A group of 40 patients diagnosed with relapsing-remitting multiple sclerosis (MS-RR), 6 with primary progressive multiple sclerosis (MS-PP), 5 with secondary progressive multiple sclerosis (MS-SP), 10 with clinically isolated syndrome (CIS) and 10 with other neurological disease (OND) participated in the study ( Table 1 ). Nine sera derived from healthy donors were used as negative controls. The patients provided their written informed consent to participate in this study. All groups were randomly selected and consisted of age- and sex-matched, HIV-1 negative patients.

IgG were purified from the sera of patients by HiTrap Protein G HP column according to the manufacturer’s instructions. The volume of the purified IgG from each sample was measured and the dilution factor with respect to the purified serum starting volume was calculated. The purified IgG were then tested in TZM-bl neutralization assays.

Two HIV-1 pseudoviruses, tier 1 and tier 2 virus strains, respectively with a higher and lower level of neutralization susceptibility were employed in the TZM-bl assay: SF162 (Clade B, Tier 1 virus), QH0692 (Clade B, Tier 2). Viral stocks were generated and titered. In particular, 293 T cell-derived stocks of pseudoviruses were generated by proviral DNA transfection using FuGENE HD, according to the manufacturer’s protocol (Promega, Madison, WI). Viral supernatants were harvested 72 h post transfection and clarified at 1800 rpm for 20 min. The virus stocks were analysed for firefly luciferase expression in the TZM-bl cell line. Four replicates of five-fold dilutions of viruses were added to 96 flat-bottomed plate wells containing 10⁴ TZM-bl cells per well, in DMEM supplemented with 10% FBS and 7.5 μg/ml DEAE-dextran in a final volume of 200 μl. After 48 h of incubation at 37°C, 100 μL of culture medium were removed from each well and replaced with 100 μL of Bright Glo luciferase reagent (Promega). After 2 min of incubation, 100 μL of the cell lysate were transferred to a 96 well white solid plate and luminescence was measured using a Victor Light 2030 luminometer (Perkin Elmer). Neutralization titers were expressed as ID50 values, defined as the virus dilution required to achieve half maximal infection (Reed–Muench calculation).

In order to assess the neutralizing activity of human and mice samples, the infectivity reduction was measured as a reduction in Luc reporter gene expression after a single round of virus infection in TZM-bl cells with pseudotyped viruses. In order to demonstrate the specificity of HIV neutralization, an HIV-unrelated virus (VSV-G virus, strain SVA.MLV#922) was also included. Briefly, 200 TCID50 of pseudoviruses in 50 μL of culture media were incubated with 100 μL of two different dilutions of sera (1:20 and 1:60) by using D-MEM with 10% FBS in a 96-well plate for 1h at 37°C. A 100 μL solution of TZM-bl cells (10⁴ cells per well) containing 18.75 μg/ml DEAE dextran was added (final concentration 7.5 μg/ml); the cultures were then incubated at 37°C for 48h. Assay controls included replicate wells of TZM-bl cells alone (cell control) and TZM-bl cells with virus (virus control). Two replicates were performed per each point. Percentage of neutralization (% NT) was expressed as reduction in relative luminescence units (RLU) compared to the virus control wells, after subtraction of cell control luminescence.

Seven-week and 28 weeks-old C57BL/6 female mice were purchased from Charles River. All mice were housed in specific-pathogen-free conditions, in roomy cages, allowing free access to food and water with a constant light/dark cycle. All efforts were made to minimize animal suffering and to reduce the number of mice used, in accordance with the European Communities Council Directive of November 24, 1986 (86/609/EEC). All procedures involving animals were performed according to the animal protocol guidelines prescribed by Institutional Animal Care and Use Committee (IACUC) at San Raffaele Scientific Institute (Milan, Italy). Mice were immunized by subcutaneous injections with 100μl of 2 mg/ml Myelin Oligodendrocyte Glycoprotein (MOG)35–55 (from Multiple Peptide System) in incomplete Complete Freund’s Adjuvant (CFA) containing 8 mg/ml Mycobacterium tuberculosis (strain H37Ra; Difco). Pertussis toxin (Sigma) (500 ng/mice) was injected on the day of the immunization and again two days later. EAE was successfully induced in 100% of mice, although with sign severity and onset variability.

Mice were weighed and scored for clinical signs daily up to the day of culling. Clinical assessment of EAE was performed in a blinded fashion according to the following scoring criteria: 0 = healthy, 1 = limp tail, 2 = ataxia and/or paresis of hindlimbs, 3 = paralysis of hindlimbs and/or paresis of forelimbs, 4 = tetraparalysis, and 5 = moribund or death.

