DOI: 10.31038/EDMJ.2019345

Introduction

Diabetes is associated with a substantially increased risk of certain neurovascular and neurodegenerative complications, e.g. stroke, dementia, Parkinson’s disease, major depressive disorder [1], through complex and poorly-defined mechanisms. We previously reported the occurrence of activating 5-HT2A receptor IgG autoantibodies in plasma or serum from older adult diabetes suffering with major depressive disorder [2], Parkinson’s disease or dementia [3]. Acute neurite retraction and accelerated mouse neuroblastoma N2a cell death induced by the autoantibodies in cell culture was partially or completely prevented by co-incubation (of IgG autoantibodies) with selective antagonists of the 5-HT2A receptor [2,3].

The second extracellular loop region of several different G-protein coupled receptors lies adjacent to the receptors’ orthosteric binding pocket [4]. In subsets of human dilated cardiomyopathy [5] or in eclampsia [6], spontaneously- occurring auto antibodies which targeted the second extracellular loop region of the beta-1-adrenergic or the angiotensin II, type 1 receptor, respectively, caused G-protein coupled receptor activation. In the present study, we tested a hypothesis that IgG autoantibodies in older adult diabetes having angiopathic and/or neurodegenerative complications cause 5-HT2A receptor activation via binding to the second extracellular loop region of the 5-HT2A receptor.

The 5-HT2A receptor is highly expressed in specific brain regions underlying cognition, perception and mood regulation [7]. It is also expressed on vascular smooth muscle cells where it plays a role in the regulation of arterial vascular tone [8]. Autoantibodies targeting 5-HT2A receptors in the central nervous system and the peripheral vasculature could serve as a biomarker for complications associated with refractory hypertension (e.g. stroke, chronic kidney disease) and/or neurodegeneration (dementia, Parkinson’s disease, retinal degeneration).

Participants and Methods

Participants

Men and women enrolled in the Diabetes or Endocrinology clinics at the Veterans Affairs New Jersey Healthcare System (VANJHCS), East Orange and Lyons NJ signed an informed consent for VANJHCS Institutional Review Board-approved study participation prior to blood drawing.

Patient subgroups

Patient 1: A 57-year-old man with morbid obesity, type 2 diabetes, stage 3 Chronic Kidney Disease (CKD) and refractory hypertension who suffered a branch retinal vein occlusion and Transient Ischemic Attack (TIA). He underwent renal biopsy which revealed hypertensive glomerulosclerosis (Fig 1).

Figure 1. Enzyme linked immunosorbent assay using the synthetic peptide Q..N-18 as the solid-phase antigen. Results are arbitrary absorbance units (OD) in a one-fortieth dilution of the protein-A eluate fraction of plasma or serum from patients with uncomplicated type 2 diabetes (N=6), Parkinsons disease (PD) (N=17), cerebrovascular accident (CVA) (N=7), major depressive disorder (MDD) (N=12) or dementia (N=7). Dashed line indicates background absorbance level of 0.04 absorbance units (dotted line).

Patient 2: A 72 -year-old man with twenty-year history of type 2 diabetes, discoid lupus erythematosus, and adult onset retinitis pigmentosa who had also suffered with a central retinal artery occlusion.

Patient 3: An 81 year-old-man who suffered with autoimmune thyroid disease and juvenile-onset retinitis pigmentosa (Stargardt disease). He has a ten-year history of type 2 diabetes without microvascular complications. Family history is positive for Stargardt disease in one brother.

Blood drawing

Blood was drawn in the morning in fasted participants.

Protein-A Affinity Chromatography

Protein An affinity chromatography was carried out as previously reported [2].

Synthetic peptide synthesis

A synthetic peptide having the amino acid sequence QDDSKVFKEGSCLLADDN, hereafter referred to as Q..N-18 or peptide 1, was synthesized at Lifetein Inc.(Hillsborough, NJ) and had > 92% purity. Q…N-18 had identical amino acid sequence to the region of the human 5-HT2A receptor comprising the second extracellular loop (ECL2) (Uniprot KB- P28223 [216 – 233] 5HT2A_Human). Three shorter overlapping peptides having the following amino acid sequences: QDDSKVF (peptide 3), VFKEGSC (peptide 4), SCLLADDN (peptide 2) were synthesized at Lifetein, Inc. and had purities > 96%. The latter three peptides were used in epitope analysis of the Q..N-18 (ECL2) region targeted for binding in subsets of human neurovascular pathologies’ autoantibodies.

Enzyme Linked Immunosorbent Assay (ELISA)

Ninety-six well plates were coated with the Q..N-18 peptide in PBS at a concentration of 65 micrograms/mL for two hours at 25 degrees C. The plates were washed three times with 0.1% Triton X-100 in PBS (PBS/Triton). Next the plate was blocked with 3% BSA in PBS for 1 hour at room temperature. Following 3 washes with PBS/Triton, 0.5–10 μg/mL concentrations of the patient IgG antibody (obtained by protein an affinity chromatography of plasma or serum) or a control sample was added to wells in duplicate and incubated for 1 hour at room temperature. The plate was washed 3× in PBS/Triton. Next HRP-conjugated goat anti human IgG (Sigma, St. Louis, MO) was added to each well at a 1: 3000 final dilution. After 1 h incubation, the plate was washed 3× in PBS/Triton. Next 150μL of a substrate solution containing 0.4 mg/mL ortho-phenylenediamine (Sigma, St. Louis, MO) and H₂ O₂ was added to each well and the reaction was monitored for color development. After 5 minutes, the reaction was stopped by the addition of 50μL 8 M H₂ SO₄ and the optical density was read at 490 nm in a microtiter plate reader (Dynatech Inc). Results are expressed as basal OD₄₉₀ compared to wells to which PBS alone was added.

Mouse neuroblastoma N2 cells

N2A cells were generously provided by Dr. Smith Varia and Dr. Janet Alder (Department of Neurosciences, Rutgers-Robert Wood Johnson Medical School). The cells were cultured in DMEM with 10% fetal calf serum.

