LRRK2 was identified using a classical genetic linkage approach. In 2002 a region of chromosome 12p12 was reported to segregate ‘identical-by-descent’ with clinical parkinsonism in the Japanese Sagamihara family¹,. We found similar results for chromosome 12 in two Caucasian kindreds, Family A (German-Canadian) and Family D (Western Nebraska)², and soon after in Norwegian pedigrees³. In these families Parkinson’s disease (PD) is inherited in a dominant fashion i.e. approximately half of every generation eventually becomes affected. Although fourteen percent of patients have an affected first-degree relative (a parent, sibling or child) with PD⁴, it is uncommon to find more than two affected subjects in a family. Genetic research requires informed consent to ask questions about family history, to perform clinical exams, take blood for DNA extraction, and potentially ask permission for studies of brain pathology post-mortem. For the advances made we are indebted to those who have participated.

In 2004, we described the precise LRRK2 mutations that resulted in PD, namely LRRK2 R1441C (Families D and 469), Y1699C (Family A) and I2020T (Family 32)⁵. The work was done with Dr. Zbigniew Wszolek at Mayo Clinic, and as part of an international team. Concomitantly, a group at the US National Institutes of Health reported LRRK2 R1396G in four Basque Spanish families and Y1654C in an English kindred⁶. Although incorrectly labelled, LRRK2 R1396G is actually R1441G, and Y1654C is Y1699C, these families provided more evidence for pathogenicity. LRRK2 I2020T was also found to cause disease in the Sagamihara kindred⁷.

Notably, six affected subjects of Family A and D, and six members of the Japanese Sagamihara family donated their brains to research^8,9. All were found to have mid-brain neurodegeneration with profound neuronal loss in the substantia nigra, typical of sporadic PD. However, only about a quarter of these patients had mid-brain Lewy body disease (‘alpha-synuclein immunopositive’ aggregates). Many patients had neurofibrillary tangles (‘tau immunopositive’ aggregates) or alternatively ‘ubiquitin immunopositive’ aggregates. It was amazing the brain pathology was so different (termed ‘pleomorphic’) among individuals, despite clinically similar presentations, and even within each family with the same disease-causing LRRK2 mutation⁸. Until recently, a diagnostic requirement for ‘definite’ PD had required the presence of mid-brain Lewy body disease so this observation was really contentious. Nevertheless, as many as ~20% of autopsy-confirmed cases with “probable” PD may not have mid-brain Lewy body pathology, (although clinically these patient may have been quite typical)¹⁰. To this date, and with >37 brains examined, only half of all patients with LRRK2 mutations and clinical PD develop mid-brain Lewy body pathology¹¹.

With Dr. Jan Aasly’s help, then Head of the Department of Neurology at St. Olav’s, Trondheim, we discovered LRRK2 G2019S in Norwegian families³. The mutation also segregated with PD in a dominant fashion¹², and through many generations of affected subjects dating as far back as the 15th century¹⁸. While we also reported affected families in Poland, Ireland, Spain and the US³, LRRK2 G2019S proved to be most frequent in ‘seemingly sporadic’ PD south and east of the Mediterranean. The frequency was highest in Ashkenazi Jews in Israel (and New York) and Arab-Berber populations in North Africa (Algeria and Tunisia), where it is found in 13% and 30% of their PD, respectively^13,14. I write ‘seemingly sporadic’ as in all these families and populations – whether from Trondheim in Norway, Tel Aviv in Israel or Tunis in Tunisia – LRRK2 G2019S originates from the same ancestral founder, in effect, these patients are all distant cousins^3,15. Arguably, the LRRK2 G2019S mutation has been dated to ~1,000 BC, a time when Mediterranean sea trade was most active between Phoenician ports in North Africa (Tunisia, Algeria and Morocco), Europe (Spain and Italy) and the Levant (Lebanon, Syria, Israel)¹⁶. In the present day populations of Israel and Tunisia, 1.9% of Ashkenazi and Arab-Berber subjects have LRRK2 G2019S due to a combination of genetic selection and drift. Ancient Norse trade, that included long-term settlements in Carthage ~1000 AD, may explain LRRK2 G2019S families in Norway. In contrast, the absence of LRRK2 G2019S in Central Europe, in Austrian and German patients, may reflect migration during World War II and the holocaust.

