity of paraherquamide against the free-living stages of Haemonchus contortus, Trichostrongylus colubriformis and Ostertagia circumcincta.

Paraherquamide, an oxindole alkaloid recently reported to have potent nematocidal activity, was shown to have a marked inhibitory effect on the motility of the free-living larval stages of H. contortus, T. colubriformis and O. circumcincta. The effect of paraherquamide on larval motility could be distinguished from that caused by levamisole and the avermectins. After 72 h exposure, the concentration of paraherquamide required to inhibit the motility of 50% of L3 larvae present was 0.033, 0.058 and 2.7 micrograms ml-1 for O. circumcincta, T. colubriformis and H. contortus, respectively. Ivermectin (IVM)-resistant isolates of H. contortus were significantly more sensitive to the paralytic effects of paraherquamide than IVM-susceptible isolates of this species. Paraherquamide had no effect on the time for development from the egg to the L3 larval stage of H. contortus, T. colubriformis and O. circumcincta.

The kinetics of mebendazole binding to Haemonchus contortus tubulin.

The kinetics of the binding of mebendazole (MBZ) to tubulin from the third-stage (L3) larvae of the parasitic nematode, Haemonchus contortus, have been characterized. In partially purified preparations, the association of [3H]MBZ to nematode tubulin was rapid, k1 = (2.6 +/- 0.3) x 10(5) M-1 min-1, but dissociation was slow, k-1 = (1.58 +/- 0.02) x 10(-3) min-1. The affinity constant (K(a)) for the interaction, determined by the ratio k1/k-1, was (1.6 +/- 0.2) x 10(8) M-1. Similar results were obtained with crude cytosolic fractions. In equilibrium studies, performed with partially purified nematode tubulin under similar conditions, a K(a) of (5.3 +/- 1.6) x 10(6) M-1 was obtained. The best estimate for the K(a) of the MBZ-nematode tubulin interaction is considered to be the 'kinetic' value determined from the ratio of rate constants. The slow dissociation of MBZ from nematode tubulin, which contrasts with the rapid dissociation of MBZ from mammalian tubulin, supports the hypothesis that the selective toxicity of the benzimidazole anthelmintics results from a difference between the affinities of mammalian and nematode tubulins for these drugs.

Cryopreservation of the first-stage larvae of trichostrongylid nematode parasites.

First stage (L1) larvae of Haemonchus contortus, Trichostrongylus colubriformis and Ostertagia circumcincta can be cryopreserved in the presence of DMSO using a two-step freezing protocol involving an initial period at -80 degrees C prior to transfer to liquid nitrogen. Thawed L1 larvae continue development in vitro producing third stage (L3) larvae that are infective to sheep when dosed per os. Establishment rates for L3 larvae grown from thawed L1 larvae were 40 and 80% for H. contortus and T. colubriformis, respectively. There was no difference in survival or infectivity between benzimidazole (BZ)-susceptible and BZ-resistant H. contortus isolates and cryopreservation caused no shift in their BZ-resistance status as indicated in an in vitro larval development assay. Cryopreservation also had no effect on the sensitivity of these isolates to the avermectins or levamisole in vitro. High survival rates (60-70%), good levels of establishment and the stability of anthelmintic resistance status of isolates indicate that little if any selection occurs during the cryopreservation process. L1 larvae of all 3 species have been successfully recovered after 16 months storage in liquid nitrogen, cultured to the L3 stage and established in sheep.

Avermectin/milbemycin resistance in trichostrongyloid nematodes.

