Differential Gene Expression in Patients With Prostate Cancer and in Patients With Parkinson Disease: an Example of Inverse Comorbidity

Prostate cancer (PCa) is one of the leading causes of death in Western countries. Environmental and genetic factors play a pivotal role in PCa etiology. Timely identication of the genetic causes is useful for an early diagnosis. Parkinson’s disease (PD) is the most frequent neurodegenerative movement disorder; it is associated with the presence of Lewy bodies (LBs) and genetic factors are involved in its pathogenesis. Several studies have indicated that the expression of target genes in patients with PD is inversely related to cancer development; this phenomenon has been named “inverse comorbidity”. The present study was undertaken to evaluate whether a genetic dysregulation occurs in opposite directions in patients with PD or PCa. In the present study, next-generation sequencing (NGS) transcriptome analysis was used to assess whether a genetic dysregulation in opposite directions occurs in patients with PD or PCa. The genes SLC30A1, ADO, SRGAP2C, and TBC1D12 resulted up-regulated in patients with PD compared to healthy donors as controls and down-regulated in patients with PCa compared with the same control group. These results support the hypothesis of the presence of inverse comorbidity between PD and PCa.


Introduction
In the United States and Europe, prostate cancer (PCa) is the leading cause of cancer-related death among men [1,2]. The prognosis is good for patients with localized or regional disease; nevertheless, the survival average in patients with metastasis is only 12-15 months [3][4][5]. The exact mechanisms involved in PCa initiation and progression have not been clearly understood yet, although genetic factors are believed to play a role [6].
Parkinson's disease (PD) is the most frequent neurodegenerative movement disorder, which is histologically featured by neuronal loss (speci cally occurring in the substantia nigra) and the by the presence of Lewy bodies (LBs) [7,8]. On a clinical point of view, PD is characterized by the onset of speci c motor symptoms (e.g. akinesis, rigidity, postural instability and resting tremor), which are due to the progressive dopaminergic nigrostriatal system degeneration. PD is a very complex disorder. It indeed involves multiple biochemical de cits that are extended to the peripheral nervous system. For this reason, PD had been recently rede ned as a multisystem disorder with a genetic background [9].
Epidemiological data have shown that patients with central nervous system (CNS) disorders, including PD, have a lower susceptibility to develop cancer [10][11][12][13]. This has been de ned "inverse comorbidity". More speci cally, inverse comorbidity is a lower-than-expected probability that a disease will occur in people who have other diseases. Emerging evidence indicates lower tumor comorbidity in patients with some CNS disorders including PD [11]. Such inverse comorbidity is in uenced by environmental, pharmacological, and alimentary factors. However, genetic factors can additionally contribute to the inverse comorbidity between complex diseases [10][11][12][13]. Accordingly, several studies have indicated that the expression of target genes in patients with PD is inversely related to tumor development [14][15][16]; nevertheless, no data are available on the possible presence of inverse comorbidity between PCa and PD. Therefore, the present study was undertaken to evaluate whether a genetic dysregulation occurs in opposite directions in patients with PD or PCa. To accomplish this, next-generation sequencing (NGS) transcriptome analysis.
These data were further con rmed by RNA sequencing data from enriched prostate CTCs from clinical blood specimens. In particular, raw data generated by Wong and colleagues [17] were reanalyzed as described above and compared to the samples investigated in-house. By analyzing the expression level of the selected four deregulated mRNAs in our patients with PCa (|FC| ≥1.5 and adjusted p ≤ 0.05), we observed a direct correlation in the two PCa datasets and an inverse correlation with the PD dataset ( Fig. 1). Discussion SLC30A1, also known as ZnT1, is widely expressed in human tissues and primarily localized in the plasma membrane. It contributes to cytoplasmic zinc (Zn) balance by exporting Zn to the extracellular space [18]. Singh and colleagues [19] reported a signi cant upregulation of SLC30A1 mRNA in PCa tissue, compared to benign prostatic tissue. Increased expression of SLC30A1 was also found in patients with Alzheimer's disease [20].
ADO has a role in cysteine metabolism and is localized in the mitochondria. This metabolic pathway is responsible for the synthesis of glutathione with antioxidant properties and is involved in the pathogenesis of various disorders [21]. Sarkar and colleagues [22] highlighted the role of an altered expression and functioning of the ADO gene in the pathogenesis of neurodegenerative disorders and cancer [22]. SRGAP2C is involved in the development of the cerebral cortex [23,24]. The gene encodes for a highly conserved protein that is expressed early in development. More in detail, this protein acts as a regulator of neuronal migration and differentiation by inducing lopodia formation, branching of neurons, and neurite outgrowth [23,24].
TBC1D12 is a Tre2-Bub2-Cdc16 (TBC) domain family of proteins that colocalize with transferrin receptor (TfR), a well-known recycling endosome marker. The possible role of TBC1D12 has been highlighted in several cellular events, including transferrin (Tf) recycling, neurite outgrowth, and endosomal tra cking [25].
Noteworthy, SLC30A1, ADO, SRGAP2C, and TBC1D12 genes play crucial roles in the molecular mechanisms involved in oncological and neurodegenerative processes, as shown in Fig. 2, which summarizes their Biological Process networks. Catalá-López and colleagues [26] showed that the presence of CNS disorders is associated with a reduced co-occurrence of cancer [26]. The concept that a speci c disease is able to offer protection against cancer can be traduced in a novel model of inverse cancer morbility, in patients with CNS diseases. Particularly, epidemiological data support the occurrence of a lower-than-expected incidence of cancer in patients suffering from PD, Down's syndrome, diabetes mellitus, Alzheimer's disease, and multiple sclerosis [27]. This knowledge represents an opportunity to deepen CNS pathogenesis. Further studies are needed to understand the reason why speci c CNS disorders have a low risk of cancer development. This data may also help in the development of new and more effective therapeutic strategies in the next future.