Four different adjuvants were compared in the study; Freund’s adjuvant, aluminum hydroxide, RIBI, and Montanide ISA 720 were added to immunogen preparations. All adjuvants were provided by Sigma-Aldrich. Mice were immunized i.p. with 50 μg/mice of trimeric gp140 (CN54). Blood samples from immunized mice were collected from the eye before immunization (T0) and 7 (T1) and 14 (T2) days after immunization, centrifuged at 1500 rpm to recover plasma, heat inactivated at 56°C for 30 minutes, and stored at -20°C before analysis.

(i) Total IgG ELISA. To quantify serum immunoglobulins, Maxisorp Immunoplates (Thermo Scientific) were coated with 0,1 μg/well with Goat anti-mouse IgA + IgG + IgM (H+L) antibody (Kpl-SeraCare) in PBS for 1 h at 37°C. The plates were saturated for 1 h with 1% skim milk powder (Sigma-Aldrich) in PBS. Commercial preparation of mouse IgG (Sigma- Aldrich) was used as standard at concentrations ranging from 4 to 0.06 μg/ml (2-fold serial dilutions) to generate a calibration curve. Duplicate 2-fold dilutions of heat-inactivated sera starting from 1:1000 and standard were incubated 1 h at 37°C. Then, 1:5000 peroxidase-conjugated goat anti-mouse IgG (Sigma- Aldrich) was added and incubated for 30 min at 37°C. The enzymatic reaction was developed with the TMB Microwell Peroxidase Substrate System (KPL, Gaithersburg, MD, USA) and read at 492 nm. The IgG concentration of the samples was interpolated from four-parameter standard curve.

(ii) gp140 ELISA. Maxisorp Immunoplates (Thermo Scientific) were coated overnight at 4°C with 0.1 μg/well gp140 (CN54) diluted in 50 mM pH 9.5 carbonate buffer. The following day, both Env-coated and uncoated plates were blocked for 1 h at 37°C with PBS supplemented with 5% skim milk, 5% FBS, and 0.1% Tween 20. Duplicate 2-fold dilutions starting from 1:20 of heat-inactivated sera were added at 50 μl/well and incubated for 1 h at 37°C. The plates were washed and 1:5000 peroxidase-conjugated goat anti-mouse IgG (Sigma- Aldrich) was added for 30 min at 37°C. HIV-1-infected and healthy individuals were used as positive and negative controls, respectively, using 1:2000 biotynylated-goat anti-human-IgG (Southern Biotech) followed by 1:3000 HRP-conjugated streptavidin (Vector). Plates were developed by adding 100 μl/well of TMB and the reaction stopped with 50 μl/well of 10% H2SO4 after 5 minutes. Absorbance was read at 450 nm using a Microplate Reader (Biotek). The absorbance values recorded for uncoated plates were subtracted from those on the Env-coated plates. Cut-offs for each sample dilution was subsequently determined by calculating the mean “adjusted” absorbance + 3 SD for negative controls. The reciprocal endpoint titer of antibody in sera was defined as the last sample dilution that produced an absorbance greater than the cut-off.

The two-tailed Student or Mann–Whitney and Kruskal–Wallis test were used to compare the results obtained by different groups of animals. Kaplan–Meier survival analysis was conducted to compare animal groups for their effectiveness in predicting clinical outcomes. The level of significance was considered at p=0.05. GraphPad Prism 8.2.0 (GraphPad Software, Inc., San Diego, CA, USA) was used to perform statistical analysis.

Based on previously published findings (1), we wondered whether the production of bnAbs was restricted to SLE patients and its relative mouse models, or bnAbs were developed in other autoimmune diseases as well, such as MS. To this aim, were tested to investigate their inhibitory activity on the neutralization-resistant tier 2 virus HIV-QH0692 strain. The neutralization (NT) activity of the sera derived from patients with MS, although weak, was significantly higher than healthy controls. No difference in the percentage of neutralization among the different MS forms and patients with other neurological diseases has been observed, probably due to some soluble factors present in the sera of all patients tested ( Figure 1A ). In order to verify whether the neutralization activity was due to immunoglobulin (Igs) fractions rather than soluble factors, the Igs of the 8 sera showing the higher % NT (1 CIS, 3 MS-RR, 2 MS-PP, 1 MS-SP, 1 OND) were further purified. Only 2 sera from MS-RR patients confirmed the previously observed neutralizing activity, indicating that it was specifically due to Igs ( Figure 1B ). Interestingly, these patients were not experiencing a relapse at the moment of the sampling, suggesting that no inflammatory responses, with the consequent release of chemokine or cytokines, were involved in the block of the pseudovirus infection.

After confirming that a small number of patients with MS displayed IgG-specific HIV-1 neutralizing activity in sera, we investigated the role of autoimmunity on the production of anti–HIV-1 neutralizing antibodies. We tested, therefore, the possibility to elicit such Abs at high titres in the MS mouse model EAE.