Resistance to levamisole and the benzimidazoles appears to be achieved by one or, at most, two mechanisms in the common trichostrongyloid parasites of sheep. For the avermectin/milbemycin anthelmintic class the picture is more complex. In-vitro assays employing the free-living stages of trichostrongyloid nematodes were used to investigate structure-activity relationships for the avermectins/milbemycins. While avermectin/milbemycin-susceptible isolates of Haemonchus contortus, Trichostrongylus colubriformis and Ostertagia circumcincta were found to differ in their intrinsic sensitivities to avermectin/milbemycin inhibition of larval development and L3 motility, structure-activity profiles against all three species were similar. In-vivo avermectin/milbemycin resistance was associated with a reduced sensitivity to avermectin/milbemycin inhibition of larval motility and/or development in some, but not all, isolates. Where a reduced sensitivity to avermectin/milbemycin inhibition of larval development was observed, different groups of resistant isolates displayed different structure-activity profiles. Many avermectin/milbemycin-resistant isolates showed an increased sensitivity to paraherquamide. These in-vitro data have allowed the classification of avermectin/milbemycin-resistant isolates into a number of distinct types. Study of the inheritance of avermectin/milbemycin resistance in two resistance types suggests that the in-vitro differences between resistant isolates reflect important differences in the mechanism of resistance present. The kinetics of expulsion of H. contortus, T. colubriformis and O. circumcincta from sheep following treatment with ivermectin indicate that, in vivo, the critical action of avermectins/milbemycin against O. circumcincta may be different to that which results in H. contortus and T. colubriformis elimination. This observation provides some explanation for the differences between resistant isolates. If, for different species, the critical event(s) leading to expulsion are different, then it follows that the mechanisms of resistance observed may also differ.

Management of anthelmintic resistance: inheritance of resistance and selection with persistent drugs.

Resistance to the benzimidazole (BZ) anthelmintics is inherited as an incomplete dominant/ incomplete recessive trait and is now widespread in populations of gastrointestinal nematode parasites of sheep. Unlike benzimidazole resistance, which is common in Haemonchus contortus, Trichostrongylus colubriformis and Ostertagia circumcincta, resistance to levamisole is relatively rare in H. contortus, although common in the other 2 species. One explanation for the slow spread of resistance to levamisole in H. contortus is that it is inherited as an autosomal recessive trait, while in T. colubriformis levamisole resistance is inherited as a recessive sex-linked trait. With the introduction of the avermectin/milbemycin class resistance has developed to the relatively short-acting ivermectin, but this time it is inherited as a completely dominant trait. The potentially more serious situation of a persistent anthelmintic selecting a dominant resistance gene was investigated using a simulation model. Efficacy against incoming infective larvae (L3) was assumed to decline or remain high over the period of drug persistence (3 days to 4 weeks), thus allowing the estimation of the relative importance of selecting resistant L3s on the development of resistance in the worm population. These factors were also examined against a background of initial efficacy levels, against adults, and mode of inheritance. Persistence and initial efficacy were found to be far more important in determining the rate of selection for resistance than was selection of resistant L3 as drug efficacy declined.

Evidence of multiple mechanisms of avermectin resistance in haemonchus contortus--comparison of selection protocols.

Three isolates of Haemonchus contortus selected for avermectin resistance in sheep were compared in three in vitro pharmacological tests previously shown to discriminate between field isolates of H. contortus resistant and susceptible to the avermectins. Two isolates, F7-A and IVC, were selected for avermectin resistance in the laboratory from a reference susceptible isolate using suboptimal doses of ivermectin (LD95) for 7 and 16 generations, respectively. In these isolates avermectin resistance was not associated with a decreased sensitivity to avermectin inhibition of larval development or L3 motility but was associated with an increased sensitivity to paraherquamide. The third isolate, Warren, was derived from an overwhelmingly avermectin-susceptible, mixed species field isolate in a single generation by propagating the small number of survivors of a 0.2 mg/kg ivermectin treatment (i.e. 10 x LD95). This isolate, like previously characterised avermectin-resistant H. contortus isolates derived from the field in South Africa and Australia, showed a markedly reduced sensitivity to avermectin inhibition of larval development and L3 motility, as well as an increased sensitivity to paraherquamide. These results suggest that avermectin resistance can manifest itself in different ways and that the two selection protocols used to generate the F7-A, IVC and Warren isolates have resulted in the selection of different resistance phenotypes.

Detection of resistance to ivermectin in Haemonchus contortus.