N2A Acute Neurite Retraction assay

Acute neurite retraction assay was performed as previously reported [2]

N2 cell survival assay

Cell survival assays were carried out as previously reported [2]

Chemicals

All chemicals (with the exception of synthetic peptides) were from Sigma Chem. Co, Inc.

Protein determinations

Protein assays were carried out as previously reported [2]

Results

Baseline clinical characteristics in the study patients

The baseline clinical characteristics are shown in Table 1. The study cohort included 49 older adults with type 2 diabetes mellitus (T2DM) and 7 older adults without diabetes. The patients without diabetes were slightly younger on average, but did not differ significantly from the diabetic patients in their mean body mass index 31.0 vs 34.9 kg/m² (Table 2).

Table 1. Baseline clinical characteristics in the 56 study patients

*Major depressive disorder (n=2), Parkinsons’disease (n=3), Schizophrenia (n=1), Graves ophthalmopathy (n=1)

Table 2. Comparison of age and body mass index in diabetic and non-diabetic patients.

Mean binding to a linear synthetic 5-HT2A receptor peptide did not differ significantly in autoantibodies from Parkinson’s disease or major depressive disorder patients with or without type 2 diabetes (Table 3). These data suggest that the ‘susceptible’ population harboring increased 5-HT2A receptor peptide autoantibodies is likely to include older persons with and without type 2 diabetes mellitus.

Table 3.Mean 5-HT2A receptor peptide binding in diabetic vs nondiabetic patients having Parkinson’s disease or major depressive disorder.

Results are mean absorbance units AU +/ (SD); [ ] number of patients

Increased 5-HT2AR synthetic peptide binding in protein-A eluates from subsets of diabetic angiopathy and/or neurovascular complications

A one-fortieth dilution of the protein-A eluate of plasma (2–8 µg/mL IgG) was tested for binding to the linear synthetic peptide Q…N-18 having an amino acid sequence identical to that of the second extracellular loop of human 5-HT2A receptor. Mean binding level was significantly increased (P < 0.01) in subsets of diabetes having Parkinson’s disease (n=17), dementia (n=7), stroke or TIA n=7), or major depressive disorder (n=12) compared to the mean level in age-matched older adult type 2 DM without significant angiopathy (i.e. retinopathy or nephropathy) (n=6; Fig 1). It was also significantly increased in diabetes patients with a coexisting systemic autoimmunity condition (i.e. discoid lupus erythematosus (n=1), ankylosing spondylitis (n=1), Graves orbitopathy (n=2), rheumatoid arthritis (n=1)) (Table 4). Mean binding in the protein-A eluate fraction of T2DM plasma (IgG) was significantly increased in patients having significant retinopathy vs. no retinopathy, or diabetic nephropathy vs. no nephropathy (Table 5). There was no significant difference in plasma IgG autoantibody binding level in T2DM patients with or without atrial fibrillation, T2DM with or without obstructive sleep apnea or T2DM with or without co-morbid cancer (Table 5). Taken together, these data suggest associations among diabetic angiopathy, neurodegenerative disorders and certain systemic autoimmune diseases with increased level of IgG autoantibodies that binds to 5-HT2A receptor, second extracellular loop region, and linear synthetic peptide.

Table 4. Mean 5-HT2A receptor peptide binding in type 2 diabetes without angiopathy vs diabetes with a neurodegenerative disorders or a co-morbid systemic autoimmune condition^^.

Results are mean AU +/- SD; ^- without any neurodegenerative disorder; ^^-N=1 each having discoid lupus erythematosus, ankylosing spondylitis, Graves disease, celiac disease, rheumatoid arthritis; *t-test compared to group having diabetes without angiopathy. PD-Parkinson’s disease, MDD-major depressive disorder, CVA-cerebrovascular accident, TIA-transient ischemic attack. ** includes 3 patients without diabetes; ***includes 2 patients without diabetes

Table 5. Association between5-HT2AR, ECL2 receptor peptide binding and baseline diabetic micro-vascular or other complications.

Results are mean AU +/- SD; Patients having co-morbid systemic autoimmune condition (n=5) were excluded except in two cases having retinal neurodegeneration.

Lack of association between IgG plasma autoantibody binding to 5-HT2A receptor peptide and patient age or diabetes duration

There was no significant association between mean IgG autoantibody binding to the 5-HT2A receptor peptide and baseline patient age, body mass index, glycosylated hemoglobin, or duration of diabetes (data not shown in Tables).

Dose-dependence and titer of diabetic protein-A eluate binding to Q..N-18

Titer and potency of IgG autoantibody binding to the Q..-N-18 linear synthetic peptide was substantially increased in subsets of diabetes suffering with recurrent stroke, diabetes with co-morbid discoid lupus erythematosus and retinitis pigmentosa, diabetes having dementia, and diabetes with Parkinson’s disease. Peak binding and titer was not elevated in representative older type 2 diabetes without neurodegeneration and without significant angiopathy (i.e. retinopathy and nephropathy), i.e. < 2-fold above background absorbance at IgG concentrations tested between 1–2 µg/mL (Table 6).

Table 6. Potency and titer of receptor peptide autoantibodies in representative patients.

Absorbance units= 0.04 is background level. 1 µg/mL IgG ~ 6.7 nM.

ND– neurodegenerative disorder, DM– diabetes mellitus; Titer– concentration of autoantibodies causing ~50% of maximal binding.

Correlation of receptor peptide binding with IgG neurotoxicity (neurite retraction, N2A cell loss)

There was a significant correlation between 5-HT2A receptor synthetic peptide binding and acute N2A neurite retraction in the protein-A eluates from twenty-five patients tested (Fig 2) including representative patients from various pathologic subgroups (Table 7). High receptor peptide binding IgG was associated with significantly greater acute N2A neurite withdrawal (compared to low-binding IgG) at each of several different dilutions tested (Table 8).

