TDP-43 and HERV-K Envelope-Specific Immunogenic Epitopes Are Recognized in ALS Patients

The human endogenous retrovirus-K (HERV-K) and TAR DNA-binding protein 43 (TDP-43) have been associated with the pathophysiology of amyotrophic lateral sclerosis (ALS). Given these findings, we investigated the humoral response against HERV-K envelope surface (env-su) glycoprotein antigens and TDP-43 in the plasma of ALS patients and healthy controls (HCs). The measured levels of Abs against the different epitopes’ fragments were significantly elevated in ALS patients, both in long-survivor (LS) and newly diagnosed (ND) patients, compared to HCs. We observed a positive correlation between HERV-K and TDP-43 antibodies (Abs) levels, which seemed to strengthen with disease progression, that was not found in HCs. The TDP-43 and HERV-K epitopes identified in this study are highly immunogenic and recognized by the humoral response of ALS patients. Increased circulating levels of Abs directed against specific HERV-K- and TDP-43-derived epitopes could serve as possible biomarkers.


Introduction
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease of unknown etiology described for the first time in 1874 by Jean-Martin Charcot. It is characterized by the progressive degeneration of both the upper and lower motor neurons, which display cytoplasmatic inclusions [1,2]. Population-based studies have highlighted that women are less affected by sporadic ALS (sALS) than men, while the same incidence between men and women in familial ALS (fALS) is observed. The incidence decreases rapidly after 80 years of age, whereas the 58-63 and 47-52 age brackets represent the peak ages of onset for sALS and fALS, respectively [3]. The ALS condition is primarily characterized by the involvement of the pyramidal tract with the progressive reduction in upper motor neuron activity, which originates from the motor cortex, and lower motor neurons, which connect the spinal cord and brainstem to skeletal muscles. Common features of ALS are progressive muscle weakness and atrophy, fasciculations, dysarthria, and dysphagia. Remarkably, a significant proportion of cases presents frontotemporal lobe degeneration (FTLD) with cognitive damage. Respiratory complications are the final stage, after which death occurs within two to five years of diagnosis [4]. The main causes of ALS are still unknown, but significant advances have been made in understanding the environmental and genetic components involved in the pathology [5]. There are several hypotheses behind the development of the disease, involving both genetic predisposition, with more than 24 implicated genes, such as Superoxide dismutase 1 (SOD1) and TAR DNA-binding protein 43 (TDP-43) [6], and environmental risk factors, for instance, smoking [7], physical activity [8], and chemical exposure [9,10].
Emerging evidence has indicated that human endogenous retrovirus K (HERV-K) may play a role in the disease's etiopathogenesis. Human endogenous retroviruses (HERVs), considered until recently as junk DNA, constitute nearly 8% of the human genome. Proviral genome integrations into the DNA are remnants of infections that occurred over several million years ago. Nevertheless, numerous nonsense mutations have contributed to their defection [11].
HERV-K, in particular the HML-2 subgroup, is present in hundreds of copies in the human genome and is one of the most recently acquired and the most transcriptionally active among HERVs. The involvement of HERV-K in the pathophysiology of ALS has been documented by numerous groups. Douville et al., (2011) observed an altered expression of HERV-K pol transcripts in post-mortem brain samples of patients affected by ALS compared to other pathological conditions such as Parkinson's disease, chronic systemic illnesses, and healthy subjects [12]. Transgenic animals, in whose neurons the env gene was expressed, developed progressive motor dysfunction [13].
TDP-43 is a multifunctional protein associated with several biological functions [17], including mRNA transcription, splicing and stability [18][19][20], stress granule formation [21], and Protein-Protein Interactions [22]. TDP-43 is indispensable for the development of the central nervous system (CNS) from the earliest stages of embryonic life to adulthood [23,24]. Under physiological conditions, TDP-43 is mainly a nuclear protein, but it also shuttles to the cytoplasm for further functions. Neurons transfected with TDP-43 mutants altering nuclear trafficking exhibit cytoplasmic aggregates with phosphorylated and ubiquitinated TDP-43, which is characteristic of ALS pathology [25]. The TARDBP gene is located on chromosome 1, and the TDP-43 protein consists of 414 amino acids. The C-terminal region, which plays a crucial role in the disease, encompasses a prion-like glutamine/asparaginerich (Q/N) domain and a glycine-rich region [26]. TDP-43 represents a target for numerous post-translational modifications that may change its structure, functions, localization, and its aggregative predisposition [27,28]. The most documented post-translational modifications in ALS patients are phosphorylation in serine-409 and serine-410 [29,30], but other phosphorylated sites of pathological TDP-43 have also been described, such as serine-379, 403, and 404 [29,31,32]. Kametani et al., showed that all serine and threonine residues in the C-terminal domain can become phosphorylated [33], and this may increase the tendency of TDP-43 to be hydrolyzed into C-terminal fragments or to aggregate [34].
To date, the ALS diagnostic process requires approximately one year, since only the progression of symptoms and the presence of signs of both upper and lower motor neuron involvement can confirm the ALS diagnosis. The diagnosis is based on clinical examination, electrophysiological findings, medical history, and exclusion of confounding disorders [35]. Unfortunately, no effective therapies able to cure the disease are available; Riluzole is the only therapy that can prolong ALS survival (by approximately three months). Different treatments are available to help control ALS symptoms, prevent unnecessary complications, and make it easier for patients to live with the disease. This study aims to more deeply understand the possible relationship between TDP-43 and HERV-K through the investigation of the humoral response to different epitopes of the TDP-43 C-termina region and the HERV-K envelope. The detection of circulating autoantibodies is crucial in the diagnosis and monitoring of several diseases, and it helps to understand the etiopathogenesis. Furthermore, this approach may be useful to hypothesize new biomarkers for ALS which would allow a faster diagnosis.

