Research ArticleDrug Discovery

Effect of selective LRRK2 kinase inhibition on nonhuman primate lung

See allHide authors and affiliations

Science Translational Medicine  04 Feb 2015:
Vol. 7, Issue 273, pp. 273ra15
DOI: 10.1126/scitranslmed.aaa3634

A lung phenotype for LRRK2 inhibitors

Human genetic evidence implicates leucine-rich repeat kinase 2 (LRRK2) as a high-priority drug target for Parkinson’s disease. However, the benefit and risk of inhibiting the kinase activity of LRRK2 is unknown and is currently untested in humans. Using two selective LRRK2 kinase inhibitors, Fuji et al. report a safety liability in nonhuman primates characterized by morphological changes in lung. These results are consistent with observations in mice lacking LRRK2. These safety observations offer a cautionary note for pharmacological modulation of LRRK2 in humans.

Abstract

Inhibition of the kinase activity of leucine-rich repeat kinase 2 (LRRK2) is under investigation as a possible treatment for Parkinson’s disease. However, there is no clinical validation as yet, and the safety implications of targeting LRRK2 kinase activity are not well understood. We evaluated the potential safety risks by comparing human and mouse LRRK2 mRNA tissue expression, by analyzing a Lrrk2 knockout mouse model, and by testing selective brain-penetrating LRRK2 kinase inhibitors in multiple species. LRRK2 mRNA tissue expression was comparable between species. Phenotypic analysis of Lrrk2 knockout mice revealed morphologic changes in lungs and kidneys, similar to those reported previously. However, in preclinical toxicity assessments in rodents, no pulmonary or renal changes were induced by two distinct LRRK2 kinase inhibitors. Both of these kinase inhibitors induced abnormal cytoplasmic accumulation of secretory lysosome-related organelles known as lamellar bodies in type II pneumocytes of the lung in nonhuman primates, but no lysosomal abnormality was observed in the kidney. The pulmonary change resembled the phenotype of Lrrk2 knockout mice, suggesting that this was LRRK2-mediated rather than a nonspecific or off-target effect. A biomarker of lysosomal dysregulation, di-docosahexaenoyl (22:6) bis(monoacylglycerol) phosphate (di-22:6-BMP), was also decreased in the urine of Lrrk2 knockout mice and nonhuman primates treated with LRRK2 kinase inhibitors. Our results suggest a role for LRRK2 in regulating lysosome-related lamellar bodies and that pulmonary toxicity may be a critical safety liability for LRRK2 kinase inhibitors in patients.

INTRODUCTION

Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by both motor and cognitive dysfunction. Whereas motor symptoms may be managed with dopamine replacement therapy in the earlier stages of the disease, no disease-modifying therapy exists. Leucine-rich repeat kinase 2 (LRRK2) has been identified as a potential target for disease-modifying therapy because mutations in the catalytic core of LRRK2 have been associated with both autosomal-dominant and late-onset sporadic PD (16).

LRRK2 is a large protein that contains numerous protein-protein interaction domains and a central catalytic core composed of a Ras of complex (ROC) guanosine triphosphatase (GTPase) domain, a mitogen-activated protein kinase kinase kinase (MAPKKK) domain, and an intervening C-terminal of ROC (COR) domain (7). LRRK2 has been linked to a variety of cell processes including protein synthesis, mitochondrial function, apoptosis, cytoskeletal dynamics, and various aspects of lysosomal and proteasomal function, such as chaperone-mediated autophagy (812). However, the mechanism by which LRRK2 mutations cause or contribute to PD is still not clear. The most common LRRK2 mutation, G2019S, which is found in the activation loop of the kinase domain, results in increased kinase activity (7, 13), as does the kinase domain I2020T mutation (14). More recently, mutations in the ROC GTPase domain (N1437H, R1441C/G) have been shown to increase LRRK2 kinase activity (15). These findings support the prevailing hypothesis that increased LRRK2 kinase activity causes or contributes to PD-associated neurodegeneration. Inhibition of LRRK2 kinase activity has therefore been proposed as an attractive therapeutic strategy for slowing the progression of both genetic and sporadic forms of PD.

We set out to assess the safety risk of inhibiting LRRK2 kinase activity. We evaluated LRRK2 expression in human and murine tissues and observed similar tissue expression profiles between species. Given the similarities, we performed a phenotypic comparative analysis between wild-type (WT) and Lrrk2 knockout (KO) mice to identify potential safety concerns in tissues expressing LRRK2. We also assessed the preclinical pharmacology, toxicity, and toxicokinetic (TK) profiles of two LRRK2-specific, brain-penetrating kinase inhibitors in rodents and nonhuman primates (NHPs). These structurally distinct small-molecule inhibitors are characterized by single-digit nanomolar biochemical and cellular activities, broad kinome and receptor selectivity, and good in vivo oral exposure and unbound brain concentrations across preclinical species (16, 17). These two molecules were therefore chosen to evaluate the consequences of inhibiting LRRK2 kinase activity in preclinical repeat-dose safety assessments.

RESULTS

Similar LRRK2 mRNA expression profiles in human and murine tissues

To determine possible target tissues for LRRK2 kinase inhibition by pharmacological intervention, we investigated and compared LRRK2 mRNA expression data from human and murine tissues. In human tissues, the highest LRRK2 mRNA expression was detected in lung and white blood cells, followed by lymphoid tissue, kidney, and bone marrow (fig. S1). In mice, LRRK2 expression, as evaluated by nonisotopic in situ hybridization (ISH), was observed in peripheral organs similar to that seen in humans, including the marginal zone B lymphocytes of the spleen, macrophages/monocytes of the splenic red pulp, B lymphocytes of lymph nodes, mature neutrophils in the bone marrow, renal proximal tubular epithelial cells, and type II pneumocytes (fig. S2).

Lung and kidney abnormalities in LRRK2 KO mice

Given the similar tissue expression profiles in humans and mice for LRRK2, we considered phenotypic evaluation of Lrrk2 KO mice to be useful for obtaining further information related to the safety risk assessment of LRRK2 kinase inhibition. Similar to findings reported in an independent Lrrk2 KO mouse line (18), the Lrrk2 KO [EGFP knock-in (KI)] mice had phenotypic abnormalities in the lungs and kidneys. These were characterized by accumulation of enlarged versions of secretory lysosome-related organelles known as lamellar bodies in type II pneumocytes of the lung (fig. S3) and by increased numbers of secondary lysosomes (phagolysosomes) in the proximal tubular epithelium of the kidney (fig. S4).