Although its origin is lost in history the LRRK2 G2019S diaspora is arguably the greatest single genetic cause of PD. Why most individuals appear to have “sporadic” rather than familial disease remains unclear. From meta-analysis the lifetime probability of LRRK2 G2019S heterozygotes becoming affected (termed penetrance) is ~30% (28% at 59, 51% at 69 and 74% at 78 years)¹⁷. However, there are families in which LRRK2 G2019S PD is clearly inherited, as that is how it was originally identified^12,18,19. In 2005, I started working with Dr. Faycel Hentati, the Director of the National Institute of Neurology in Tunis, North Africa. There is a pandemic of PD in Arab-Berbers, the predominant population in North Africa, and to visit the Clinic and families in their homes has been a humbling experience, given their plight. Their generosity to help describe the clinical syndrome and its penetrance, and to help find genetic modifiers that influence when subjects with LRRK2 G2019S will become affected, has been remarkable^19–21. Many other LRRK2 genetic variants contribute to risk, albeit more modestly. For example, LRRK2 G2385R, specific to Asian populations from Korea to Taiwan, is found in 6-8% of patients and doubles the risk of PD^22,23. There are also LRRK2 genetic variants that are ‘protective’ i.e. inversely associated with PD²⁴.

LRRK2 G2019S directly affects the ‘activation segment’ of LRRK2’s kinase (Figure). In effect this mutation ‘always keeps the door ajar’, allowing LRRK2 to phosphorylate other proteins (substrates), even when it should be inactive³. This notion has been supported with each substrate identified - first auto-phosphorylation of LRRK2 itself, and most recently in phosphorylation of many members of the Ras GTPase superfamily²⁵. Thus a competitive but specific inhibitor of LRRK2 kinase activation, even with modest efficacy, might prevent disease in those susceptible.

To develop a safe and affordable drug, and without off-target side-effects for a lifetime of use, is our challenge. Several classes of LRRK2 inhibitors have now been identified, but which will meet this criteria is unclear²⁶. Recent studies have shown LRRK2 kinase inhibition more often destabilizes the protein, lowering its expression, and may result in undesirable side-effects in lung and kidney. More research funding is needed in three areas: i) to identify and measure “PD-relevant” outcomes of LRRK2 kinase inhibition; ii) to develop preclinical models in which such effects can be easily and reliably assessed, and; iii) to prospectively identify and follow the natural evolution of disease in those patients and families with LRRK2 genetic variability, most likely to benefit from such a therapeutic approach.

“PD-relevant” outcomes of LRRK2 dysfunction may be contested but in neurons many agree mutant effects are subtle, chronic and initially synaptic²⁷. Animal models that faithfully recapitulate the etiology and physiology of germline LRRK2 mutations in humans have been successfully developed, with measureable differences^28,29. While drug screening requires more simple, innovative and informative assays of LRRK2 function, several avenues hold promise and not least in human “inducible-pluripotent stem cell’ derived dopaminergic neurons³⁰. LRRK2 G2019S is also appreciated to cause PD in >50,000 Arab-Berber patients, and Tunisia, among other North African nations, definitely has the infrastructure, expertise and desire to develop neuroprotection, anticipating such drugs might be made affordable.

Now we know genetic causes of PD, the associated biologic mechanisms, and the patients most likely to benefit, “neuroprotection” to halt or prevent symptoms can be realized. LRRK2-directed therapeutics are likely to have broad efficacy as neuroprotective agents and beyond patients with specific mutations²⁴. A similar approach has been taken for patients with PD and glucocerebrosidase (GBA) mutations by Sanofi-Genzyme who recently launched as Phase II clinical trial of their ceramide substrate inhibitor (GZ/SAR402671)³¹. Several clinical trials have also been initiated to clear or prevent Lewy body formation by lowering alpha-synuclein expression³². While neuropathologic examination of LRRK2 patients suggests Lewy bodies may be neither cause nor consequence of the disease process, clearing those protein aggregates is likely to be of benefit to many patients who develop that pathology.