Figure 2. . Correlation between 5-HT2A receptor synthetic peptide binding and acute N2A neurite retraction in the protein-A eluates from twenty-five patients with neurovascular or neuropsychiatric complications. A 1/40^th dilution of the protein-A eluate fraction was incubated with N2A cells and acute neurite retraction was determined as described in Materials and Methods. Results were correlated with binding to Q..N-18 in an ELISA using a 1/40^th dilution of the protein-A eluate of serum or plasma.

Figure 3. A 1/40^th dilution of the Pt 1 refractory hypertension plasma autoantibodies(AAB) was incubated with N2A neuroblastoma cells alone (solid bar) or in the presence of a 20 µg/mL concentration of peptide 2 ((SCLLADDN), peptide 3 (QDDSKVF),or peptide 4 (VFKEGSC)(open bar). Results are (mean + SD) acute neurite retraction determined as described in Materials and Methods.

Table 7. Association between autoantibody- 5HT2ARreceptor peptide binding and autoantibody-induced acute neurite retraction by representative patient subgroup.

*Neurite retraction scale: % neurite shortening after 5 minutes’exposure to autoantibodies , i.e. 0–10% = (-); 11–24% (+); 25–33% (++); 34–50% (+++); 51–74% (++++); 75–85% (+++++).

Table 8. Dose-dependent, increased autoantibody-induced neurite retraction in DM having neurodegeneration or cancer vs. DM without neurodegeneration or cancer

Results are mean+ SD % neurite retraction after 5 minutes exposure to indicated dilution of autoantibodies; NR- nodetectabeneurite retraction. ^ P <0.002; * P < 0.02– compared to neurite retraction at same autoantibody dilution in low peptide binding IgG autoantibodies subgroup; AU- absorbance units, SD-standard deviation.Uncomplicated diabetes mellitus- without microvascular complications.

‘High- binding’ IgG caused significantly greaterN2a neuron loss (after 24 hours incubation) than an identical concentration of ‘low-binding’ IgG (Table 9). Taken together these data suggest that IgG autoantibody binding to the linear synthetic 5-HT2A (ECL2) receptor peptide is correlated with IgG autoantibody-induced N2A neurite retraction and accelerated N2A neuron loss.

Table 9. Association between autoantibody- receptor peptide binding and autoantibody-induced accelerated N2A neuron loss.

Results are mean +/- SD. Neuron survival was determined after 24 hours incubation as described in Materials and Methods.

Soluble ECL2 peptide inhibits DM autoantibody-induced N2A neurite retraction

An 18-mericlinear synthetic peptide (Q..N-18) having an identical amino acid sequence to the ECL2 of the 5-HT2A receptor dose-dependently inhibited mouse N2A neuroblastoma cell neurite retraction induced by a sixty Nano molar concentration of the protein-A eluate fraction of plasma from Patient 1 type 2 diabetes suffering with refractory hypertension, retinal vein occlusion, and transient ischemic attack (TIA). The Patient 1 refractory hypertension autoantibodies alone caused 90% neurite retraction in N2a cells (Table 10). Half-maximal inhibition of autoantibody-induced acute neurite retraction occurred at a 2 µg/mL concentration of Q..N-18, corresponding to ~ 1 µM concentration of the peptide (Table 10).

Table 10. Dose-dependent inhibition of autoantibody-induced acute N2A neurite withdrawal by co-incubation with 18-meric Q..N linear synthetic peptide Autoantibody (one sixtieth dilution) + (Q…N-18) at

The indicated concentration of theQ..N-18 linear synthetic 5HT2A receptor peptide was co-incubated with a 100 nM concentration of the Patient 1, diabetes (DM) refractory hypertension autoantibodies. Acute N2A neurite retraction was assessed after 5 minutes as described in Materials and Methods.

Epitope mapping of region within 5-HT2AR, ECL2 peptide targeted by autoantibodies

We next performed epitope ‘mapping’ of the region within the second extracellular loop region of 5-HT2A receptor targeted by the type 2 diabetes and neurodegenerative pathologies’ autoantibodies. Each of three short overlapping ECL2-region peptides was preincubated alone (for 5 minutes) with N2A cells prior to the addition of Patient 1, refraction hypertension IgG autoantibodies. Peptide 3or4 (at 20 µg/mL concentration) had no significant effect on Patient 1IgG autoantibody-induced neurite withdrawal. Peptide 2(20µg/mL) dose-dependently nearly completely prevented acute neurite retraction induced by the Patient 1, refractory hypertension IgG autoantibodies (Table 11).

Table 11. Identification of a short linear peptide that blocks autoantibody (IgG)-induced neurite retraction.

*A one-fortieth dilution of the Pt 1 diabetic refractory hypertension IgG autoantibodies was incubated in the presence or absence of a 20 µg/mL concentration of three different short linear synthetic peptides comprising portions of the second extracellular loop of 5-HT2A receptor. Results are mean +/SD acute N2A neurite retraction after 5 minutes. Peptide 2 (20 µg/mL) was associated with 95% protection against the diabetic stroke pathologies IgG-induced neurite retraction. Peptide 3 or Peptide 4 at identical (20 µg/mL) concentration had no significant protective effect on IgG autoantibody-induced N2A neurite retraction.

Sub region-specific 5-HT2AR, ECL2 peptide protects against autoantibody neurotoxicity

A brief pre incubation with Peptide 2 (20 µg/mL) nearly completely prevented (99%) IgG-induced acute neurite withdrawal induced by a 50–100nM concentration of pathologies’ IgG autoantibodies in ten of ten patients tested (Table 12).

Table 12. Neutralization of diabetic autoantibody-induced N2A acute neurite retraction by a linear synthetic short peptide contained within the 5-HT2A receptor ECL2, i.e. Peptide 2.
Acute N2A neurite retraction in IgG autoantibodies

ECL2- second extracellular loop region of 5-HT2A receptor; PD-Parkinson’s disease

Co-incubation with peptide 2 (20 µg/mL) completely protected N2A cells from accelerated neuron loss caused by a1/100^th dilution of Patient 1, refractory hypertension IgG autoantibodies (Table 13). After 16 hours’ incubation (at 37 degrees C) with Patient 1 or the IgG autoantibodies from four different patients suffering with stroke, Parkinson’s disease, dementia or major depressive disorder, peptide 2 (20 µg/mL) afforded substantial neuro protection against accelerated neuron loss (Table 14).

