The New p.F1700L LRRK2 Variant Causes Parkinson's Disease by Extensively Increasing Kinase Activity

Variants in LRRK2 represent the most frequent cause of clinically classical monogenic Parkinson’s disease (PD). Altered LRRK2 protein function boosts neuroinflammation, impairs vesicle trafficking, and affects ciliogenesis within the striatum. Because a relevant fraction of the more than 1000 identified LRRK2 variants is not pathogenic, determining pathogenicity for single variants is crucial, particularly because LRRK2 kinase inhibitors have entered phase 3 trials. Notably, we have already established an analytic workflow to determine kinase activity and decipher the pathogenicity of single LRRK2 variants in vitro and in vivo. In this letter, we report on a 74-year-old male patient from northern Germany with advanced typical PD (Unified Parkinson’s Disease Rating Scale Part III: 33/108 points, Hoehn and Yahr stage 3–4) without relevant tremor. The age at onset was 67 years. The disease course was slowly progressive, he experienced a good response to dopaminergic therapy, and dementia was absent. Family history was suggestive of autosomal dominant inheritance, with the father and two brothers also diagnosed with PD. However, the father was already deceased, one brother was not available for examination, and the second brother died early after giving blood for genetic investigation. Both brothers carried a new p.F1700L (NM_198578.4: c.5098T>C) variant in LRRK2, initially detected by gene panel analysis in the deceased brother and further investigated by Sanger sequencing. The variant is rated as a variant of uncertain significance according to the criteria of the American College of Medical Genetics (assessed by Franklin: https://franklin.genoox.com/) and is not listed in gnomAD (https://gnomad.broadinstitute.org/). In silico prediction suggested pathogenicity based on a Combined Annotation Dependent Depletion (CADD) score (https://cadd.gs. washington.edu/) of 27.3. The variant is located within the C terminus of the Ras of complex protein B scaffolding domain. We investigated LRRK2 kinase pathway activity in a heterologous transient overexpression system in HEK293 cells as described previously (Fig. 1A). We then analyzed LRRK2 activity in vivo in fresh peripheral blood, simultaneously collected from a p.F1700L carrier and a sex-matched healthy control subject (Fig. 1B). The resulting LRRK2-dependent pRab10 phosphorylation level mirrors LRRK2 kinase activation status. Both experiments demonstrated significant LRRK2 kinase hyperactivation because of the p.F1700L variant. Notably, p.F1700L demonstrated LRRK2 kinase hyperactivation similar to the neighboring p.Y1699C variant, which has the highest degree of kinase activity among all LRRK2 variants investigated thus far in the HEK293 assay (Fig. 1A). Moreover, we found a 7to 8-fold increase in Rab10 phosphorylation levels in vivo (Fig. 1B). Together, we provide robust evidence for the pathogenicity of the newly identified p.F1700L variant in LRRK2, demonstrating that this substitution is among the variants with the highest kinase activity of all LRRK2 variants investigated to date and impacts protein function more profoundly than, for example, the frequent p.G2019S variant. Thus, carriers of a p.F1700L variant should be included in clinical trials targeting LRRK2 kinase activity already initiated or soon commencing. Finally, we confirmed the applicability and usefulness of the applied assays to determine the pathogenicity of LRRK2 variants of uncertain significance. Further studies should focus on an association between the degree of kinase activity and penetrance, disease onset, and disease severity.