Samples Collection
We evaluated an ALS group of 45 patients (17 females and 28 males; mean age ± SD: 64.67 ± 9.11 years), and an age-and gender-matched healthy control group (HCs) collected from the Blood Transfusion Centre of Sassari (17 females and 28 males; mean age ± SD: 63.87 ± 4.82 years). Data are summarized in Table 1. The ALS population, sampled from January 2016 to December 2019, comprised newly diagnosed (ND) ALS patients (7 females and 20 males; mean age ± SD: 64 ± 7.8 years), hospitalized at the Neurology Unit Clinic of the University Hospital of Sassari and longsurvivor (LS) ALS patients (10 females and 8 males; mean age ± SD: 65.8 ± 10.4 years) reported by primary doctors and doctors in the Sassari local district.
Peripheral venous blood samples were collected in a K2-EDTA tube. Whole blood was gently layered over an equal volume of Ficoll (Sigma-Aldrich, St. Louis, MO, USA) in a 15 mL tube and centrifuged for 20 min at 1800 RPM without brake. The plasma contained in the uppermost layer was collected by pipetting and tested for the presence of Abs against TDP-43-and HERV-K-env-derived epitopes. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of ALS 1 Sassari (2149/CE).

Statistical Analysis
All data were analyzed using GraphPad Prism 8.2.0 software (GraphPad Software, San Diego, CA, USA). A Mann-Whitney U test was used to analyze non-parametric data and compare differences among two groups. Kruskal-Wallis and Dunn's post hoc tests were carried out to compare differences among three groups. Statistically significant difference was set with a p < 0.05. Receiver-operating characteristic (ROC) was used to choose the cut-off value to assess the sample positivity, which was consequently tested through Fisher's exact test. A Spearman correlation test was performed among levels of antibodies to HERV-K env-and TDP-43-derived peptides.

Results
We enrolled 45 ALS patients and 45 matched HCs in our study. Details of the ALS patients' clinical information are shown in Table 3.