Table 13. Neutralization of diabetic IgG-induced accelerated N2A cell loss by Peptide 2, a short linear synthetic peptide corresponding to a portion of 5-HT2AR second extracellular loop region.

A 1/100^th dilution of the diabetes (DM) refractory hypertension (HTN), plasma IgG autoantibodies was incubated for 16 hours in the presence or absence of a 20 µg/mL concentration of the linear synthetic peptide, Peptide 2. Results are mean +/- (SD) percent of basal N2A neuroblastoma cell survival.

Table 14. Neutralization of diabetic (IgG) autoantibody-induced accelerated N2A cell loss by Peptide 2, a short linear synthetic peptide corresponding to a portion of 5-HT2AR, ECL2.

Results are mean +SD N2A cell survival after 16 hours incubation of the indicated diabetic pathologies IgG autoantibodies in the presence or absence of 20ug/mL Peptide 2. PD-Parkinson’disease, MDD-major depressive disorder.

In silico test of 5-HT2A receptor peptide’s binding affinity for major histocompatibility complex class II (MHC-II) molecules

We used the Immune Epitope Database (IEDB) and an improved prediction tool [9] to test whether a portion of the Peptide 2 amino acid sequence corresponds to an ‘immune epitope,’i.e. a sequence which displays significant binding affinity to one or more alleles comprising MHC class II,HLA‐DR, HLA‐DQ, or HLA‐DP molecules. The 15-meric linear peptide sequence(s) SKVFKEGSCLLADDN which includes Peptide 2 (underlined) was tested for binding to MHC class II molecules. There was no significant high affinity binding to alleles comprising the HLA‐DR or HLA‐DP loci. The core amino acid sequence, KEGSCLLAD displayed significant binding affinity (i.e. K_a217nM), to one specific HLA-DQ allele, DQA10102-DQB10602.

Discussion

Type 2 diabetes is not an autoimmune disease. Yet the present data suggest humoral immune responses to the 5-hydroxytryptamine (serotonin) 2A receptor in subsets of older adult obese type 2 diabetes suffering with diabetic macular edema, and proteinuric nephropathy. Aging, diffuse vascular injury and visceral obesity may each contribute to humoral immunity to the 5HT2A receptor. A role for chronic inflammation and hemodynamic factors is suggested by the observation of significantly increased ECL2 receptor peptide binding in subsets of patients with nephropathy and chronic kidney disease. The high prevalence of increased 5-HT2A receptor peptide binding in autoantibodies from patients suffering with Parkinson’s disease (14/17), major depression (9/12), stroke (6/7), and dementia (4/7), is unlikely to represent ‘non-specific binding’ since autoantibody binding (in the ELISA) was significantly associated with both autoantibody-induced neurite retraction and accelerated N2A cell loss. The correlation between receptor peptide binding and neurotoxicity demonstrated here suggests that neurovascular IgG autoantibody binding to the5-HT2AR second extracellular loop causes receptor activation which as previously reported was positively coupled to PLC/IP3/Ca2+ and RhoA/ROCK signaling pathway activation[2,3].

The second extracellular loop region loop lies in close proximity to the Orthosteric Binding Pocket (OBP) [4] and has a role in preventing normal constitutive receptor activation in biogenic and trace amine G-Protein Coupled Receptors (GPCRs) [10]. Neurovascular pathologies’ autoantibodies appeared to target a conserved sub region in ECL2 [10] important in promoting sustained activation of the 5HT2B and 5HT2A receptors by the hallucinogenic drug Lysergic Acid Diethylamine (LSD) [4]. In a recent study by Wacker et. al. [4], the atomic structure of LSD complexes to the 5-HT2B and -2A receptors was reported. Lysergic acid diethylamide-known to have long-lasting effects in humans-exhibited unusually long residence times at the 5-HT-2B and -2A receptors, i.e. off-reaction times of ~ 45 minutes and ~220 minutes, respectively [4]. Wacker and co-authors reported that amino acid residues 207–214 in 5-HT2BR form a ‘lid’ which blocks the opening to the orthosteric binding pocket likely accounting for LSD’s prolonged off-reaction time at the receptor [4]. Of interest, seven of the eight amino acid residues comprising the inhibitory Peptide 2 sequence SCLLADDN (underlined here) form the corresponding ‘lid’ region in the 5-HT2A receptor, including the conserved amino acid residue, leucine 229 (bolded), which is present in all three 5-HT2 (-A, -B, and -C) receptor isoforms [10] One of the striking features of the neurovascular pathologies’ autoantibodies effect in N2A cells was its long duration of action, i.e. neurite retraction was essentially irreversible after 45 minutes’ or longer observation times. Serotonin or the 5HT2A receptor agonist DOI promoted neurite retraction which peaked after (3–5 minutes) and was reversible at longer times [3]. Taken together with the structural data on the LSD bound 5-HT2R [4], these data suggest that neurovascular autoantibodies may binding at a critical region in the second extracellular loop involved in stabilizing a ‘persistently-active’ conformation of the 5HT2A receptor.

Systemic autoimmunity is associated with a wide variety of autoantibodies capable of targeting diverse auto antigens. In the present study, highest binding to the 5-HTR2AR linear synthetic peptide occurred in two patients with a systemic autoimmune condition, i.e. discoid (cutaneous) lupus erythematosus (Pt 2) or HLAB27- positive ankylosing spondylitis. Heparan Sulfate Proteoglycans (HSPG) is among a small group of overlapping auto antigens reported in both systemic lupus erythematosus [11] and ankylosing spondylitis [12]. Heparan sulfate proteoglycans are strongly anionic at physiologic pH. They are abundantly expressed in extracellular matrix and on vascular and neuronal surfaces where they have role(s) in the maintenance of tissue barriers and serve as low-affinity co-receptors for cationic ‘heparin-binding’ growth factors, e.g. basic fibroblast growth factor (FGF-2) [13]. Neurotoxic and endothelial cell inhibitory autoantibodies previously reported in subsets of diabetes complicated by painful neuropathy, nephropathy, macular edema or primary open angle glaucoma [14] displayed increased affinity for heparan sulfate proteoglycan [14] or a heparin Sepharose column [15]. Glomerular HSPG are elaborated in diabetic nephropathy [16], and may be a target of heightened humoral autoimmunity especially under conditions of chronic inflammation which prevail in chronic kidney disease.