Variants in LRRK2 represent the most frequent cause of clinically classical monogenic Parkinson's disease (PD). 1 Altered LRRK2 protein function boosts neuroinflammation, impairs vesicle trafficking, and affects ciliogenesis within the striatum. 2 Because a relevant fraction of the more than 1000 identified LRRK2 variants 3 is not pathogenic, determining pathogenicity for single variants is crucial, particularly because LRRK2 kinase inhibitors have entered phase 3 trials. 4 Notably, we have already established an analytic workflow to determine kinase activity and decipher the pathogenicity of single LRRK2 variants in vitro 5 and in vivo. 6 In this letter, we report on a 74-year-old male patient from northern Germany with advanced typical PD (Unified Parkinson's Disease Rating Scale Part III: 33/108 points, Hoehn and Yahr stage 3-4) without relevant tremor. The age at onset was 67 years. The disease course was slowly progressive, he experienced a good response to dopaminergic therapy, and dementia was absent. Family history was suggestive of autosomal dominant inheritance, with the father and two brothers also diagnosed with PD. However, the father was already deceased, one brother was not available for examination, and the second brother died early after giving blood for genetic investigation. Both brothers carried a new p.F1700L (NM_198578.4: c.5098T>C) variant in LRRK2, initially detected by gene panel analysis in the deceased brother and further investigated by Sanger sequencing. The variant is rated as a variant of uncertain significance according to the criteria of the American College of Medical Genetics (assessed by Franklin: https://franklin.genoox.com/) and is not listed in gnomAD (https://gnomad.broadinstitute.org/). In silico prediction suggested pathogenicity based on a Combined Annotation Dependent Depletion (CADD) score (https://cadd.gs. washington.edu/) of 27.3. The variant is located within the C terminus of the Ras of complex protein B scaffolding domain. 5 We investigated LRRK2 kinase pathway activity in a heterologous transient overexpression system in HEK293 cells as described previously 5 (Fig. 1A). We then analyzed LRRK2 activity in vivo in fresh peripheral blood, simultaneously collected from a p.F1700L carrier and a sex-matched healthy control subject (Fig. 1B). The resulting LRRK2-dependent pRab10 Thr73 phosphorylation level mirrors LRRK2 kinase activation status. 6 Both experiments demonstrated significant LRRK2 kinase hyperactivation because of the p.F1700L variant. Notably, p.F1700L demonstrated LRRK2 kinase hyperactivation similar to the neighboring p.Y1699C variant, which has the highest degree of kinase activity among all LRRK2 variants investigated thus far in the HEK293 assay ( Fig. 1A). Moreover, we found a 7-to 8-fold increase in Rab10 phosphorylation levels in vivo (Fig. 1B).
Together, we provide robust evidence for the pathogenicity of the newly identified p.F1700L variant in LRRK2, demonstrating that this substitution is among the variants with the highest kinase activity of all LRRK2 variants investigated to date and impacts protein function more profoundly than, for example, the frequent p.G2019S variant. 5,6 Thus, carriers of a p.F1700L variant should be included in clinical trials targeting LRRK2 kinase activity already initiated or soon commencing. 4 Finally, we confirmed the applicability and usefulness of the applied assays to determine the pathogenicity of LRRK2 variants of uncertain significance. Further studies should focus on an association between the degree of kinase activity and penetrance, disease onset, and disease severity.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. *Correspondence to: Dr. Esther Sammler, MRC Protein Phosphorylation and Ubiquitylation Unit, School of Life Sciences, Dundee DD1 5EH, UK. E-mail: e.m.sammler@dundee.ac.uk Relevant conflicts of interest/financial disclosures: Nothing to report.
Funding agencies: E.S. was supported by a Chief Scientist Office Scottish Senior Clinical Fellowship, and N.P. was supported by a Carnegie Trust PhD studentship. This work was supported by a grant from the Medical Research Council (MC_UU_00018/1 to D.R.A.). C.K. and K.L. were supported by the German Research Foundation (FOR2488).
Full financial disclosures and author roles may be found in the online version of this article.

Ethics Statement
This study was approved by the ethics committee of the University of Lübeck and performed according to the Declaration of Helsinki.