In vitro assessment of LRRK2 kinase inhibitors GNE-7915 and GNE-0877

Two small-molecule LRRK2 kinase inhibitors, GNE-7915 (7915) and GNE-0877 (0877) (Fig. 1), were chosen for preclinical safety assessment. Their choice was based on single-digit nanomolar LRRK2 enzyme inhibitor constant (Ki), cell-based, and in vivo IC50 (drug concentration causing 50% inhibition of the desired activity) values, and well-balanced in vitro and in vivo profiles (Table 1) (16, 17). These molecules exhibited high LRRK2 specificity over other kinases and receptors at 0.1 and 1 μM (Table 1 and table S1). The only potential off-target kinase of overlapping concern for both molecules was human dual-specificity protein kinase (TTK). Subsequently, biochemical activities were measured, and these studies demonstrated 50- and 212-fold selectivity for LRRK2 over TTK for 7915 and 0877, respectively. Off-target receptor profiling, which included assessing LRRK2 inhibitor binding or activity against an expanded panel of brain receptors, resulted in each molecule showing >70% inhibition at 10 μM for only 5-hydroxytryptamine receptor 2B (5-HT2B). Follow-up functional IC50 determination confirmed that both molecules were antagonists of 5-HT2B (7915 at 1 μM; 0877 at 0.56 μM).

Fig. 1. Molecular structures of 7915 and 0877.

These diaminopyrimidine small-molecule LRRK2 kinase inhibitors have single-digit nanomolar biochemical and cellular activities, broad kinome and receptor selectivity, and good in vivo oral exposure and unbound brain concentrations across preclinical species. Hence, they were selected for evaluation of the in vivo effects of LRRK2 kinase inhibition in rodents and NHPs.

Table 1. 7915 and 0877 in vitro potency and selectivity profiles.

ND, not done.

View this table:

No lung or kidney pathology in rodents after selective LRRK2 kinase inhibition

Male C57BL/6 mice were orally administered 7915 at 200 or 300 mg/kg or 0877 at 30 or 65 mg/kg twice daily (BID) for 15 days. Similar plasma concentrations at 1 hour after dose were observed on days 1 and 15, and dose-related increases in exposure were observed for both 7915 and 0877 groups (table S2). The unbound brain/unbound plasma (Cub/Cup) ratio for 7915 was 0.26 ± 0.03, and was 0.48 ± 0.01 for 0877. For both molecules, the total concentrations in lung and kidney were similar to the total brain concentration. An estimated multiple of the observed exposure over the anticipated efficacious concentration (“fold above IC50”), calculated by dividing the average Cup by the pLRRK2 cell IC50, was used to provide relative exposure comparisons of LRRK2 kinase inhibition between molecules (notably, in vivo Cub IC50 values determined by a PD inhibition model using BAC Tg G2019S LRRK2 mice gave similar values to the pLRRK2 cell IC50; Table 1). Estimated 5- and 36-fold over the IC50 exposures were achieved with 7915 at 300 mg/kg BID and 0877 at 65 mg/kg BID, respectively.

Despite the differences in exposure, evidence of LRRK2 kinase inhibition was observed in brain, kidney, and lung in all 7915- and 0877-dosed groups as assessed by reduction in the ratio of pS935/total LRRK2 (Fig. 2 and fig. S5). [Note: pS935 (indirect) was used over pS1292 (direct) as a readout of LRRK2 kinase activity because pS1292 could not be detected in tissue lysates from animals with nonmutant Lrrk2 with our anti-pS1292 antibody.] Slightly greater reduction of pS935 was seen in groups given 0877, most notably in brain, consistent with higher exposures over the cell IC50. Whereas comparably significant decreases were observed in pS935 in lungs and kidneys with 7915 and 0877, reduced total LRRK2/actin ratios associated with decreased pS935 were only evident in kidneys and no microscopic effects were observed in lungs or kidneys in any dose group.

Fig. 2. Pharmacodynamic assessment of LRRK2 kinase inhibitors in mice.

(A and B) LRRK2 pS935/total LRRK2 and total LRRK2/actin ratios were measured to evaluate LRRK2 kinase inhibition in mouse brain, kidney, and lung after 15 days of dosing with (A) 7915 or (B) 0877. Dose groups administered LRRK2 inhibitors were compared to the vehicle control group using an unpaired t test. *P < 0.05; **P < 0.01; ***P < 0.001.

Safety studies in male and female Sprague-Dawley rats were conducted for each compound to also assess toxicity profiles. Male and female rats were orally administered 7915 at 10, 50, or 100 mg/kg or 0877 at 30, 75, or 200 mg/kg once daily (QD) for 7 days. Dose-related increases in exposure were observed for both 7915 and 0877 over the range of doses administered (table S3). Administration of 7915 was tolerated for 7 days up to 100 mg/kg per day, and no effects were observed in lungs or kidneys. Dosing of 0877 was not tolerated for 7 days at 75 or 200 mg/kg per day, with mortality and/or moribundity requiring euthanasia before end of study in both dose groups. Clinical and anatomical findings in moribund animals were nonspecific and consistent with physiologic stress and poor health status. No effects were observed in lungs or kidneys in any 0877 dose group.

Microscopic lung effect of 7915 in a 7-day safety study in NHPs

Given that 7915 was well tolerated for seven daily doses in rats at the highest dose tested (~20-fold over the cellular IC50), 7915 was evaluated in a single-administration, dose-ranging TK and tolerability study in cynomolgus monkeys designed to select doses for a repeat-dose study. Monkeys were subsequently administered 7915 at 10, 25, or 65 mg/kg QD via oral gavage for 7 days. The exposure of 7915 in monkeys generally increased with dose, and there were no sex-related differences (Table 2 and fig. S6). Similar exposures were observed on days 1 and 7 for the 10 mg/kg group, but day 7 exposure was higher than day 1 for the 25 and 60 mg/kg groups.

Table 2. TK data of 7915 in NHPs.
View this table:

Although monkeys maintained their body weight over the course of the study, several clinical signs were attributed to 7915 and were dose-related in terms of severity, incidence, and day of onset. Tremors were observed at ≥10 mg/kg per day, hypoactivity and decreased reactivity to stimulus were observed at ≥25 mg/kg per day, and additional abnormal behaviors were observed at 65 mg/kg per day. These clinical observations were transient (observed 3 to 9 hours after dose) and resolved before the subsequent dose.

Clinical pathology findings were limited to increases in plasma fibrinogen and serum cholesterol in monkeys administered ≥25 mg/kg per day. The only 7915-related anatomic pathology finding was increased vacuolation of type II pneumocytes in the lungs of both sexes administered ≥25 mg/kg per day (Fig. 3). Confirmation that the vacuolated cells were type II pneumocytes was made by immunohistochemical (IHC) staining for prosurfactant protein C (proSPC) (Fig. 3, A to D). Transmission electron microscopy (TEM) clarified the vacuolation as an increase in the size and number of lamellar bodies (Fig. 3, E to J). Image analysis was performed on lung tissue stained by IHC for proSPC to evaluate the change in cell size distribution in 7915 dose groups compared to controls. There was a dose-related increase in log10 median cell size (Fig. 3K).