Infective, third-stage (L3) larvae of Haemonchus contortus isolates resistant to ivermectin (IVM) show a decreased sensitivity to IVM-induced paralysis in vitro. The inhibition of larval motility by IVM can be detected in L3 larvae incubated in the dark on an agar matrix containing IVM, by the failure of affected larvae to move when stimulated by exposure to light. Optimally, avermectin (AVM) potency is quantified after three cycles, each involving storage in the dark for 24 h followed by a brief exposure to light. For IVM-susceptible isolates, a 50% inhibition of motility (LP50) was achieved with IVM concentrations between 0.30 and 0.49 microM, while LP50 values in IVM-resistant isolates ranged from 0.8 to 2.6 microM depending on the in vivo resistance status of the isolate. A limited study of structure-activity relationships within the AVM class indicated that in vitro inhibition of L3 motility was consistent with the known in vivo efficacy of each analogue. Resistance factors for IVM-resistant isolates were dependent on AVM structure with the more polar AVM B2 analogue being a particularly sensitive probe of IVM-resistance status.

Avermectin inhibition of larval development in Haemonchus contortus--effects of ivermectin resistance.

Avermectin (AVM) inhibition of the development of the free-living stages of Haemonchus contortus has been quantified in an assay in which nematode eggs are placed on an agar matrix containing serial dilutions of a drug in the wells of a microtitre plate. Development is allowed to proceed for 6 days by which time larvae in control wells (no drug) have reached the infective, third (L3) stage. At high concentrations (> 30 nM) ivermectin (IVM) paralyses L1 larvae soon after hatching, however, much lower concentrations (approximately 1 nM) are sufficient to inhibit development to the L3 stage which suggests that effects of the drug other than those relating to gross motor activity are responsible for the latter effect. The larval stages of IVM-susceptible H. contortus isolates from both Australia and South Africa, including isolates known to be resistant to levamisole or rafoxanide and/or the benzimidazoles, were equally sensitive to inhibition by AVMs. In contrast, 6 isolates of H. contortus resistant to IVM in vivo showed a reduced sensitivity to AVM inhibition of development. The order of potency of a limited range of AVMs as inhibitors of larval development was consistent with in vivo efficacy. Resistance ratios for IVM-resistant isolates were dependent on AVM structure, with AVM B2 the most sensitive probe for IVM resistance in the isolates tested.

The survival of Ostertagia circumcincta and Trichostrongylus colubriformis in faecal culture as a source of bias in apportioning egg counts to worm species.

When cultured alone or concurrently with Trichostrongylus colubriformis in sheep faeces, Ostertagia circumcincta produced fewer infective larvae per 100 eggs than did T. colubriformis. Averaged over five trials 60% of T. colubriformis eggs were recovered as infective larvae while for O. circumcincta the figure was only 39%. This result was observed for two strains of O. circumcincta and was independent of when larvae were harvested from culture (days 6-10 at 25 degrees C). The mortalities of both species occurred at the first and second larval stages. These observations are of concern when using larval differentiation from faecal culture to make quantitative estimates of worm egg numbers for each species present. Species such as T. colubriformis which have a low mortality during culture are likely to have their egg numbers overestimated when cultured with a species, like O. circumcincta, that suffers high mortality in culture.

Characterisation of an avermectin resistant strain of Australian Haemonchus contortus.

A strain of Haemonchus contortus (CAVR) isolated in Australia was found to be resistant to ivermectin (IVM) with 0.4 mg kg-1 of the anthelmintic failing to significantly reduce worm burdens. Resistance to IVM was sex-influenced in the CAVR strain with adult males showing a greater sensitivity to IVM. Cross resistance to moxidectin was evident with approximately 15% of the population surviving a dose of 0.1 mg kg-1. The free-living stages of the CAVR isolate had a reduced sensitivity to avermectin (AVM) inhibition of development and motility. Similar structure-activity patterns and resistance factors were obtained for a series of related AVMs as inhibitors of larval development and L3 motility in CAVR and White River II, an IVM-resistant H. contortus isolate from South Africa. Further, both isolates were found to be 3 times more sensitive to paraherquamide than a susceptible H. contortus isolate. This suggest that the same resistance mechanism is operating in both isolates. The CAVR strain is susceptible to the benzimidazoles, levamisole and closantel.

Inheritance of avermectin resistance in Haemonchus contortus.