Data Availability Statement
The data presented in this study can be received from the corresponding author upon reasonable request.  Y1699C variant in an established HEK293 overexpression system, followed by LI-COR Odyssey immunoblotting (i) and quantification of LRRK2 kinase activity relative to LRRK2 wild type (wt). LRRK2-dependent phosphorylation of endogenous Rab10 at threonine 73 (pRab10 Thr73 ) was used as a readout for LRRK2 kinase activity, and the LRRK2-specific small molecule inhibitor MLi-2 at 200 nM for 90 minutes to demonstrate LRRK2 kinase dependency of pRab10 Thr73 as before. 5 The p.F1700L variant demonstrated significant LRRK2 kinase activity with a 4-fold increase of LRRK2-dependent Rab10 phosphorylation compared with LRRK2 wild type similar in effect size to LRRK2 p.Y1699C. Endogenous Rab10 levels did not differ (ii). Each data point represents a biological replicate experiment. Data were analyzed using one-way ANOVA with multiple comparisons test. Statistical significance was determined from three replicate values for each variant and represented with P values (****P < 0.0001). (B) In vivo analysis of kinase activity in patient-derived neutrophils and monocytes. Fresh blood was taken from the Parkinson's disease patient carrying the LRRK2 p.F1700L variant and a sex-matched healthy control subject. Immunomagnetic negative isolation of peripheral blood neutrophils and monocytes was performed, and each sample was then split into two batches for treatment with and without the specific LRRK2 kinase inhibitor MLi-2 before cell lysis as before. 6,7 As with the HEK293 overexpression assay, LRRK2-dependent Rab10 phosphorylation (pRab10 Thr73 ) as a readout for LRRK2 kinase activity was significantly increased in both peripheral blood neutrophils and monocytes derived from the LRRK2  The very recent discovery of intronic GAA monoallelic repeat expansions in the FGF14 (fibroblast growth factor 14) gene, 1,2 as well as the previously described biallelic repeat expansions in the RFC1 (replication factor C subunit 1) gene, 3,4 causing cerebellar ataxia, neuropathy, vestibular areflexia syndrome, have greatly enhanced the diagnostic yield in ataxia, which was previously limited to <50% even with exome sequencing. 5 We describe a Chilean nonconsanguineous family with ataxia from a rural area (Fresia). Seven affected women and five affected men were identified, including 11/20 affected among the offspring in the three younger generations, suggesting high penetrance of an autosomal dominantly inherited cause. Age at onset (AAO) ranged from 17 to 50 years, with seemingly genetic anticipation (Fig. 1A). The index patient is a 29-year-old man who presented with cramps and leg pain at 21 years of age. He progressively developed vertigo, clumsiness, and gait ataxia. He had a history of rapid eye movement sleep behavior disorder (RBD), constipation, and depressive symptoms. His physical examination showed dysarthria, gait and limb ataxia, and absence of deep tendon reflexes. Oculomotor abnormalities or pyramidal features were absent. A sensorimotor chronic moderateto-severe polyneuropathy was diagnosed by nerve conduction studies, and superior vermis brain atrophy was seen on magnetic resonance imaging. All affected family members shared this phenotype (Video S1), with additional features in his grandmother, ie, orolingual dyskinesias, choreoathetosis, and dementia. Genetic testing for ATXN1/2/3/7, CACNA1A, TBP, ATN1, FXN, HTT, and exome sequencing was negative. FGF14 testing demonstrated an expanded allele (325 repeats) in three of the four affected and in one unaffected (aged 62 years) family member (Fig. 1A,B). Incomplete segregation could not be explained by a sample mix-up as relationship and sex were tested by nine polymorphic microsatellite markers, including two markers each on the X and Y chromosomes. Notably, repeat-primed PCR suggested repeat interruption in all four tested FGF14 expansion carriers in the Chilean family ( Figure 1C). Due to vertigo in one affected, we also tested for RFC1 expansions in all available family members revealing a biallelic repeat expansion in one unaffected family member, aged 18 years, ie, well younger than the average AAO of RFC1 patients (Fig. 1A,C). This is the first Latin American family in whom an expanded FGF14 repeat was found. Although the definitive range for normal vs. intermediate/reduced penetrance vs. pathogenic/full penetrance alleles has not yet been determined, 250 and 300 repeats have been suggested as the respective boundaries. 2 Importantly, however, the GAA repeat in our family is interrupted, and only pure expanded alleles have currently been identified in affected individuals. 1 The phenotype in this family with the interrupted FGF14 repeat expansion of uncertain clinical relevance differs from the two series published to date, as ocular abnormalities were absent and additional features were consistently observed, such as polyneuropathy, RBD, constipation, and depression. Notably, there is co-occurrence of an only heterozygous RFC1 repeat expansion in at least one affected and one unaffected family member with an additional heterozygous expansion in FGF14 (Fig. 1A). However, at this stage, it is unclear whether there is an additive effect of expanded repeats in both genes. Interestingly, patient L-14997 appears to be the most severely affected in this family.

References
Since the role of the interrupted FGF14 repeat expansion could not be determined unambiguously in this family, another, as yet unidentified genetic variant/repeat expansion could contribute to the disease. Furthermore, it is important to not only determine the length of the FGF14 repeat but also its architecture in unexplained spinocerebellar ataxia.