The putative dominant epitope targeted by 5-HT2A receptor activating autoantibodies in the present study contains a di-aspartic acid motif (SCLLADDN) not present in the 5-HT2B receptor [10]. It is interesting to speculate that the negative charge(s) associated with having di-aspartic acid residues in a solvent-exposed region of the receptor may provide a basis for strong electrostatic interaction with circulating IgG pathologies autoantibodies. In a prior report, neurotoxicity in diabetic dementia and PD dementia IgG autoantibodies was enhanced eight-fold following chromatography of plasma on a dextran sulfate affinity chromatography column [2]. Dextran sulfate (Liposorber) apheresis issued to lower excessive concentrations of cationic lipoprotein particles (very low density, and low-density lipoproteins) in the circulation in patients with familial hypercholesterolemia. It is possible that neurovascular pathologies autoantibodies co-purified with cationic lipoprotein particles based on having similar cationic surface charge characteristics which may contribute to an electrostatic interaction with anionic sulfate groups.

Prostate expresses high level of HSPG [17] which acts as a co-receptor for several different fibroblast growth factors having key role(s) in stromal proliferation underlying prostate cancer progression [18]. Among nine patients in the present study who had co-morbid prostate (n=8) or bladder (n=1) cancer, the mean level of 5-HT2AR autoantibody peptide binding was 0.14 AU, i.e. 3.5-fold higher than background (0.04 AU). Eight of the nine prostate or bladder cancer patients suffered with a neurodegenerative (n=5 PD, n=1 dementia) or neuropsychiatric disorder (N=1 MDD, N=1 schizophrenia). In most of these patients, the cancer diagnosis and treatment was established prior to the onset of neurologic or neuropsychiatric disease manifestations perhaps consistent with a paraneoplastic mechanism for humoral autoimmunity to 5-HT2A receptor activating autoantibodies. In a prior report, prostate cancer autoantibodies in patients suffering with fatigue/depression were highly neurotoxic and evoked large increases inward cationic current in (whole-cell patch clamped) rat hippocampal pyramidal neurons associated with long-lasting desensitization of excitatory synaptic inputs [14]. Taken together with the present data, these findings suggest that 5-HT2A receptor found on cortical pyramidal neurons or in other brain regions may mediate glutamatergic excitatory post-synaptic actions in response to 5-HT2A receptor- activating autoantibodies [19].

A short linear synthetic peptide comprising a region of the second extracellular loop of the 5HT2A receptor, i.e. peptide 2, largely prevented autoantibody- induced neurotoxicity perhaps by competing for binding to the region of the 5-HT2R targeted by a majority of the autoantibodies tested. The peptide antagonist (SCLLADDN) retained its neuro protective effect (against autoantibody-induced N2A cell loss) even after 16 hours’ incubation time (at 37 degrees C) in the presence of 10% fetal calf serum. A hallmark property of an auto antigen’s immune dominant epitope is its ability to survive complete intracellular proteolytic processing [20]. Our in silico analysis suggests that Peptide 2 having the sequence SCLLADDN may comprise an immune dominant peptide which is specifically ‘recognized’ by the major histocompatibility complex class II molecular system through high affinity binding interaction with the specific HLA-DQ allele, DQA10102-DQB10602. Of interest, DQA10102-DQB10602 was previously reported to have conferred markedly increased susceptibility to narcolepsy-cataplexy in certain populations [21]. Narcolepsy is a putative neurodegenerative process affecting hypocretin neurons located in the lateral hypothalamus [22]. Narcolepsy incidence rates in young adults were reported to have peaked following the 2009 H1N1 influenza pandemic [23] and related H1N1 vaccination regimes [24] suggesting molecular mimicry with certain strains of influenza viruses as one possible mechanism for T-cell driven auto immunity in the unknown etiology of a subset of ‘autoimmune’ narcolepsy-cataplexy in HLA-DQ susceptible persons [25].

Highest level of 5-HT2AR, ECL2 autoantibody binding, i.e. 6.25 times above background, occurred in an older man (Patient 2) having discoid lupus erythematosus, and near blindness secondary to central retina artery occlusion and retinitis pigmentosa. It is of interest that 5-HT2AR immune reactivity was reported on the terminals of photoreceptors and bipolar cells in rabbit retina [26]. In addition, 5-HT2AR antagonists afforded protection against light-induced photoreceptor degeneration in certain susceptible genetic strains of mice [27]. Plasma autoantibodies in Patient 3, an 81 year-old-man who suffered with autoimmune thyroid disease and juvenile-onset retinitis pigmentosa (i.e. Stargardt disease), displayed significantly increased binding (2.5-fold background) in the ELISA. Taken together, these data suggest systemic autoimmunity in which humoral autoantibodies are directed against many different autoantigen epitopes might provide a humoral-mediated mechanism contributing to retinal degeneration occurring in genetically-susceptible patients. The 5-HT2AR ECL2 peptide-based ELISA might be useful in screening family members of affected patients suffering with genetic forms of retinal degeneration. More in vivo study of the antagonist peptide is needed to determine whether it might be neuro protective in an animal model of retinal degeneration.

The source(s) of 5-HT2AR autoantigen in angiopathic type 2 diabetes is not known. Obese diabetes is associated with increased inflammation and the latter is associated with higher risk of thrombosis. Since 5-HT2AR is expressed on platelets, it is possible that diabetic microangiopathy predisposes to micro thrombi formation at sites of inflammation which may cause 5HT2A receptor to be taken up and processed by macrophages and/or other professional antigen presenting cells. Monocytes have been reported to adhere to platelets at sites of inflammation via specific interactions not observed between platelets and other subtypes of white blood cells [28].