A larval development assay was used to compare the responses of the Chiswick Avermectin Resistant (CAVRS) isolate of Haemonchus contortus, an avermectin-susceptible isolate (VRSG) and their crosses to avermectins. The F(1) and F(2) generations of reciprocal crosses between CAVRS and VRSG were denoted as CAVRS malesxVRSG females=CXV, and VRSG malesxCAVRS females=VXC. The levels of avermectin resistance in the developing larvae of the F(1) of both CXV and VXC were indistinguishable from that in the avermectin-resistant parent, indicating that the resistance trait is completely dominant. Avermectin dose-response curves for the CXV F(1) did not show a 50% mortality rate at low concentrations, indicating that avermectin resistance is not sex-linked. This conclusion was confirmed when adult male worms of the F(1) of the CXV mating were found to have survived treatment of the host with 200microgkg(-1) ivermectin. This dose rate (200microgkg(-1) ivermectin) caused a 50% reduction in the number of adult males in the F(1) from both CXV and VXC crosses, but only a non-significant reduction in the number of adult females in the F(1). Dose-response curves obtained for the F(2) generations in the larval development assay indicated the presence of 25% of avermectin-susceptible individuals, suggesting that a single major gene largely controls the avermectin-resistance trait. This genetic analysis of avermectin resistance in an Australian H. contortus isolate indicates that the expression of the gene for avermectin resistance is an autosomal, complete dominant in the larvae; however, in adults its expression is sex-influenced, with males having a lower resistance to avermectin than females.

Bafilolides, potent inhibitors of the motility and development of the free-living stages of parasitic nematodes.

Three Streptomyces isolates were identified as producing macrolide antibiotics of the bafilomycin or leucanicidin types during an evaluation of Australian actinomyces for the production of inhibitors of larval development in the parasitic nematode, Haemonchus contortus. Bafilomycins A1, B1, C1, and D were obtained from culture A239 and the 2-O-methyl-L-rhamnosyl derivative of bafilomycin A1, leucanicidin, from cultures A223 and A240. All these 'bafilolides' gave similar patterns of inhibition typified by an initial paralysis of newly hatched L1 larvae and a lethal toxicity within 24 h. LD50 values for inhibition of larval development of McMaster H. contortus ranged from 0.23 micrograms ml-1 for leucanicidin to 2.5 micrograms ml-1 for bafilomycin D. The bafilolides had broad spectrum nematocidal activity, being equi-potent as inhibitors of H. contortus, Trichostrongylus colubriformis and Ostertagia circumcincta larval development. Further, all bafilolides caused some inhibition of H. contortus L3 motility, with the semi-synthetic analogue, bafilomycin B2, the most potent inhibitor (LP50 against McMaster H. contortus 1.9 microgram ml-1). Nematode strains resistant to the known benzimidazole, levamisole and avermectin anthelmintics showed no cross resistance to the bafilolides, supporting the hypothesis that the bafilolides act by an independent mechanism.

Biochemistry of benzimidazole resistance.

Heavy reliance on the benzimidazole (BZ) anthelmintics since their introduction in the 1960's for the control of gastrointestinal parasites of livestock has led to widespread BZ resistance in target parasite species. The BZs exert their primary action by binding to tubulin, the major protein component of microtubules. This review discusses the biochemistry of the interaction between the BZs and tubulin from mammalian and BZ-resistant and -susceptible parasite sources, exploring aspects of the selective toxicity of these drugs and examining the mechanism of BZ resistance. Although tubulin is a highly conserved protein present in both the host and the parasite, the BZs demonstrate relatively low mammalian toxicity. The selectivity of these drugs can be explained by the much higher affinity of the BZs for tubulin from the parasite at 37 degrees C compared to their affinity for tubulin from the host. This difference in affinity reflects the considerably slower rate of BZ dissociation from parasite tubulin. BZ-resistance in parasitic nematodes is characterised by a loss of high affinity BZ-parasite tubulin interactions and a corresponding increase in lower affinity interactions, although there are still significant differences between BZ-resistant parasite tubulin and tubulin from the host. These differences suggest the potential for the design of new generation BZs active against 'BZ-resistant' parasites.