Discussion
In this study, we investigated the humoral response directed against specific epitopes of the HERV-K envelope and TDP-43 proteins in human ALS plasma samples. Autoantibodies were found to be significantly higher in ALS patients compared with healthy control samples. These results confirm our previous data and highlight the reactivation of the human endogenous retrovirus in ALS pathology [14].
Likewise, Conti et al., documented increased levels of both anti-TDP-43 Abs and TDP-43 protein in ALS-serum patients compared with healthy controls, motor neuron disease mimics, and Alzheimer's and frontotemporal lobar degeneration patients, but autoantibodies and protein serum levels failed to correlate [36]. However, recent work has documented a decreased level of anti-TDP-43 Abs in ALS plasma samples [37].
Several studies [38,39] suggest an association between TDP-43 plasma levels and the cerebral accumulation of the protein; however, there is no strong scientific evidence that corroborates this intriguing hypothesis, and further studies are required.
The hypothesis regarding the possible physiological role of autoantibodies in the prevention of excessive accumulation of neuronal inclusions is in agreement with the data reported [40].
Otherwise, the literature has quite clearly proven that the cytoplasmic aggregation of phosphorylated, ubiquitinated, and truncated TDP-43 represents a pathologic hallmark across the clinical spectrum of ALS [6]. The C-terminal region of TDP-43, in particular, is a complex domain, which is essential in the mediation of liquid-liquid phase transitions in the protein, which is central for the biogenesis of various membrane-less organelles such as stress granules, even if makes them prone to misfolding and aggregation [41]. Cterminal fragments (CTFs) are found in cytoplasmic aggregates along with the full-length TDP-43 [6]. Regarding this, our results highlight the increased humoral response directed against selected C-terminal epitopes, including phosphorylated ones, of TDP-43 in ALS patients when compared with HCs ( Figure 1). Associating the plasma antibody response with the cerebral accumulation of TDP-43 is particularly intriguing to have an overview of the cerebral accumulation of protein; nevertheless, further investigations are necessary to gain more insight into this aspect.
In order to investigate the role of autoantibodies as a novel biomarker in ALS, we searched for the presence of different autoantibodies in both ALS-ND and ALS-LS patients ( Figure 2). The humoral response against the four epitopes selected did not exhibit any statistical difference, indicating that the Abs searched may be present in ALS individuals from the first months after the diagnosis. It would be very interesting to investigate the humoral response in a prospective study looking for autoantibodies at the first symptoms before a definite ALS diagnosis is made.
Different studies have documented the implication of HERV-K in ALS. An altered expression of HERV-K pol transcripts in post-mortem brain samples of ALS patients has been confirmed compared to patients with other nervous system diseases and healthy subjects [12]. HERV-K env transgenic cells and mice showed ALS-like pathological and neurological manifestations. Significantly increased levels of anti-HERV-K-env-su Abs were observed by Arru et al., in the serum and cerebrospinal fluid of ALS patients, hinting at the development of a humoral immune response to HERV-K in patients [14].
Interestingly, no correlations have arisen in HCs. This endorses the results obtained in ALS patients and the specific relationship between HERV-K and TDP-43 in pathological conditions ( Figure 3). Additional analysis, associated with the duration of the disease, has highlighted a slightly stronger correlation in LS-ALS compared to ND-ALS regarding the Ab levels between TDP-43 and HERV-K (Figure 4). If further studies confirm this observation, that relationship might be considered as a possible marker of ALS progression. With regard to this connection, it would be useful to conduct a prospective study and monitor the patients in ALS evolution.
Currently, the regulatory role of TDP-43 in the HERV-K expression is well known [13], but further and more in-depth studies are needed to understand whether a correlation between the humoral response against TDP-43 and HERV-K exists.

Conclusions
Our investigation concludes that the epitopes of TDP-43 and HERV-K envelope identified by the Immune Epitope Database (IEDB) are highly immunogenic and recognized by the humoral response found in ALS patients in comparison with the weak response of the healthy control subjects analyzed. We are currently enrolling for a larger study to deepen our hypothesis. In particular, clinical data regarding ALS are going to be collected, in order to point out any putative relationship. It would be interesting to conduct a prospective study to better understand if our results are the consequences of the pathology, or if they are the result of a connection that might exist between TDP-43 and HERV-K env.
Currently, no specific biomarkers are available for the differential diagnosis of ALS, thus diagnostic and prognostic biomarkers for ALS remain a major unmet clinical need, and the results obtained in this work suggest a need for further study in this direction with a larger number of patients and controls.