Our study was cross-sectional and included mostly older men. More longitudinal study in diverse (unselected) populations is needed to estimate the actual prevalence of 5-HT2AR- activating autoantibodies and whether a dose-response relationship may exist between the autoantibodies and one or more kind of neurodegenerative disorder.

In summary, the present data provide the first evidence that subsets of major depressive disorder, Parkinson’s disease, dementia, small vessel stroke (e.g. retinal artery occlusion), refractory hypertension, and proteinuric nephropathy (in both diabetes and obese non-diabetic older persons) harbored IgG autoantibodies which bound to a region of the second extracellular loop of the human 5HT2A receptor [4] previously implicated in causing long-lasting receptor activation.

References

1. Hu G (2007) Type 2 Diabetes and the Risk of Parkinson’s Disease. Diabetes Care 30: 842–847.
2. Zimering MB (2018) Circulating Neurotoxic 5-HT2A Receptor Agonist Autoantibodies in Adult Type 2 Diabetes with Parkinson’s Disease. J Endocrinol Diabetes 5.
3. Zimering MB (2017) Diabetes Autoantibodies Mediate Neural- and Endothelial Cell- Inhibitory Effects Via 5-Hydroxytryptamine- 2 Receptor Coupled to Phospholipase C/Inositol Triphosphate/Ca2+ Pathway. J Endocrinol Diabetes 4.
4. Wacker D, Wang S, McCorvy JD, Betz RM, Venkatakrishnan AJ, et al. (2017) Crystal Structure of an LSD-Bound Human SerotoninReceptor. Cell 168: 377–389.
5. Holthoff HP, Zeibig S, Jahns-Boivin V, Bauer J, Lohse MJ, et al. (2012) Detection of anti-β1-AR autoantibodies in heart failure by a cell-based competition ELISA. Circ Res111: 675–684.
6. Dechend R, Gratze P, Wallukat G, Shagdarsuren E, Plehm R, et al.( 2005) Agonistic autoantibodies to the AT1 receptor in a transgenic rat model of preeclampsia. Hypertension 45: 742–746.
7. Xu T, Pandey SC (2000) Cellular localization of serotonin (2A) (5HT (2A)) receptors in the rat brain. Brain Res Bull 51: 499–505.
8. Alsip NL, Harris PD, Durrani GE (1991) Multiple Serotonin Receptors on Large Arterioles in Striated Muscle. J Vasc Res 28: 537–541.
9. Jensen KK, Andreatta M, Marcatili P, Buus S, Greenbaum JA, et al. (2018) Improved methods for predicting peptide binding affinity to MHC class II molecules Immunology 154: 394–406.
10. Michino M, Beuming T, Donthamsetti P, et al. (2015) What Can Crystal Structures of Aminergic Receptors Tell Us about Designing Subtype-Selective Ligands? Pharmacol Rev 67: 198–213
11. Shibata S, Sasaki T, Harpel P, Fillit H (1994) Autoantibodies to vascular heparan sulfate proteoglycan in systemic lupus erythematosus react with endothelial cells and inhibit the formation of thrombin-antithrombin III complexes. Clin Immunol Immunopathol 70: 114–123.
12. Wright C, Sibani S, Trudgian D, Fischer R, Kessler B, et al. (2012) Detection of multiple autoantibodies in patients with ankylosing spondylitis using nucleic acid programmable protein arrays. Mol Cell Proteomics 11: 00384.
13. Bernfield M, Sanderson RD (1990) a developmentally regulated cell surface proteoglycan that binds extracellular matrix and growth factors. Philos Trans R Soc Lond B Biol Sci 327: 171–186
14. Zimering MB, Alder J, Pan Z, Donnelly RJ (2011) Anti-endothelial and anti-neuronal effects from auto-antibodies in subsets of adult diabetes having a cluster of microvascular complications. Diabetes Res Clin Pract 93: 95–105.
15. Zimering MB, Anderson RJ, Moritz T, Ge L (2009) Investigators for the VADT Low plasma basic fibroblast growth factor is associated with laser photocoagulation treatment in adult type 2 diabetes mellitus from the Veterans Affairs Diabetes Trial. Metabolism 58: 393–400
16. Makino H, Ikeda S, Haramoto T, Ota Z (1992) Heparan sulfate proteoglycans are lost in patients with diabetic nephropathy. Nephron 61: 415–421.
17. Kosir MA, Quinn CC (1995) Sorting of heparan sulfate proteoglycan into matrix compartments of prostate adenocarcinoma cells. J Surg Res 58: 46–52.
18. Wu X, Kan M, Wang F, Jin C, Yu C, et al. (2001) A rare premalignant prostate tumor epithelial cell syndecan-1 forms a fibroblast growth factor-binding complex with progression-promoting ectopic fibroblast growth factor receptor 1.Cancer Res 61: 5295–5302.
19. Aghajanian GK, Marek GJ (1997) Serotonin induces excitatory postsynaptic potentials in apical dendrites of neocortical pyramidal cells. Neuropharmacology 36: 589–599.
20. Kim A, Sadegh-Nasseri S (2015) Determinants of immunodominance for CD4 T cells. CurrOpin Immunol 34: 9–15.
21. Mignot E, Lin L, Rogers W, Honda Y, Qiu X, Lin (2001) Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups. Am J Hum Genet 68: 686–699.
22. Nishino S, Ripley B, Overeem S, Lammers GJ, Mignot E (2000) Hypocretin (orexin) deficiency in human narcolepsy. Lancet 355: 39–40.
23. Nohynek H, Jokinen J, Partinen M, Vaarala O, Kirjavainen T, et al. (2012) S03 adjuvanted AH1N1 vaccine associated with an abrupt increase in the incidence of childhood narcolepsy in Finland. PLoS One 7: 33536.
24. Han F, Lin L, Warby SC, Faraco J, Li J, et al (2011) Narcolepsy onset is seasonal and increased following the 2009 H1N1 pandemic in China. Ann Neurol 70: 410–417.
25. Black JIII, Avula R, Walker D, Silber M, KrahnL, et al. (2005) HLA DQB1*0602 positive narcoleptic subjects with cataplexy have CSF IgG reactive to rat hypothalamic protein extract. Sleep 28: 1191–1192
26. Pootanakit K, Prior KJ, Hunter DD, Brunken WJ (1999) 5-HT2a receptors in the rabbit retina: potential presynaptic modulators. Vis Neurosci 16: 221–230.
27. Chen Y, Okano K, Maeda T, Chauhan V, Golczak M, et al.(2012) Mechanism of all-trans-retinal toxicity with implications for stargardt disease and age-related macular degeneration. J Biol Chem 287: 5059–5069.
28. Passacquale G, Vamadevan P, Pereira L, Hamid C, Corrigall V, et al (2011) Monocyte-platelet interaction induces a pro-inflammatory phenotype in circulating monocytes. PLoS One 6: 25595.

