Parkinson’s disease (PD) is a neurodegenerative condition that affects 1% of the population over 65 years of age. The two pathological hallmarks of PD are a loss of dopaminergic neurons in the substantia nigra (SN) and the presence of cytoplasmic eosinophilic inclusions termed Lewy bodies (LBs), whose main component is phosphorylated α-synuclein.1 This degeneration of SN neurons leads to a dopamine deficiency responsible for the major motor symptoms. Nevertheless, it has become increasingly evident that PD is a multicentric neurodegenerative process that also affects neuronal structures outside the SN.2 In this context, various reports performed on surgical or autopsy specimens have shown that the enteric nervous system (ENS) is affected during PD.3 4 However, it is still a matter of debate whether these alterations occur early in the course of the disease. This is mainly due to a lack of accessibility to the ENS in the living patients. Therefore, demonstrating (1) the ability to study the ENS using routine colonic biopsies and (2) the presence of lesions characteristics of PD could be relevant for an early diagnosis of the disease and to better understand its pathophysiology.

We therefore performed routine colonic biopsies in five PD patients complaining of functional constipation (63 (SD 7) years, three men; all had disease duration >5 years). Five healthy age-matched patients (61 (SD 6.5) years, one man) requiring a total colonoscopy for colorectal cancer screening were included as controls. They had no known neurological disease. None suffered from functional digestive symptoms. In order to avoid any specific role for chronic constipation, we included three additional non-age-matched patients (52 (SD 5) years, no men) who underwent total colonoscopy for the assessment of a chronic intractable constipation as additional controls. Written consent was obtained according to the principles of the Declaration of Helsinki.

Four biopsies were taken from the ascending colon during colonoscopy. Biopsies were performed using standard biopsy forceps without needles (FB210K; Olympus, Tokyo, Japan). Samples were immediately immersed in 4°C saline solution and microdissected in order to separate the submucosa (containing the internal submucosal plexus) from the mucosa. The submucosa were then fixed in 4% paraformaldehyde. Imunohistochemical studies were then performed on these tissues using a combination of antibodies against rabbit anti-tyrosine hydroxylase (TH) (1:500, Pel-Freez, Rogers, Arkansas, USA), rabbit anti-dopamine-β-hydroxylase (DBH) (1:250, Millipore, Saint-Quentin-en-Yvelines, France), mouse anti-Hu C/D (1:200, Invitrogen, Cergy-Pontoise, Fance), rabbit anti-phosphorylated α-synuclein (1:5000, WAKO, Osaka, Japan) or rabbit anti-neurofilament 200 kDa (1:250; Millipore) as previously described.5

In control patients, individual biopsies contained 11.2 (SD 7.9) ganglia and each ganglia contained 5.6 (SD 1.9) Hu-immunoreactive (IR) neurons. In PD patients, the number of ganglia per biopsies was similar to controls (13.6 (SD 5.3); p = 0.22). In addition, the number of Hu-IR neurons per ganglion in PD was unchanged as compared to controls (7.0 (SD 1.6); p = 0.25) (fig 1A,C). Constipated controls did not differ from PD patients in the number of ganglia per biopsy (11.3 (SD 1.5); p = 0.57) or in the number of neurons per ganglion (5.5 (SD 0.6); p = 0.19) (fig 1C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Submucosal neuron counts and dopaminergic phenotype are unchanged in patients with Parkinson’s disease (PD). Hu-immunoreactive (IR) submucosal neurons were identified in the colon of controls (n = 5) (A), PD patients (n = 5) (B) and constipated patients (n = 3). There was no change in the number of Hu-IR submucosal neurons per ganglion in the three conditions (C). Double labelling with antibodies against Hu (A,B) and tyrosine hydroxylase (TH) (D,E) showed that occasional submucosal neurons were TH-IR (arrow heads). No significant decrease in the proportion of TH-IR submucosal neurons occurred in PD and in constipated patients (F). Each circle, square and triangle represents one control, PD or constipated patient, respectively. Horizontal bars represent the mean. Scale bar: 20 μm. CTL, control; CP, constipated patient.

In healthy controls, 11.6 (SD 5.0)% of Hu-IR neurons were TH-IR. In PD patients, the proportion of TH-IR neurons was unchanged as compared to controls (12.3 (SD 3.3)%; p = 0.80) (fig 1D–F). In constipated patients, the proportion of TH-IR neurons was similar to the one of PD patients (8. (SD 2.7)%; p = 0.12) (fig 1F). In all groups no neuronal body was DBH positive, suggesting that all TH-IR neurons in the submucosal plexus were dopaminergic. These results are consistent with a previous report by Singaram et al6 showing the absence of loss of TH-IR neurons in the submucosal and myenteric plexuses of PD patients, suggesting that it is not a marker of choice for detecting PD lesions in the ENS.

However, immunohistochemical staining with an antibody against phosphorylated α-synuclein, revealed that 4 out of 5 PD patients had phospho-α-synuclein-IR neurites (identified with neurofilament (NF) in the submucosa (fig 2A,F). These phospho-α-synuclein-IR neurites were absent in both control and constipated patients. In some cases, large aggregates were observed in dystrophic NF-IR neurites (fig 2E), a pattern reminiscent of Lewy neurites.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Phospho-α-synuclein-positive submucosal neurites differentiate Parkinson’s disease patients from controls. Double labelling with antibodies against neurofilament (NF) (A,B) and phosphorylated α-synuclein (C,D) revealed that some NF-immunoreactive (IR) neuritic structures were also phospho-α-synuclein-IR (merged image in E,F) in the majority of Parkinson’s disease patients, but in none of the controls. Occasionally the inclusion-bearing neurites displayed dystrophic alterations (A,C,E). Scale bar: 30 μm.

Taken together, our pilot study showed that routine colonic biopsies can be used to study the submucosal plexus of the ENS. In addition, we identified for the first time in the gut of living PD patients lesions similar to the ones observed in the brain. This technique could be a reliable tool to detect early lesions in the gut during the course of PD in order to better understand the pathogenesis of the disease and/or to identify novel biomarkers.

Acknowledgements

The authors wish to thank M Roy and F Vavasseur for their help in the assessment of patients and controls.