We compared four different adjuvants to assess their efficacies in terms of IgGs yield and safety profiles: Freund’s adjuvant, aluminum hydroxide, RIBI, and Montanide ISA 720. Freund’s adjuvant induces systemic humoral and cellular immunity, but it can often cause chronic inflammation and tissue necrosis at the injection site; similar to Freund’s adjuvant but less aggressive is montanide ISA 720, a stable oil-water emulsion; aluminium, widely used as adjuvant for its overall safety, generates mild inflammatory responses and achieves immune memory; RIBI induces the presentation of hydrophobic antigenic epitopes and generates humoral responses to epitopes in their native conformation (19). Based on the comparative safety profiles and efficacies on the induction of IgGs between adjuvants, montanide was chosen for further experiments ( Figure 2A ).

We evaluated the primary anti-Env antibody response in 28 weeks-old control and EAE C57BL/6 mice. Control and EAE C57BL/6 mice were immunized with montanide alone or together with the trimeric gp140 to produce specific anti-HIV Abs. Trimeric gp140 represents the ectodomain of Env, composed by the entire gp120 and 20 KDa of gp41, and it is used as immunogen to induce a more potent humoral immune response against HIV-1 compared to the monomeric gp120, since it mimics native Env on the virion (20).

We first measured the amount of total IgGs in serum at 7 (T1) and 14 (T2) days post immunization (p.i.). Immunization with gp140 did not show substantial differences compared to montanide alone on the amount of total IgGs in both control mice and EAE mice at any experimental time ( Figure 2B ). As observed for the total amount of IgGs, neither control mice nor EAE mice mounted a statistically significant specific anti-gp140 IgG response at any experimental time point ( Figure 2C ). Taken together, these results suggest that the severity of the disease status we have registered ( Figure 2D ) might affect the elicitation of the humoral immune responses in EAE mice against the gp140 antigen.

We next asked whether the Env-specific antibody response elicited in EAE mice (although not statistically significant, as shown in Figure 2C ) was able to neutralize HIV-1 using a standardized in vitro TZMbl assay. Serum neutralization was tested against two HIV-1 clade B viruses representing the neutralization-sensitive tier 1 (SF162) and neutralization-resistant tier 2 virus strains (QH0692). As showed in Figure 3 , none of these sera displayed neutralizing activity against neither the SF162 strain nor the QH0692 strain, indicating that gp140 Abs elicited in control mice did not contribute to neutralization.

In order to investigate if the negative correlation between HIV-1 and MS might be related to the presence of bnAbs, we attempted at inducing the development of such Abs in mice to study their potential effect on the onset of the EAE or its clinical course.

To this aim, 7 weeks and 28 weeks-old C57BL/6 EAE mice were immunized with montanide alone or together with the trimeric gp140 to elicit specific anti-HIV Abs and gp-140-specific Abs were measured 7 and 14 days p.i. by ELISA. As showed in Figure 4 , no statistically significant difference in terms of gp140-specific Abs production was measured neither 7 days p.i. nor 14 days p.i., confirming the challenges of the induction of anti-HIV Abs through vaccination.

To further analyze the sera of immunized mice, we evaluated their neutralization activity on the HIV-QH0692 strain. As represented in Figure 5 , neither 7 weeks-old mice sera ( Figure 5A ) nor 28 weeks-old mice sera ( Figure 5B ) showed any statistically significant neutralizing activity after immunization with adjuvant alone or in combination with gp140.

We wanted to verify the effect of the immunization of onset and/or the clinical course of EAE. To this aim, EAE was induced in 7- and 28-weeks old C57BL/6 mice 14 days after the immunization with gp140. The experimental groups composition of mice was as follows: group 1: 10 untreated C57BL/6 mice (healthy controls); group 2: 10 untreated C57BL/6 mice (EAE controls); group 3: 10 C57BL/6 mice receiving montanide ISA 720 and EAE induction; group 4: 10 C57BL/6 mice receiving gp140 and EAE induction. Mice were weighed and scored for clinical signs daily up to the day of culling. As shown in Figure 6 , the disease was less severe in 7-weeks old mice immunized with gp140 compared to untreated mice, but no difference was observed between those mice receiving gp140 and those receiving the adjuvant alone ( Figure 6A ). The same phenotype was observed for mice weights ( Figure 6C ). No differences neither in terms of disease severity nor mice weight was observed in 28 weeks-old mice immunized with gp140 ( Figures 6B, D ). Taken together, these results showed that gp140 immunization did not have any effect on the clinical manifestations of EAE, probably due to the inefficiency in gp-140-specific Abs elicitation.