Parkinson’s disease (PD) is the second most common neurodegenerative disease, affecting 1-2% of the population over the age of 60 [2,3]. The cardinal clinical features include tremor, rigidity, bradykinesia and/or postural instability, as well as neuropathological loss of dopaminergic neurons in the substantia nigra (SN), decreased dopamine (DA) neurotransmission, and the presence of neuronal intracellular Lewy body (LB) inclusions [2]. In addition, non-motor features such as depression, constipation, pain, and sleep disorders are important manifestations of the disease [4].

Historically believed to have no strong genetic component, genetic mutation or variation in a number of genes is now recognized as causal or risk-associated factors involved in a growing number of PD cases [5–7]. Mutations in the leucine-rich repeat kinase 2 (LRRK2) gene are the most common cause of familial and late-onset PD identified to date [3]. The most common LRRK2 mutation, G2019S, accounts for as much as 30-40% of Parkinsonism in Ashkenazi Jews and North African Arab-Berber populations [8,9]. Furthermore, LRRK2 mutations account for up to 2% of sporadic Parkinsonism [10]. The LRRK2 gene encodes a large multi-domain protein containing an ankyrin repeat region, a leucine-rich repeat domain, a Ras of complex protein GTPase domain, a C-terminal of Roc domain, a kinase domain, and a WD40 domain [11]. The LRRK2 G2019S mutation in the kinase domain appears to increase its enzymatic activity [12] and since LRRK2-related PD and sporadic PD display a similar phenotype [13], pharmaceutical companies are pursuing LRRK2 kinase inhibitors to reduce this gain-of-function as a promising therapeutic option for people with PD.

To be viable for human therapeutic development, drug makers must demonstrate that inhibition of LRRK2 activity is safe. In the absence of optimal tool compounds (i.e., potent and selective to LRRK2), researchers have utilized genetically modified rodent models to explore potential liabilities of targeting LRRK2 kinase activity. Studies in LRRK2-deficient mice have found morphological and histopathological abnormalities in both kidney and lung tissue that have been associated with impairments in the autophagy pathway [14–17]. LRRK2 knockout (KO) mice display large kidneys that are dark red with microscopic presence of microvacuoles in the proximal tubule epithelial cells. A lung phenotype (increased number and size of lamellar bodies) has also been found in LRRK2 KO but not kinase-dead (KD) mice, suggesting that the LRRK2 protein-protein binding domains, rather than the kinase domain, may be crucial for normal lung function [16]. However, the LRRK2 mouse studies that have been published to date have not examined any clinical chemistry or other biomarkers that may be associated with these deficits This information could be critical to guiding the development of appropriate safety measures for future clinical trials.

Recently, it was reported that LRRK2 KO rats at 4-months of age exhibit perturbations in renal morphology accompanied by significant decreases of lipocalin-2 (NGAL) in both urine and plasma [1]. Although consistent with reports in KO mice, this finding is inconsistent with renal damage since an increase in NGAL is an early responder of nephrotoxicity and tubular damage. The authors speculate that the decrease in NGAL may be independent of renal function but associated with alterations of immune homeostasis [1]. Furthermore, significant alterations in the cellular composition of the spleen between LRRK2 KO rats and wild-type (WT) animals were detected with subtle differences in response to dual infection with rat-adapted influenza virus and Streptococcus pneumoniae. A molecular pathway analysis of LRRK2 revealed links between LRRK2 and the thioredoxin system, which interacts with PRDX3, TXNIP, and TXNRD1. These proteins are associated with nutrient sensing, adiposity, and human obesity. The authors suggest that there might be a link between the reported LRRK2 KO weight gain, LRRK2 deficiency, and the thioredoxin pathway [1].

Given that this characterization of LRRK2 KO rats was limited to one age and there is still ambiguity regarding the clinical pathology markers associated with the LRRK2 KO renal phenotype, the present study extends these findings by examining the morphology, histopathology, ultrastructure, blood, and urine chemistry in LRRK2 KO and WT rats in 6 different age groups spanning a 16-month period.

Ethics Statement: All animal work in these studies is in compliance with the National Institutes of Health for humane animal welfare and has been approved by WIL Research and VA Medical Center/Portland IACUC committees.