DOI: 10.31038/EDMJ.2019344

Abstract

Aims: To test whether neurite-inhibitory plasma autoantibodies in chronic schizophrenia activate Gq/11- and Gi- coupled signaling pathways downstream of 5-hydroxytryptamine 2A receptor activation; and for modulation of serotonergic signaling by the metabotropic 2/3 receptor agonist LY379268.

Methods: Plasma from five older adults with chronic schizophrenia and eight age-matched patients having another neuropsychiatric, immune or metabolic disorder was subjected to Protein-A affinity chromatography to obtain IgG autoantibodies. Mean neurite retraction (5 minutes) or cell survival (24 hours) was determined in mouse N2A neuroblastoma cells incubated with autoantibodies in the presence or absence of specific antagonists of the Gq/11/PLC/IP3R signaling pathway, Gi-coupled, beta-arrestin2-directed pathways, or LY379268.

Results: Chronic schizophrenia plasma autoantibodies- mediated dose- and time-dependent acute N2A neurite retraction was completely prevented by M100907, a selective 5-hydroxytryptamine 2A receptor antagonist. LY379268 promoted autoantibody-induced neurite retraction causing a shift-to-the-left in the dose-response curve. Antagonists of the RhoA/Rho kinase and Gq/11/PLC/IP3R signaling pathways blocked autoantibody-mediated neurite retraction. Chronic schizophrenia plasma autoantibodies mediated increased N2A cell survival which was blocked by LY379268, pertussis toxin, and antagonists of PI3-kinase- mediated survival signaling.

Conclusion: Schizophrenia plasma autoantibodies activate the 5-hydroxytryptamine 2A receptor positively coupled to Gq/11/PLC/IP3R pathway and RhoA/Rho kinase signaling activation in promoting acute N2A cell neurite retraction. Autoantibodies in a subset of patients experiencing hallucinations promoted increased N2A cell survival mediated (in part) via a pertussis-toxin sensitive, Gi-coupled, PI3-kinase-dependent mechanism. Positive modulation of 5-HT2AR-mediated neurite retraction by LY379268 suggests the autoantibodies may target (in part) the 5-HT2AR/mGlu2R heteromer.

Introduction

The serotonin 2A receptor (5-HT2AR) is highly expressed in cortical brain regions underlying normal perception [1]. The hallucinogenic drug lysergic acid diethylamine (LSD) causes long-lasting 5-HT2AR activation which is positively coupled to Gq/11- and β-arrestin-2- dependent signaling pathway activation [2]. Head twitch in mice infused with the hallucinogenic (reversible 5HT2AR agonist) 1-[2,5-dimethoxy-4-iodophenyl]- 2-aminopropane (DOI) required an additional contribution from metabotropic glutamate (mGlu)2R-mediated, Gi-coupled signaling since head twitch was not observed in DOI-treated mice harboring an mGlu2R knock-out mutation [3]. Heteromeric 5-HT2AR/mGlu2R complexes occur in the mammalian prefrontal cortex [4] and are thought to integrate serotonergic and glutamatergic signals via allosteric receptor-receptor interactions alters the balance between Gq11- and Gi-coupled signaling pathway activation [4].

Paranoid schizophrenia is a common disabling disease affecting ~1% of adults [5]. Evidence from epidemiologic studies [6] and a recent genome-wide association study [7] suggests a role for dysregulated acquired immunity in the pathophysiology of schizophrenia. A possible role for brain reactive autoantibodies was suggested by prior studies [6] including our report that plasma IgG autoantibodies in a subset of chronic schizophrenia potently suppressed neurite outgrowth and mediated strong depolarization in N2A mouse neuroblastoma cells [8]. Since major depression and Parkinson’s disease autoantibodies mediated long-lasting 5-HT2A receptor activation positively coupled to Gq/11 signaling [9,10], here we tested whether chronic paranoid schizophrenia plasma autoantibodies activate 5-HT2A receptor, Gq/11-mediated signaling leading to neurite outgrowth inhibition in N2A mouse neuroblastoma cells. A role for signaling cross-talk involving 5-HT2AR/mGlu2R heteromers was tested by comparing N2A acute neurite retraction induced by schizophrenia plasma autoantibodies in the presence or absence of the mGlu2/3R agonist LY379268. Since LY379268 was previously reported to suppress hallucinogen-induced Gi-coupled signaling at the 5HT2AR [4], we investigated biased 5-HT2AR-dependent, Gi-coupled signaling evoked by autoantibodies from subgroups of psychosis-prone vs. patients not experiencing recurrent visual or auditory hallucinations.

Participants and Methods

Participants– Outpatient men ranging in age from 47–78 years old provided informed consent for participation in the Institutional Review Board-approved study and were consecutively enrolled from the diabetes and endocrinology clinics at the Veterans Affairs New Jersey Health Care System (East Orange and Lyons, New Jersey).

Psychosis-prone patients

Patient 1: A 61- year- old man with a history of chronic paranoid schizophrenia and multiple recurrent hospitalizations for auditory hallucinations and one previous suicide attempt. The patient has type 2 diabetes mellitus of approximately eight years duration without microvascular complications.

