The mucopolysaccharidosis (MPS) disorders are a group of inherited metabolic diseases resulting from abnormalities in enzymes required for the turnover of intracellular glycosaminoglycan (GAG) chains. MPS I is considered the classic MPS type and is the most common MPS occurring in approximately 1 in 100,000 newborns worldwide; however incidence and prevalence of phenotypic groups vary from region to region [1
]. It is caused by a reduction in, or a deficiency of, the lysosomal enzyme α-l
-iduronidase (IDUA, E.C. 22.214.171.124), which is required for the intracellular degradation of heparan sulphate (HS) and dermatan sulphate (DS) GAG chains. In the absence of IDUA, HS and DS GAGs progressively build up within cells, disrupting normal cell function and resulting in multiple organ failure. Both severe and attenuated forms of MPS I are observed, reflecting differences in residual enzyme activity [3
]. Because of the widespread build-up of HS and DS GAGs, affected children display a range of symptoms including organomegaly, skeletal abnormalities, mental deterioration and corneal clouding. Death most commonly arises from cardiac or respiratory failure in early adolescence.
Because of the range of organs affected by MPS I and the progressive nature of pathology, finding an effective multi-tissue treatment for MPS I has been difficult, as indeed it has been for all MPS types. Bone marrow transplantation (BMT) and enzyme replacement therapy (ERT) are currently in the clinic; however, each has its own limitations.
Bone marrow transplantation has been shown to be effective in prolonging life and reducing some clinical symptoms associated with MPS I, such as slowing down central nervous system (CNS) deterioration [4
]. However, this improvement is generally only seen when transplantation is undertaken at a very early age (<2 years with 6–12 months being the optimal age) and therefore before the onset of significant CNS pathology. Bone pathology also responds to BMT but the outcome is variable [6
] and orthopaedic procedures are still needed after successful BMT. Other problems associated with BMT include the high mortality rate (10%–20% in the first year post-BMT) and finding a suitable donor before the onset of symptoms [7
]. Despite these limitations, BMT remains the treatment of choice for MPS I patients with CNS involvement who are 2 years old or less.
ERT using recombinant IDUA (Aldurazyme®
, Genzyme, Cambridge, MA, USA) was introduced in 2003 (USA). Advantages of ERT over BMT are the higher level of circulating enzyme that can be achieved and the negation of graft-versus-host disease. Patients on ERT have improved heart and lung function, reduced urinary GAG excretion and decreased liver and spleen size [8
]. Skeletal disease response to ERT treatment is variable, with limited improvement seen in some cases [9
]. Joint flexion shows a small improvement after ERT but improvement soon plateaus [9
] and joint function does not normalise [9
]. Corneal clouding also does not respond to ERT [9
]. With a molecular weight of 87 kDA [10
-iduronidase is too large to cross the blood-brain barrier and as such is ineffective in treating the neurological aspect of the disease. MPS I patients with CNS disease are therefore not placed on ERT regimens. However, there is a role for ERT in stabilising respiratory disease in MPS I patients with CNS disease prior to BMT [11
The limitations of current therapies, especially towards the major pathologies observed in MPS I (brain and skeletal disease), have led to an intense research effort by many laboratories to devise alternative approaches that can be used as either stand-alone therapies or adjuvant therapies to ERT or BMT. Substrate deprivation or reduction therapy (SDT/SRT) reduces the initial synthesis of the GAG chains that are the substrate for the MPS enzyme and restores the balance between synthesis and degradation of GAG in MPS cells. The advantage of SDT over ERT/BMT lies in the use of small molecular weight chemical inhibitors of GAG synthesis that are capable of crossing the blood-brain barrier or diffusing into poorly vascularised tissues such as cartilage and cornea.