Fig. 3. Effect of 7915 on NHP type II pneumocytes.

(A to D) Lung tissue stained with (A and C) hematoxylin and eosin and (B and D) IHC for proSPC from (A and B) a vehicle-dosed control monkey as a comparator, and from (C and D) a monkey dosed with 7915 (65 mg/kg). The 7915-dosed monkey has enlarged, vacuolated cells identified as type II pneumocytes (black arrows). All monkeys on the study [two per sex per group administered vehicle or 7915 (10, 25, or 65 mg/kg per day)] were evaluated, and these are representative microscopic images of the lung effect. Images were taken at ×600 original magnification. (E to J) Resin-embedded lung tissue from representative monkeys dosed with (E to G) vehicle or (H to J) 7915 (65 mg/kg) was also evaluated by (E and H) light microscopy and (F, G, I, and J) TEM. Images were taken at ×1000 original magnification for (E) and (H). Scale bars, 1 μm (F and I) and 0.5 μm (G and J). (K) Quantification of type II pneumocyte cell size was assessed by image analysis of proSPC IHC-positive cell staining areas. These are shown as size distribution curves (left), median log10 area (middle), and interquartile range (IQR) log10 area (right). Data are shown for all monkeys in the study, with group means plotted for median log10 area and IQR log10 area.

7915 was detected in brain and cerebrospinal fluid (CSF) samples collected at ~24 hours after day 7 dosing in all dose groups (fig. S6). The Cub/Cup ratio was 0.60 ± 0.16, and the unbound CSF concentration (CuCSF)/Cup ratio was 0.82 ± 0.24. These results demonstrate good central nervous system (CNS) penetration of 7915 in monkeys. Evidence of LRRK2 kinase inhibition in brain (striatum) was measured by a decrease in pS935 (10 mg/kg per day, P < 0.05; 65 mg/kg per day, P < 0.001) (Fig. 4A and fig. S7). A trend of decreased LRRK2 brain protein was observed in all dose groups.

Fig. 4. Pharmacodynamics of LRRK2 kinase inhibitors in NHPs.

(A to E) LRRK2 pS935/total LRRK2 and total LRRK2/actin ratios were measured to evaluate LRRK2 kinase inhibition in (A) brain (striatum) after 7 days of dosing with 7915, and in (B) brain (striatum), (C) kidney, (D) PBMCs, and (E) lung after 29 days of dosing with 0877 or 7915. Plots include all animals, except those for lung, where only animals with detectable total LRRK2 were included. Dose groups administered LRRK2 inhibitors were compared to the vehicle control group using an unpaired t test. *P < 0.05; **P < 0.01; ***P < 0.001.

A similar microscopic lung effect with 7915 and 0877 in a 29-day safety study in NHPs

A follow-up repeat-dose 29-day toxicity study in cynomolgus monkeys was performed to compare the toxicity profiles of 0877 and 7915 and to specifically assess whether the structurally distinct 0877 also resulted in lung toxicity. To adjust for the faster in vivo clearance of 0877, monkeys were orally administered 0877 at 6 or 20 mg/kg or 7915 at 30 mg/kg BID.

The exposures for both 7915 and 0877 were consistent with previous single- and/or multiple-dose studies (Table 3 and fig. S8). Higher exposure multiples over the IC50 were achieved with both 0877 dose groups compared to the 7915 dose group.

Table 3. Repeat-dose TK data of 0877 and 7915 in NHPs.

ND, not done.

View this table:

Clinical observations for 7915 at 30 mg/kg BID were similar to those observed in the previous study and suggestive of mild sedation and gastrointestinal (GI) effects. Clinical observations with 0877 at 20 mg/kg BID included hunched posture (also considered a possible sign of sedation) and mild GI effects.

There were no clinical pathology effects with 0877 at either dose. Consistent with the results from the 7-day study, 7915 at 30 mg/kg BID demonstrated a minimal increase in serum cholesterol concentrations in females. In addition, microscopic examination of blood smears of 7915-dosed monkeys showed an increase in monocyte cytoplasmic vacuolation in both sexes on day 29 (fig. S9). Evaluation of histologic sections did not reveal similar findings in tissue macrophages, or in cells of the lymphoid organs or bone marrow. Light microscopic and TEM evaluation of the lung confirmed similar lamellar body changes in type II pneumocytes in monkeys administered either LRRK2 inhibitor to those seen in the 7-day study with 7915 (Fig. 5, A and B). Erosion and ulceration of the stomach were observed in one animal per group given 0877 at 20 mg/kg BID and 7915 at 30 mg/kg BID. Although this finding was of low incidence and is observed incidentally in cynomolgus monkeys, a relationship to the test molecules could not be excluded.

Fig. 5. Effect of 0877 and 7915 on NHP type II pneumocytes.

(A and B) Lung tissue sections from a monkey dosed with 0877 (20 mg/kg BID) were examined by (A) light microscopy of toluidine blue–stained plastic sections and (B) TEM. All monkeys on the study were evaluated, and these are representative microscopic images of the lung effect of 0877, which is morphologically identical to that by 7915. Image taken at ×1000 original magnification for (A). Scale bar, 1 μm (B). (C) Quantification of type II pneumocyte cell size was assessed by image analysis of proSPC IHC-positive cell staining areas (as in Fig. 3K). These are shown as size distribution curves (left), median log10 area (middle), and IQR log10 area (right). Data are shown for all monkeys in the study, with group means plotted for median log10 area and IQR log10 area.

CSF, brain, lung, and kidney were collected for concentration analysis of 7915 and 0877. The CuCSF/Cup and Cub/Cup ratios were 0.83 ± 0.25 and 0.69 ± 0.21, respectively, for 7915, and 1.1 ± 0.4 and 0.46 ± 0.16, respectively, for 0877, indicating good CNS penetration for both molecules (fig. S8). The total concentrations in lungs and kidneys were similar to total brain concentrations for both compounds; that is, no unusual accumulation of either compound was detected in the lung.