Conclusion
In conclusion, the data of the present study support the existence of inverse comorbidity between PCa and PD and may represent the starting point to design more in-depth and well-sized studies to con rm these ndings.

Patient selection
A total of 27 men were included in this study; 7 controls, 13 patients with PD, and 7 with PCa. For the NGS experiments, we recruited 23 participants, including 7 controls with an age range of 54-87 years, 13 patients with PD with an age range of 50-88 years, and 3 patients with PCa aged 54, 71, and 72 years, respectively. Moreover, data deposited by Wong and colleagues [17] on a case history of 4 patients with PCa were analyzed.
Patients with PD and controls were recruited at the Oasi Research Institute-IRCCS, Troina (Italy), whereas patients with PCa were recruited at the Urology Unit, Cannizzaro Hospital, Catania (Italy). Family history was collected through a detailed interview of the wife or the rst-degree relatives. The present and past clinical history of each patient was gathered and all available medical records (e.g., drug prescriptions, medical records, certi cates) were carefully evaluated. All the diagnoses of PD were made by an experienced neurologist, according to the current PD diagnostic criteria [28]. PCa diagnoses were performed after radical prostatectomy and histological analysis. The diagnosis and grading of PCa were made according to the 2005 International Society of Urological Pathology (ISUP) Consensus Conference [29]. All patients and controls or their relatives gave their informed consent for study participation. The study was carried out in accordance with the Declaration of Helsinki of 1964 and its later amendments, and the Ethics Committee of the Oasi Research Institute-IRCCS, Troina (Italy) approved the protocol on June 03, 2017 (2017/05/31/CE-IRCCS-OASI/9).

RNA extraction
PBMCs separation was performed using Ficoll-Paque (Ficoll Plaque PLUS -GE Healthcare Life Sciences, Piscataway, USA) and the RNA was extracted using TRIzol reagent (TRIzol Reagent, Invitrogen Life Technologies, Carlsbad, CA, USA), according to the manufacturer's instructions [30]. The RNA obtained was stored at -80 o C.
RNA sequencing and data analysis RNA sequencing was performed by Genomix4Life Srl (Baronissi, Italy). RNA concentration in each sample was assayed with an ND-2000c spectrophotometer (NanoDrop) and its quality was evaluated with the TapeStation 4200 (Agilent Technologies). Indexed libraries were prepared from 1 µg/ea puri ed RNA with TruSeq Stranded Total RNA (Illumina) Library Prep Kit according to the manufacturer's instructions.
Libraries were quanti ed using TapeStation 4200 (Agilent Technologies) and pooled such that each index-tagged sample was present in equimolar amounts, with a nal concentration of the pooled samples of 2 nM. The pooled samples were subject to cluster generation and sequencing using an Illumina NextSeq 500 System (Illumina) in a 2x75 paired-end format. The raw sequence les generated (fastq les) underwent quality control analysis using FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Subsequently, the quality-checked reads were trimmed with cutadapt v.1.10 and then aligned to the human genome (hg38 assembly) using STAR v.2.5.2 [31], with standard parameters. The read-count for the genes of interest was computed using feature Counts and normalized using negative binomial generalized linear models, considering all genes expressed in the samples, by Bioconductor DESeq2 package [31,32]. Transcripts showing fold-change ≤−1.35 or ≥1.35 (|FC| ≥1.35), with adjusted p-values ≤ 0.05 (padj) were considered as differentially expressed. Gene Ontology (GO) annotation analysis was performed using GO net web-application [33]. The sequencing data are available upon request, as they are partial data of a larger research project not yet completed.
RNA sequencing data from enriched prostate circulating tumor cells from clinical blood specimens Raw data of enriched prostate circulating tumor cells (CTCs) from 12 (3 blood samples for each patient) clinical blood specimens were downloaded by NCBI GEO (accession number GSE104209) [17]. RNA sequencing data were analyzed as described above and compared to the in-house control.

Declarations
Con ict of interests