The isoflavone genistein has been used as an SDT/SRT candidate for reducing GAG synthesis. Studies in MPS II mice have shown decreased urinary, liver, spleen and heart GAG, as well as a small reduction in brain GAG storage at low concentrations, with a higher dose of genistein not providing any additional benefit [13
]. MPS IIIB mice treated with genistein showed reduced tissue GAG content [14
] and a correction in aberrant open field exploration [15
These promising results in mice led to human clinical trials to treat MPS IIIA and MPS IIIB patients, which showed decreased urinary GAG, improved hair composition, improved/stabilised sleep habits, improved general behaviour, speech performance and comprehension in some patients as well as improved or stabilised disease in some patients [16
]. No long-term reports (>3 years) of patient use have been reported for genistein. A double-blinded, placebo-controlled 6-month clinical trial showed no statistically significant clinical improvement in treated patients, despite significantly lower levels of urinary GAG detected in treated patients [18
] and higher doses (15 mg/kg/day) did not correlate to significantly improved behaviour despite a dose-dependent decrease in urinary GAG observed [19
]. A supra-high dose genistein (150 mg/kg/day) trial in MPS patients with CNS disease showed limited effect on disease, potential estrogenic side effects and variable effect on urinary GAG excretion [20
]. Studies have also shown limited improvement in shoulder joint range of motion in MPS II patients [21
]). No adverse reactions were observed in any of these studies to genistein treatment.
Our laboratory has developed and tested substrate deprivation therapy (SDT) using rhodamine B (also termed substrate reduction therapy) for MPS disorders [22
]. While we have shown that this approach using rhodamine B improved somatic [24
] and neurological function in the MPS IIIA mouse [23
], its effect on the multiple sites of pathology in MPS I is unknown. MPS I presents with subtle differences in CNS disease progression compared to MPS III disease, which can present with severe aggression and temperament issues not generally seen in MPS I patients [2
]. MPS I pathology also has a skeletal disease component not seen in MPS IIIA, which may affect the clinical efficacy. Before we can put forward SDT as a therapy for the pathology observed in MPS I, it must be trialled in the appropriate animal model.
There are animal models of MPS I in the cat [28
], dog [29
] and mouse [30
], with the latter being well characterised. The MPS I knockout mouse (Idua−/−
] displays somatic pathology that includes increased urinary output, widespread lysosomal storage as determined by histology and severe skeletal and physical abnormalities [30
]. Behaviour deficits are also observed as measured by open field exploration, marble burying anxiety, spatial learning and memory and novel object recognition [32
]. This mouse model is an accurate model of human MPS I disease and is a valuable tool for characterising disease progression and the evaluation of new therapies.
In this study, we report SDT treatment of MPS I mice using rhodamine B, a non-specific inhibitor of GAG synthesis. In a six-month in vivo therapy trial, rhodamine B altered several clinical parameters of disease progression toward normal including a reduction in female bodyweight gain, decreased lung GAG, decreased bone mineral volume and treated female mice showed improved learning ability as assessed by the water cross maze. This suggests that substrate deprivation therapy should not be overlooked in models without residual enzyme activity particularly as an adjunct therapy.
SDT targeting inhibition of GAG synthesis is an emerging therapy for MPS disorders. MPS I is the most common MPS disease with storage of dermatan (DS) and heparan sulphate (HS) GAGs resulting in neurological and skeletal pathology. Because of the range of organs affected by MPS I and the progressive nature of pathology, finding an effective multi-tissue treatment for MPS I has been difficult, as indeed it has been for all MPS types. SDT has the advantage of using small molecules which can diffuse into typically hard-to-reach tissues. Genistein has been used in MPS animal models, which showed an improvement in tissue GAG [13
] and also improved behaviour in MPS IIIB [15
]. Clinical trials with genistein have shown variable effect on MPS disease with slight improvement in cognition or stabilisation of disease in some patients, [16
] with higher doses not correlating to improved behaviour outcome [19
Rhodamine B is a non-specific inhibitor of GAG synthesis and has been used in vitro
and in vivo
in MPS IIIA, where an improvement in somatic and neurological pathology was observed [23
]. The aim of this study was to determine if SDT alters CNS and skeletal pathology in the knockout MPS I mouse model when residual enzyme activity is minimal or absent. Although no adverse effects have been observed in mouse studies [22
], reports of rhodamine B use in humans are limited. One study reports that an acute overexposure to aerosolised rhodamine B in humans resulted in transient mucous membrane and skin irritation, however, these symptoms resolved within 24 h [41
], while acute overexposure to oral rhodamine B resulted in increased urinary excretion of the compound but no other symptoms [42
]. Studies in MPS IIIA mice show no effect on litter size or liver function, with low levels of rhodamine B safe to administer in mice in utero
and over multiple generations [22
MPS I mice treated with weekly injections of 1 mg/kg rhodamine B showed no side-effects and female bodyweights were significantly reduced compared to their untreated MPS I littermates. A slight delay in male bodyweight gain was observed with treatment however, overall, treatment had little effect on male weight gain. This is in contrast to previous studies in our laboratory in MPS IIIA mice which showed a reduction in bodyweight gain in both males and females [24
], suggesting that the mechanism of weight gain may be different in MPS I compared to MPS IIIA.