Pharmacodynamic effects of 0877 and 7915 in the brain, as evaluated by pS935, were observed at 6 and 20 mg/kg BID with 0877 and at 30 mg/kg BID with 7915 (Fig. 4B and fig. S10A). The decrease in pS935 was greatest with 20 mg/kg BID of 0877 (P < 0.001) as expected given the highest fold over LRRK2 IC50 achieved in this group. This group also had a significant decrease in total LRRK2 (P < 0.05). All groups showed almost complete inhibition of pS935 in kidneys (Fig. 4C and fig. S10B) and peripheral blood mononuclear cells (PBMCs) (Fig. 4D and fig. S10 D). The trend toward a decrease in total LRRK2 was also observed in kidneys and PBMCs. Pharmacodynamic assessment of the lungs showed high variability of total LRRK2 in control animals but confirmed inhibition of pS935 in all inhibitor-dosed groups, where total LRRK2 was detectable (Fig. 4E and fig. S10C).

In kidneys, no light microscopic differences were observed between inhibitor-dosed groups and control animals; however, the suggestion of a decrease in cytoplasmic neutral lipid globules was noted by high-magnification light microscopy and TEM in monkeys given 0877 at 20 mg/kg BID and 7915 at 30 mg/kg BID (fig. S11). No effect on renal function was detected by routine clinical pathology assessments.

As in the 7-day study with 7915, the effect on lamellar bodies was also quantified in this study by measuring type II pneumocyte cytoplasmic area by digital image analysis of proSPC IHC-stained slides. A shift in type II pneumocyte log10 size distribution curves to the right was noted in the 0877 at 20 mg/kg BID and 7915 at 30 mg/kg BID groups compared to controls, with higher median cell sizes and a higher IQR for 7915 (Fig. 5C). The results confirmed the qualitative assessment of severity by pathologists; the group given 7915 had the greatest effect, and there was a dose-related effect with 0877. No increase in the effect was seen following the longer (29-day) dosing period of 7915 at 30 mg/kg BID 7915 compared to monkeys given 65 mg/kg QD in the 7-day study.

Because the vacuolation noted in pneumocytes with our LRRK2 kinase inhibitors was similar to that described for cationic amphiphilic drug (CAD)–induced phospholipidosis, as well as lysosomal storage diseases, we hypothesized that affected monkeys may show an increase in blood and/or urine di-docosahexaenoyl (22:6) bis(monoacylglycerol) phosphate (di-22:6-BMP) concentrations, a marker of lysosomal dysregulation, as has been reported in these conditions (1923). Di-22:6-BMP was measured by liquid chromatography–tandem mass spectrometry (LC-MS/MS) before dose and at necropsy in urine and plasma, and in CSF at necropsy only, for monkeys administered 7915 or 0877. The day 29 creatinine-adjusted urinary concentrations of di-22:6-BMP were lower with 0877 at 6 mg/kg BID (P < 0.05) and 20 mg/kg BID (P < 0.001) and with 7915 at 30 mg/kg BID (P < 0.001) compared to controls (Fig. 6A). The day 29 plasma concentrations of di-22:6-BMP in LRRK2 inhibitor groups were not lower than controls, but showed an average decrease of 1.9- and 2-fold with 0877 at 6 and 2 mg/kg BID, respectively, and an average decrease of 1.6-fold with 7915 at 30 mg/kg BID compared to baseline (Fig. 6B). No di-22:6-BMP concentration differences were observed in CSF in inhibitor-dosed monkeys compared to controls (fig. S12).

Fig. 6. Di-22:6-BMP in urine, plasma, and serum.

(A and B) Concentrations of di-22:6-BMP were measured in (A) urine and (B) plasma in all monkeys (two per sex per group) at baseline and after administration of LRRK2 kinase inhibitors 7915 and 0877 (day 29), with the exception of one animal in the 6 mg/kg BID 0877 group for which day 29 samples were not available. Baseline concentrations for urine were taken on day −8 and for plasma on days −8 and −2. The baseline value plotted for plasma (predose) is the average of the day −8 and day −2 measurements. (C and D) Concentrations of di-22:6-BMP were measured in (C) urine and (D) serum of WT versus Lrrk2 KO mice. P values were derived from unpaired t tests.

Because measurement of surfactant protein D has also been proposed as a serum biomarker for CAD-induced phospholipidosis in the lung (24), assessments of serum surfactant protein D between control animals and kinase inhibitor–dosed groups were performed. No changes in inhibitor-dosed groups were detected by enzyme-linked immunosorbent assay (ELISA) (table S4).

A similar decrease in di-22:6-BMP in Lrrk2 KO mouse urine

To follow up on the finding in monkeys, we also measured di-22:6-BMP by LC-MS/MS in urine and serum from Lrrk2 KO and WT mice. Similarly, KO mouse urine concentrations of di-22:6-BMP were significantly lower than WT (P < 0.05) with an average 17-fold decrease in urinary di-22:6-BMP (Fig. 6C). The serum concentrations of di-22:6-BMP were comparable in the Lrrk2 KO and WT animals (Fig. 6D). No differences were detected between Lrrk2 KO and WT mice with respect to surfactant protein A or D concentrations in serum (table S5).

DISCUSSION

The purpose of our work was to characterize the potential safety liabilities associated with targeting LRRK2 kinase activity in order to develop a small-molecule LRRK2 kinase inhibitor to treat PD. Similar to previous reports (1, 2, 25, 26), our human and mouse tissue mRNA expression analysis suggested that the effects of LRRK2 kinase inhibition were most likely in the nervous (brain), urinary (kidneys), respiratory (lungs), and immune (monocytes, neutrophils, B lymphocytes) systems. Phenotypic comparative analysis between Lrrk2 KO and WT mice substantiated the effects of LRRK2 deficiency on kidneys and lungs, as has been seen in a different line of Lrrk2 KO mice and in Lrrk2 KO rats (18, 2729). These data provide the basis for assessing whether the effects seen in our studies with two distinct and specific LRRK2 kinase inhibitors could be ascribed to on-target inhibition.

As was expected according to literature, we did not observe any morphologic changes in brain related to Lrrk2 KO in mouse, nor in LRRK2 inhibitor–dosed rodents or monkeys. Clinical observations considered consistent with sedation were observed in NHPs with both 7915 and 0877, but we concluded that these observations were unlikely to be due to LRRK2 kinase inhibition because there was no correlation between the severity of clinical signs (7915 ≫ 0877) and the relative degree of LRRK2 kinase inhibition expected at Cmax (0877 > 7915) or the measured pharmacodynamic effects (decrease in pS935) in brain (0877 > 7915). The cause of these clinical signs remains undetermined.