The evidence is unequivocal that LRRK2 KO mice and rats exhibit an abnormal kidney phenotype [1,15–17]. We have replicated these findings but also found that these abnormalities in the LRRK2 KO rat progress with age, coincide with clinical pathology biomarkers, and extend to the lung and liver. Interestingly, the youngest LRRK2 KO cohort examined (1-month old), display clinical pathology that is not observed with gross examination, histochemical or immunohistochemical stains. Prior to the emergence of any abnormal phenotype, this cohort exhibits cholesterol, creatinine, phosphorous, chloride, sodium, and SDH alterations (see Table 3). Furthermore, the lung and liver of the oldest cohort examined (16-month old) display abnormal ultrastructure phenotypes that have never been previously reported. This highlights the importance in examining a wide-range of age groups and employing a myriad of techniques to uncover LRRK2 KO induced phenotypes. While we have replicated some of the morphological and clinical pathology findings found in the Ness et al. study [1] (e.g., increased body weight; altered cholesterol, red blood cell counts, and hematocrit percentage), extending the analysis to other age groups has further uncovered alterations in kidney, lung, and liver. It should be noted though that the liver findings appear to reflect more of a metabolic process abnormality than a lysosomal type observed in the kidney and lung since the EM liver results of increased lipid droplets are similar to the renal glomeruli phenotype.

The identification of age-related phenotypes in the LRRK2 KO rats has important implications. First, it suggests that LRRK2 deficiency has deleterious effects over time that may first emerge prior to any gross morphological alterations. These early peripheral (e.g., blood or urine) signals may become safety biomarkers for future LRRK2 kinase inhibitor clinical trials. Secondly, it facilitates the selection of which LRRK2 KO rat aged cohorts to use for pharmacological mechanism-based safety studies. Examining potential on- or off-target effects of LRRK2 kinase inhibitors requires a LRRK2 KO animal model with a phenotype that will not mask potential safety liabilities of LRRK2 kinase inhibitors. The 1-month old LRRK2 KO cohort may be a better animal model than 2-, 4-, 8-, 12-, or 16-month old animals for LRRK2 kinase inhibitor safety experiments as it exhibits a milder phenotype with regards to gross morphology and histopathology.

The challenge remains to ascertain the therapeutic window for a LRRK2 kinase inhibitor. It is important to note that all of the present studies were conducted using homozygous LRRK2 KO rats. Given that the LRRK2 heterozygous KO mouse kidney is devoid of kidney abnormalities and the lung abnormality is only associated with KO and not KD mice [16], pharmacological LRRK2 kinase inhibition of less than 50% may be tolerable. To ascertain this safety window, predictive safety and efficacy animal models are needed to determine the minimal amount of LRRK2 kinase inhibition that is required for the treatment of Parkinson’s disease. One challenge in developing a LRRK2 kinase inhibitor is that there is no robust in vivo model and only a few pharmacodynamic (PD) readouts (e.g., pSer935 and pSer1292) that can be used to screen the efficacy of potential LRRK2 kinase inhibitors. Without knowing the minimal LRRK2 kinase inhibition that is required to obtain efficacy, a therapeutic index is unobtainable.

It is plausible, however, that genetically induced abnormal phenotypes in rodents may not translate to other species (e.g., dog, non-human primates, and humans) and/or be predictive of LRRK2 pharmacological induced toxicity. For example, the elevated cholesterol observed in the LRRK2 KO rats is not a good model of human cholesterol related diseases such as atherosclerosis [25] since rat serum cholesterol is primarily composed of high density lipoproteins. Also, elevated SDH (a marker of liver damage) observed in LRRK2 KO rats was not associated with hepatocellular degeneration. SDH is expressed in the kidney, but typically has its highest expression in the regions of the kidney that did not show any abnormalities in the LRRK2 KO rats (i.e., glomeruli and distal convoluted tubules) [26]. Therefore, future experiments need to determine the implications of chronic exposure to potent and selective LRRK2 kinase inhibitors on both rodent and non-rodent species.

If these pathological observations are related to LRRK2 kinase inhibition in non-rodent species and are predictive of clinical pathology then the identification of cerebral spinal fluid (CSF) biomarkers along with peripheral (e.g., blood and/or urine) safety markers could be crucial for the development of LRRK2 kinase inhibitors. One recent promising PD approach is to measure LRRK2 released from exosomes in the CSF and urine [27]. The prediction is that LRRK2 kinase inhibitors will diminish the total LRRK2 levels secreted into exosomes and allow for the measurement of LRRK2 target engagement from accessible sampling compartments [27]. A CSF biomarker would especially facilitate a first-in-class LRRK2 kinase inhibitor human trial by allowing the clinician to monitor the relationship between brain LRRK2 kinase activity and safety.

Although LRRK2 genetic rodent evidence suggests potential issues in inhibiting LRRK2 kinase activity, it is important to note that none of the LRRK2 KO induced phenotypes reported to date translate to detrimental functional deficits [1,15–17]. Further pre-clinical studies examining pharmacological inhibition of kinase activity in non-rodent species and the identification of safety/efficacy biomarkers are needed. With pharmaceutical companies making advances in developing LRRK2 kinase inhibitors, it is crucial that we exhaust all means to bring a safe drug into the clinic.