Tissue GAG was elevated in MPS I untreated mice in all tissues examined compared to normal mice. Treatment with rhodamine B resulted in a significant reduction in the level of GAG in the lung, with no change in tissue GAG levels in liver, kidney, spleen, heart and brain. Again, this is in contrast to our results in the MPS IIIA mouse in which tissue GAG decreased in rhodamine B treated mice [23
]. The MPS I mouse model is a knockout with no detectable residual enzyme activity, whereas the MPS IIIA mouse model retains 3%–4% residual enzyme activity [30
]. Because tissue GAG levels reflect the rate of synthesis (reduced with rhodamine B) combined with the rate of degradation (nil in MPS I, reduced but still present in MPS IIIA), it is possible that by 6 months of age the cumulative amount of GAG storage in treated MPS I mice reached the same level as untreated MPS I mice. This notion can be tested by short-term rhodamine B treatment of MPS I mice to determine if there is a transient decrease or delay in accumulation of GAG.
MPS I untreated mice also showed significantly elevated levels of β-hexosaminidase and significantly reduced α-l
-iduronidase enzyme activity. No change in either enzyme activity was observed in MPS I mice treated with rhodamine B, suggesting that rhodamine B’s mode of action is specifically targeted to inhibition of GAG synthesis and not stabilisation of enzymes [44
MPS I mice display behavioural deficits in the water cross maze, open field, rotarod and inverted grid tests reflecting their CNS and skeletal pathology [32
] (Figure 4
, Figure 5
and Figure 6
). SDT with rhodamine B improved learning capacity in female mice but was unable to preserve long-term memory. Although male MPS I mice had a reduced learning capacity this was not significantly different to normal mice, and treatment had no effect. The reason for this difference in behaviour between the sexes, which both accumulate similar levels of brain GAG and ganglioside, is unknown. We have previously shown that both male and female MPS IIIA mice display learning deficits in the water cross maze [23
], again indicating differences in the manifestation of CNS pathology in MPS I and MPS IIIA. SDT also improved performance on the inverted grid test, which may reflect an improvement in skeletal disease. SDT did not alter anxiety or balance/co-ordination as indicated by open field and rotarod tests respectively. In view of the differences observed between the response of MPS I and MPS IIIA mice to SDT in biochemical and behavioural tests, a study of MPS IIIA response to SDT using the open field and rotarod tests is currently underway.
MPS I also develop skeletal pathology as shown by a significant increase in L5 vertebral bone mineral volume, primarily due to an increase in trabecular number. BV/TV was significantly decreased in rhodamine B treated MPS I mice. This is the first report of SDT targeting inhibition of GAG synthesis to have a positive outcome on bone. Skeletal disease is not adequately addressed using current therapies and this could provide a potential treatment for reaching this disease site. Improvement in the inverted grid may also be related to improvement in skeletal disease.
In agreement with previous data, we observed both CNS and skeletal deficits in the MPS I mouse and in addition observed differences in water cross maze performance between males and females. A differential response to therapy between the sexes was also observed with respect to bodyweight gain and the water cross maze. The MPS I mouse model thus displays subtle differences in manifestation of pathology and in response to SDT of the MPS IIIA model, underscores the need to test therapies on models of specific MPS types. That SDT can be used to target CNS and skeletal pathology in MPS I, a model with no detectable residual enzyme activity, suggests that SDT may have a role to play in modulating pathology in severe MPS by delaying the onset of symptoms. However, a more thorough time course, incorporating both longer and shorter response times than analysed here, is required to confirm this concept.