A robust pharmacodynamic effect on LRRK2 pS935 was seen in PBMCs in monkeys administered either of the small-molecule inhibitors. This provides corroboration for the utility of this phospho-epitope in PBMCs as a peripheral marker of LRRK2 kinase inhibition that could be used in clinical trials of PD. Dzamko et al. (30) reported that similar ratios of total LRRK2/actin and pS935/total LRRK2 were present in PBMCs of controls and idiopathic PD patients. These ratios were not affected by age, disease duration, disease severity, or levodopa medication, providing further support for using this phospho-epitope as a clinical pharmacodynamic marker. It is interesting to note the increased vacuolation of peripheral monocytes after 7915 (30 mg/kg BID) for 29 days because monocytes have high LRRK2 expression; however, this finding was considered most likely an off-target effect of 7915 given that it was not seen with 0877 or in Lrrk2 KO mice.

The effect of greatest concern for our LRRK2 kinase inhibitors was the abnormal accumulation of lamellar bodies in type II pneumocytes in NHPs. This finding in lung tissue was morphologically identical to that seen in Lrrk2 KO mice. Hence, we propose that this is an on-target effect of inhibiting LRRK2 function. Assessment of lung pS935 showed marked pharmacologic inhibition by both molecules at necropsy. The reason for a slightly greater morphologic effect by 7915 compared to 0877 at both doses as measured by image analysis evaluations is not clear, because pS935 reduction in lung is comparable between all three dose groups, and fold over IC50 calculated for 7915 is lower than that for either dose of 0877. One possibility is that 7915 has an off-target effect that compounds lysosomal dysregulation. Evidence for this may be the increase in monocyte cytoplasmic vacuolation observed only with 7915. Another possibility, although not considered likely, is that the effect on the lungs in monkeys seen after administration of our LRRK2 inhibitors may not be caused by kinase activity inhibition per se, but by an altering of another aspect of LRRK2 function. The absence of a similar lung effect in Lrrk2 kinase-dead (KD) mice as reported by Herzig et al. (18) supports this possibility, but the discrepancy of the lung phenotype between KO and KD is more likely due to incomplete loss of LRRK2 kinase function in the KD model. Needless to say, evaluation of additional structurally distinct LRRK2 kinase inhibitors in monkeys would add confidence to the assessment that this pneumocyte effect is due to on-target kinase inhibition.

We did not observe a similar change in lamellar bodies in type II pneumocytes in mice or rats dosed with 7915 or 0877, despite reexamining plastic-embedded and toluidine blue–stained lung sections at ×1000 magnification. The lack of effect on lungs in mice, particularly those dosed with 0877, at drug concentrations comparable by fold over IC50 inhibition to that at which the effect was observed in NHPs, and in which a similar degree of pS935 reduction in lung was observed, suggests that mice may need almost complete inhibition of LRRK2 for this effect. Unfortunately, both inhibitors were absorption-limited in mice, so testing higher exposures was not possible. In rats, 7915 and 0877 drug concentrations achieved greater fold over IC50 inhibition than that at which the lung effect was seen in monkeys (although pS935 in lung was not assessed) and approached or went beyond the limit of tolerability for both molecules. Recently, published reports indicate that even Lrrk2 KO rats do not manifest a type II pneumocyte vacuolar change when assessed by light microscopy; however, a more subtle finding was observed—an increase in size and number of lamellar bodies by TEM and subcellular imaging of primary type II pneumocytes (28, 29, 31). Similar to mice, we attribute the absence of the change in rats to an inherently lower sensitivity to this LRRK2-mediated effect (that is, complete inhibition is necessary for even a subtle effect).

No renal toxicity by pharmacologic LRRK2 kinase inhibition was observed in any species. In contrast to cytoplasmic accumulation of secondary lysosomes in Lrrk2 KO and KD mice (18), no effect on lysosomal organelles was observed in inhibitor-dosed rats or mice, despite confirmation of LRRK2 kinase inhibition and reduction in total LRRK2/actin ratios in mice. One explanation that is similar to that for the lung is that the decreases in pS935 induced by both inhibitors were not sufficient to recapitulate the effect of the Lrrk2 KO or KD phenotype. Monkeys given 7915 or 0877 did not accumulate secondary lysosomes in proximal tubular epithelium, despite near-complete inhibition of pS935 in the kidney. In the kidney, in contrast to the lung, it seems possible that NHPs may be less sensitive than rodents. The only notable finding on TEM was a decrease in neutral lipid globules in the proximal tubular epithelium of monkeys given the higher doses of 0877 and 7915 compared with controls, the pathogenesis of which is unclear.

Genetic deletion and pharmacologic inhibition of LRRK2 kinase activity resulted in a reduction of urine di-22:6-BMP, a phospholipid localized within the internal membrane of lysosomes and late endosomes, which functions in lysosomal degradation (32). Di-22:6-BMP has been shown to increase in the blood and urine in CAD-induced phospholipidosis and in lysosomal storage diseases, where there is altered trafficking and/or function of the lysosomal compartment (1923). Because the vacuolation noted in pneumocytes with our LRRK2 kinase inhibitors was similar to those conditions, we anticipated a possible increase in di-22:6-BMP in Lrrk2 KO mice and kinase inhibitor–dosed monkeys, but found the opposite—a unique demonstration of a decrease in urine di-22:6-BMP. This decrease in di-22:6-BMP further suggests an important role of LRRK2 in lysosomal regulation and related autophagic mechanisms. A decrease of di-22:6-BMP was not observed in the serum of Lrrk2 KO mice, suggesting that the decrease in urine primarily reflects the kidney phenotype. Mean di-22:6-BMP was lower compared to predose in all inhibitor-dosed groups in monkey plasma and may reflect the effect on lung. Although additional evaluations are needed, these results establish di-22:6-BMP as a potential safety marker of LRRK2 kinase inhibition. Its utility as a CNS pharmacodynamic biomarker is unclear, because no effect on CSF di-22:6-BMP was observed in inhibitor-dosed monkeys.

Our results suggest that lung toxicity may be the primary clinical safety liability of LRRK2 kinase inhibitors in patients. The fact that this preclinical safety finding was observed in NHPs but not in rodents highlights the importance of species selection when assessing LRRK2 kinase inhibitors and the necessity of evaluating preclinical safety in NHPs (because it is not known whether dogs would predict this lung effect). Although affected monkeys did not show pulmonary-based clinical signs or a progressive change following longer duration of dosing, and Lrrk2 KO mice do not develop progressive lung disease, theoretical concerns based on genetic diseases affecting lamellar bodies and toxicities observed with CAD-induced phospholipidosis include acute and chronic pulmonary sequelae (3336). Given the uncertainty about short-term clinical consequences or longer-term sequelae of type II pneumocyte lamellar body accumulation, the safety risk to PD patients is unclear.

The data also highlight a role for LRRK2 in regulating lamellar bodies in lung and, perhaps more generally, in lysosomal regulation. The identification of these two otherwise well-tolerated, highly selective LRRK2 kinase inhibitors provides a means to better understand the biological function of LRRK2 in tissues where it is expressed and, most importantly, in the pathogenesis of familial and sporadic PD.

MATERIALS AND METHODS

Study design

The purpose of this study was to assess the safety risk of inhibiting LRRK2 kinase activity by in vitro and in vivo evaluations. LRRK2 expression in human was queried in an mRNA expression database, and murine tissues were evaluated by non-isotopic ISH. Lrrk2 KO transgenic mice were generated to further assess potential safety concerns of inhibiting LRRK2. The preclinical pharmacology, toxicity, and TK profiles of two structurally distinct LRRK2-specific, brain-penetrating kinase inhibitors were then evaluated in rodents and NHPs. Because they both were characterized by single-digit nanomolar biochemical and cellular activities, broad kinome and receptor selectivity, and good in vivo oral exposure and unbound brain concentrations across preclinical species, they were appropriate molecules for evaluating the consequences of inhibiting LRRK2 kinase activity in rodent and monkey pharmacodynamic and safety assessments. For in vivo LRRK2 inhibitor studies, animals were assigned randomly to groups. Group sample sizes of eight male mice, six rats per sex, or two monkeys per sex were used to account for biological variability among animals and the need for reliable statistical analysis yet be balanced with the ethical use of animals. All analyses were predetermined by experimental protocol or amendments. Except where noted, all data were included and no animals were excluded from the study.

Animal welfare

All rodent studies were performed in accordance with Institutional Animal Care and Use Committee–approved guidelines. Rodents had free access to water and Rodent Diet 5010 (LabDiet) for the duration of all studies. The cynomolgus monkey studies were conducted at Covance Laboratories Inc. (in Chandler, AZ, or Madison, WI) in compliance with the Animal Welfare Act, the Guide for the Care and Use of Laboratory Animals, and the Office of Laboratory Animal Welfare. Monkeys had free access to water and were fed Certified Primate Diet #2055C (Harlan Laboratories Inc.) one or two times daily and given various cage enrichment devices and fruits, vegetables, or dietary enrichment for the duration of the studies.

Lrrk2 KO (Lrrk2 EGFP KI) mouse generation

The construct for targeting the Lrrk2 locus in embryonic stem (ES) cells was made using a combination of recombinant engineering (37) and standard molecular cloning techniques (fig. S13). From a mouse BAC clone containing the Lrrk2 genomic region, a 5944–base pair (bp) fragment containing Lrrk2 exons 1 + 2 (chr15:91,670,808-91,676,751 GRCm38/mm10 assembly) was retrieved into the plasmid pBlight-TK (37). Next, a cassette containing homology to the 5′ untranslated region in exon 1, an EGFP cassette with an SV40 pA, a loxP-PGK-em7-Neo-BGHpA-loxP selection marker, and intron 1 homology was used to delete the Lrrk2 exon 1 coding region and the first 30 bp of intron 1 and replace it with an EGFP cassette exactly at the Lrrk2 ATG, giving rise to the final targeting vector. This vector was confirmed by DNA sequencing, linearized with Not I, and used to target C57BL/6 C2 ES cells using standard methods (G418-positive and ganciclovir-negative selection). Positive clones were identified using polymerase chain reaction (PCR) and TaqMan analysis and confirmed by sequencing of the modified locus. Correctly targeted ES cells were transfected with a Cre plasmid to remove the Neo selection marker and create the final Lrrk2 EGFP KI allele. KI ES cells were injected into blastocysts using standard techniques, and germline transmission was obtained after crossing resulting chimeras with C57BL/6N females.

Animals were genotyped using a three-primer approach, the Qiagen Type-it Fast SNP PCR master mix, and the following conditions: 1 cycle of 94°C for 4 min followed by 30 cycles of 94°C for 1 min, 60°C for 30 s, 72°C for 1 min, and finally 1 cycle of 72°C for 10 min. The following primers were used: Lrrk2, 5′-CAACGCCCCTTTGCTATTC-3′ (forward) and 5′-ACAATCAACAGAGGCACGTG-3′ (reverse); EGFP, 5′-GGCATGGACGAGCTGTACA-3′ (reverse). Expected sizes were as follows: WT, 618 bp; KI/KO, 496 bp.

LRRK2 KO and EGFP expression were confirmed in Lrrk2 EGFP KI mice by Western blot, immunofluorescence, and ISH.

LRRK2 KO and WT mouse tissue sampling

Tissues were collected for routine histology, IHC, and electron microscopy from five Lrrk2 KO and five WT female mice at 28 to 35 weeks of age and from three Lrrk2 KO and three WT male mice at 40 to 48 weeks of age.

IHC for proSPC

IHC for proSPC was performed on 4-μm formalin-fixed, paraffin-embedded sections from mice and NHP tissues using a rabbit polyclonal antibody (catalog no. ab3786; Millipore). For the mouse tissues, sections were deparaffinized in xylene and rehydrated through graded alcohols to distilled water. Sections were pretreated with target retrieval solution for 20 min at 99°C followed by 3% hydrogen peroxide in phosphate-buffered saline (PBS). Sections were then treated with avidin/biotin block and 10% donkey serum containing 3% bovine serum albumin in PBS, respectively. Sections were incubated for 1 hour at room temperature with anti-proSPC antibody at 1:4000 followed by a donkey anti-rabbit biotinylated secondary antibody (catalog no. 711-065-152; Jackson ImmunoResearch) for 30 min. Sections were incubated in Vectastain ABC Elite reagent (Vector) followed by Pierce metal-enhanced diaminobenzidine (DAB), and sections were counterstained, dehydrated, and coverslipped. For NHP tissues, all steps were performed on the Ventana Discovery XT autostainer. Pretreatment was done with Cell Conditioner 1 mild followed by proSPC at 1:2000 for 1 hour at 37°C. Ventana Rabbit OmniMap was used as the detection system, and Ventana DAB and Hematoxylin II were used for chromogenic detection and counterstain. Isotype control antibodies were used to determine the specificity of the IHC assay.

Electron microscopy

Lung and kidney tissues were fixed in 1/2 Karnovsky’s fixative (2% paraformaldehyde, 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.2), washed in the same buffer, and postfixed in 1% osmium tetroxide. The samples were then dehydrated through a series of graded ethanols, followed by propylene oxide, and embedded in Eponate 12 (Ted Pella). Semithin sections (300 μm) were cut on a Leica Ultracut UCT (Leica Microsystems) and stained with toluidine blue for light microscopic examination. Thin sections (70 μm) were stained with uranyl acetate and lead citrate and examined in a JEOL JEM-1400 transmission electron microscope.

Digital image analysis

proSPC-stained lung sections were digitized at ×40 magnification on a SCN400 bright-field whole-slide scanner (Leica Biosystems). Image analysis of the digitized sections was performed using Developer XD 2.1 software (Definiens). For each animal, the central tendency and heterogeneity of cell size were summarized by the median log10 and IQR log10 of the stain-positive cell area distribution.

In vitro assessments of 7915 and 0877

pLRRK2 cell IC50 was determined by quantifying immunoblots of pS1292 in HEK293 cells transfected with R1441G/Y1699C/G2019S LRRK2. In vitro off-target selectivity profiles (DiscoveRx, Invitrogen, and Cerep) of both LRRK2 inhibitors were assessed using 0.1 to 10 μM dimethyl sulfoxide stock solutions (>95% chemical and optical purity by high-performance liquid chromatography).

For all animal studies, both LRRK2 inhibitors were administered as free base suspensions in vehicle [1% (w/v) Avicel RC-591 and 0.2% (v/v) polysorbate 80 (Tween 80) in reverse osmosis water].

Pharmacology studies in mice with 7915 and 0877

Male C57BL/6 mice 10 to 12 weeks of age were obtained from Charles River Laboratories. Eight animals per group were orally administered the vehicle control article or 7915 (200 or 300 mg/kg BID) or 0877 (30 or 65 mg/kg BID) at a dose volume of 5 ml/kg for 15 days, with necropsy performed on day 15 at 1 hour after dose (~Cmax). Five animals per group were used for toxicity assessments, and three animals per group for TK analyses. Toxicity evaluation was based on clinical observations, body weights, clinical pathology (cholesterol and triglycerides), organ weights (brain, kidneys, lungs, liver, and spleen), TK analyses, and macroscopic and microscopic anatomic pathology (lungs and kidneys).

Mouse Western blots (brain, lungs, and kidney)

Frozen tissues were homogenized with TissueLyser (Qiagen) in lysis buffer [50 mM tris-HCl (pH7.4), 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 2 mM EDTA] containing cOmplete phosphatase and PhosSTOP protease inhibitor cocktails (Roche). Tissue lysate was cleared via centrifugation at 20,000g for 30 min at 4°C. Protein concentration of the supernatant was measured with BCA (bicinchoninic acid) assay (Pierce). Lysate (30 μg) was loaded onto 3 to 8% tris-acetate gels or 4 to 12% bis-tris gels (Invitrogen). Samples were transferred onto nitrocellulose membrane and immunoblotted with the following antibodies: anti-pS935-LRRK2 (catalog no. 5099-1; Epitomics), anti-LRRK2 (catalog no. 3514-1; Epitomics), and anti-actin (catalog no. A3853; Sigma). The LI-COR Odyssey system was used for Western blot detection and quantitation. Unpaired t tests were conducted to compare LRRK2 kinase inhibitor–dosed groups with vehicle control group.

Rat safety studies with 7915 and 0877

Male and female Sprague-Dawley rats (Rattus norvegicus) 8 to 12 weeks of age were obtained from Charles River Laboratories. Three rats per sex per group were administered the vehicle control article or 7915 (10, 50, or 100 mg/kg QD) or 0877 (30, 75, or 200 mg/kg QD) orally via gastric lavage at a dose volume of 5 ml/kg for 7 days and necropsied on day 8 (~Cmin). An additional three rats per sex per group were similarly dosed and assessed for TK on days 1 and 7. Toxicity evaluation was based on clinical observations, body weights, clinical pathology (hematology, clinical chemistry), organ weights, and macroscopic and microscopic anatomic pathology [brain, cecum, colon, duodenum, heart, ileum, jejunum, kidneys, liver, lungs, lymph node (mesenteric), rectum, skeletal muscle (quadriceps femoris), spleen, sternum with bone marrow, stomach, and thymus].

NHP 7-day safety study with 7915

Sixteen male and female cynomolgus monkeys (n = 2 per sex per group), 2 to 5 years old and 2 to 5 kg, were assigned to four groups and administered the vehicle control article or 7915 at 10, 25, or 65 mg/kg QD via oral gavage at a dose volume of 5 ml/kg for 7 days and necropsied on day 8 (~Cmin). Blood was collected on days 1 and 7 before dose and about 1, 3, 6, 9, and 24 hours after dose for TK evaluation. Assessment of toxicity was based on clinical signs, body weights, qualitative assessment of food consumption, body temperatures, and neurological examinations. Clinical pathology evaluations included hematology, coagulation, and clinical chemistry. In addition, monocyte cytoplasmic vacuolation was assessed by counting the number of large (≥1 μm in diameter), clear vacuoles per 20 monocytes on peripheral blood smears. Anatomic pathology evaluations included organ weights, macroscopic observations, and microscopic evaluations (adrenal gland, aorta, brain and pituitary gland, spinal cord, eye, esophagus, stomach, small and large intestines, liver and gallbladder, heart, kidney, lung, lymph nodes, skeletal muscle, pancreas, salivary gland, sciatic nerve, skin, femur with bone marrow, sternum with bone marrow, spleen, thymus, thyroid and parathyroid glands, tongue, trachea, urinary bladder, and male and female reproductive tracts). Blood, cerebral spinal fluid, and brain samples were collected for TK evaluation, and brain samples were collected for pharmacodynamic evaluations.

NHP 29-day safety study with 7915 and 0877

Sixteen male and female cynomolgus monkeys (n = 2 per sex per group), 2 to 4 years old and 2.6 to 3.4 kg, were assigned to four groups and administered the vehicle control article or 0877 at 6 or 20 mg/kg or 7915 at 30 mg/kg. All animals were dosed via oral gavage twice daily (about 12 hours apart) at a dose volume of 5 ml/kg on days 1 to 28 and once on day 29. Animals were necropsied about 4 hours after dose on day 29 (~Cmax). Evaluation of toxicity was based on clinical observations, body weights, qualitative assessment of food consumption, ophthalmic examinations, physical and neurological examinations, vital signs, blood pressure assessments, and electrocardiogram evaluation. Clinical and anatomic pathology assessments were as described for the 7-day safety study. Blood, urine, CSF, and brain, lung, and kidney samples were collected for TK, biomarker assessments, and/or pharmacodynamic evaluations.

NHP Western blots (brain, lungs, and kidney)

Frozen cynomolgus tissues were placed in a mortar in liquid nitrogen, and the tissue was pulverized into powder using a pestle on dry ice. Tissue powder was then homogenized with TissueLyser (Qiagen) in lysis buffer [50 mM tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 2 mM EDTA] containing cOmplete phosphatase and PhosSTOP protease inhibitor cocktails (Roche). Tissue lysate was cleared via centrifugation at 20,000g for 30 min at 4°C. Protein concentration of the supernatant was measured with BCA assay (Pierce). Lysate (30 μg) was loaded onto 3 to 8% tris-acetate gels or 4 to 12% bis-tris gels (Invitrogen). Samples were transferred onto nitrocellulose membrane and immunoblotted with the following antibodies: anti-pS935-LRRK2 (catalog no. 5099-1; Epitomics), anti-LRRK2 (catalog no. 3514-1; Epitomics), and anti-actin (catalog no. A3853; Sigma). The LI-COR Odyssey system was used for Western blot detection and quantitation. Unpaired t tests were conducted to compare LRRK2 kinase inhibitor–dosed groups with vehicle control group.

NHP Western blots (PBMCs)

PBMCs were isolated from whole blood by density gradient separation using 90% Ficoll-Paque PLUS (Sigma-Aldrich), and the resulting PBMC pellet was snap-frozen. Cells were lysed in lithium dodecyl sulfate sample buffer containing sample reducing agent (Invitrogen) and cOmplete phosphatase and PhosSTOP protease inhibitor cocktails (Roche). To reduce viscosity of the samples, the lysates were incubated with Benzonase (Sigma-Aldrich) and run on NuPAGE Novex 3 to 8% tris-acetate gels (Invitrogen). Samples were transferred onto polyvinylidene difluoride membranes and probed with the following antibodies: anti-pS935-LRRK2 (catalog no. 5099-1; Epitomics), anti-LRRK2 (catalog no. 75-253; NeuroMab), and anti–β-actin (catalog no. 4967; Cell Signaling); chemiluminescent detection (Pierce Dura) was then performed. Immunoblots were imaged using a GE ImageQuant LAS 4000 digital imaging system, and quantitation was performed using ImageQuant TL software. Unpaired t tests were conducted to compare LRRK2 kinase inhibitor–dosed groups with vehicle control group.

Di-docosahexaenoyl (22:6) bis(monoacylglycerol) phosphate

The extraction and quantitation of di-22:6-BMP by LC-MS/MS analysis in the urine, plasma, and serum samples were conducted by Nextcea Inc. Unpaired t tests were conducted to compare LRRK2 kinase inhibitor–dosed groups with vehicle control group and to compare Lrrk2 KO with WT mice.

Mouse biomarker collections

Serum and urine were collected for biomarker assessments (di-22:6-BMP and surfactant proteins A and D) from two male and seven female Lrrk2 KO and four male and six female WT mice at 17 to 24 weeks of age. Urine was analyzed for total protein and creatinine on a Cobas Integra 400 chemistry analyzer (Roche Diagnostics).

Statistical analysis

All individual animal values were plotted with group means, and P values were assessed by unpaired t tests conducted to compare each LRRK2 kinase inhibitor–dosed group with the vehicle control group.

SUPPLEMENTARY MATERIALS

www.sciencetranslationalmedicine.org/cgi/content/full/7/273/273ra15/DC1

Materials and Methods

Fig. S1. LRRK2 mRNA expression in human tissues.

Fig. S2. LRRK2 mRNA expression in selected WT mouse tissues.

Fig. S3. Lung phenotype of Lrrk2 KO mice.

Fig. S4. Kidney phenotype of Lrrk2 KO mice.

Fig. S5. Western blots of mouse pharmacodynamic data.

Fig. S6. Plasma pharmacokinetic and brain and CSF data of 7915 in NHPs from 7-day study.

Fig. S7. Western blots of NHP brain pharmacodynamic data from 7-day study.

Fig. S8. Plasma pharmacokinetic and brain and CSF data of 0877 and 7915 in NHPs from 29-day study.

Fig. S9. Effect of 7915 on NHP monocytes.

Fig. S10. Western blots of NHP pharmacodynamic data from 29-day study.

Fig. S11. Effect of 0877 and 7915 on NHP kidney.

Fig. S12. Di-22:6-BMP in NHP CSF.

Fig. S13. Diagram of gene targeting strategy for Lrrk2 EGFP KI mice (Lrrk2 KO).

Table S1. Kinase profile data from Invitrogen at 1 μM.

Table S2. TK data of 7915 and 0877 in mice.

Table S3. TK data of 7915 and 0877 in rats.

Table S4. Monkey surfactant protein D ELISA results.

Table S5. Results for surfactant protein A (SpA) and surfactant protein D (SpD) in mouse serum.

Reference (38)

REFERENCES AND NOTES

Acknowledgments: We thank the pathology core laboratories at Genentech for their support of this project; G. Cosma, D. Dambach, N. Dybdal, and D. Diaz for guidance on the safety studies; R. Hall for legal assistance; H. Taylor and S. Tadesse-Bell for help with graphs; L. Berezhkovsky, X. Ding, and X. Zhang for assistance with pharmacokinetic data; R. Pattni for targeting vector construction and members of the core laboratories in the Department of Transgenic Technology for technical assistance; F. Hsieh and E. Tengstrand of Nextcea Inc. (Woburn, MA) for assistance with the di-22:6-BMP LC-MS/MS analysis; and Covance Laboratories (Madison, WI) for the conduct of the NHP studies. Funding: This work was funded by Genentech Inc. and The Michael J. Fox Foundation for Parkinson’s Research. Author contributions: R.N.F., M.F., M.A.S.B., B.K.F., J.P.L., T.S., A.U., H.Z., A.A.E., and R.J.W. designed the project. R.N.F., M.F., M.A.S.B., L.H., P.K., D.W.L., T.L., S.L., X.L., J.P.L., J.O., A.S., J.T., H.Z., A.A.E., and R.J.W. designed, performed, oversaw, and analyzed the safety studies. S.W. and M.R.-G. designed and produced the Lrrk2 KO mouse. A.M.J., M.B., M.R., M.S., and M.Y. performed the evaluations on the transgenic mice and human tissue expression and provided additional pathology analyses on the safety studies. R.N.F. and J.B. performed the digital pathology image analysis. S.-C.L.-K. performed the statistical analyses. B.K.C., Z.K.S., A.A.E., and J.P.L. designed the LRRK2 kinase inhibitors. L.H., A.S., H.Z., Z.S., and S.Z. performed the PD analyses. R.N.F., X.L., A.A.E., and R.J.W. wrote the manuscript. Competing interests: Most of the authors are employees of Genentech Inc. and hold equity in Roche. A.U., B.K.F., M.A.S.B., and T.S. are employees of The Michael J. Fox Foundation for Parkinson’s Research and do not have any financial or competing interests in the results of this study.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article