Patient 2: A 55- year- old man with chronic paranoid schizophrenia, three previous suicides attempts, and type 2 diabetes (of five years known duration) without microvascular complications.

Patient 3: A 61-year-old man having major depressive disorder with mood-incongruent psychotic features and type 2 diabetes of fourteen years duration.

Patient 4: A 47-year-old man with chronic schizophrenia, and no history of type 2 diabetes. He had paranoid delusions without any suicide attempt.

Blood samples: Blood was drawn in the morning after an overnight fast. Plasma or serum was stored at -20 C.

Protein A affinity chromatography- Protein-A affinity was performed as previously described [10].

N2A mouse neuroblastoma cells: were cultured in Dulbecco’s modified Eagles medium (DMEM) containing 10% fetal calf serum. Cells were fed with fresh medium every 1–2 days, except for survival assays in which fresh medium was not added for up to 5 days prior to the addition of test autoantibody fractions.

Acute neurite retraction assay: % of basal neurite length in N2A cells expressing one or more proximally-located dendrite-like process was determined after interval exposure to test substances as previously reported [10].

N2A cell survival: MTT assay was performed 24 hours after incubation with test substances as previously reported [10].

Chemicals: all chemicals were obtained from Sigma, Co., Inc. (St Louis, MO), except YM-254890 (Focus Biomolecules) and LY 379268 (Tocris Bioscience).

Protein assay: protein concentration was determined using a modified bichichonic acid assay (Bio Rad. Inc.).

Statistics- Paired and unpaired T-tests were used to assess for statistically significant differences between groups or between treatments.

Results

Baseline characteristics in the study patients

The clinical characteristics in the study patients are shown in (Table 1). Patients having chronic schizophrenia did not differ significantly in their mean age, body mass index or glycosylated hemoglobin level from patients having Parkinson’s disease (n=5), dementia (n=1), or diabetic nephropathy (n=1).

Table 1. Clinical characteristics in the study participants

*Parkinsons disease (PD) (n=4), major depressive disorder (n=1), diabetic nephropathy (n=1), dementia (n=1). MVD- microvascular disease.

Acute N2A neurite retraction from schizophrenia plasma IgG fractions

Schizophrenia plasma autoantibodies (at IgG concentrations > 10 nM) caused dose-dependent N2A neurite retraction which significantly exceeded neurite retraction in an identical concentration of IgG from two older adult type 2 diabetes patients without microvascular, neuropsychiatric or neurodegenerative complications (Figure 1A). Neurite retraction in response to potent schizophrenia plasma autoantibodies, e.g. Pt 1, was linear, irreversible and more than 50% neurite withdrawal occurred after 5 minutes’ exposure time (Figure 1B).

A

B

Figure 1. A) Dose-dependent or B) time-dependent acute neurite retraction induced by diabetic schizophrenia plasma autoantibodies (solid bars; Pt 1) or (A) by the autoantibodies from two older diabetic patients not suffering with diabetic microvascular complications or any neuropsychiatric or neurodegenerative complication (open bars). Results represent mean + SE determined as described in Materials and Methods. B) Similar results were obtained in the plasma autoantibodies from all five of five schizophrenia patients tested.

Schizophrenia (SCZ) plasma autoantibody-induced neurite retraction was completely prevented by co-incubation with a two-hundred nanomolar concentration of the selective 5HT2A receptor antagonist, M100907 (Table 2). Slightly higher concentrations (500 nM) of the 5-HT2A receptor antagonists spiperone and ketanserin afforded significant protection (~50–80%) against SCZ plasma autoantibody-induced neurite retraction (Table 2). A higher concentration (1–10 µM) of selective antagonists of the endothelin A, angiotensin type 1, alpha-1-adrenergic and 5HT2B receptors (all Gq/11-coupled GPCRs) i.e. bosentan, losartan, prazosin and SB204741, did not significantly protect (0- 28%) against schizophrenia plasma IgG-induced neurite retraction (Table 2).

Table 2. Pharmacology of schizophrenia plasma autoantibody(AutoAB)-induced neurite retraction

A twenty nanomolar concentration of the Pt 1 schizophrenia protein-A eluate fraction of plasma was incubated with N2A cells in the presence or absence of the indicated concentration of each Gq/11, GPCR antagonist. Results are the mean +/- SD of two experiments

Mechanism of schizophrenia plasma IgG-induced neurite retraction

Major depressive disorder and Parkinson’s disease plasma IgG autoantibodies were previously reported to cause acute N2A neurite retraction through a mechanism involving activation of the RhoA/Rho kinase and PLC/IP3R/Ca2+ signaling pathways [9,10]. In the present study, neurite retraction induced by SCZ plasma IgG autoantibodies was completely prevented by co-incubating N2A cells with a ten micromolar concentration of the selective Rho kinase inhibitor Y27632 or a one micromolar concentration of the selective Gq11 inhibitor Y254890 (Table 3). In addition, a one micromolar concentration of the phospholipase C inhibitor U73122, and a 50 micromolar concentration of the IP3R antagonist 2-APB each significantly protected (72–83%) against SCZ IgG-induced neurite retraction (Table 3). Taken together, these data suggest that SCZ IgG induced neurite retraction likely involves activation of both RhoA/Rho kinase and Gq/11/PLC/IP3R/Ca2+ signaling pathways.

Table 3. Mechanism of acute N2A neurite retraction induced by schizophrenia plasma AutoAB

A forty nanomolar concentration of the Pt 1 or Pt 2 schizophrenia plasma autoantibodies (AutoAB) was incubated in the presence or absence of the indicated concentration of RhoA/Rho kinase (ROCK) inhibitor, or individual antagonists of the Gq11/PLC/IP3R signaling pathway. Results are (mean +/- SD) acute N2A neurite retraction occurring in response to Pt 1 or Pt 2 IgG autoantibodies.

Modulation of SCZ IgG-induced neurite retraction by LY379268, a potent mGlu2/3R agonist

Cross-signaling between Glu2/3R and 5HT2AR was inferred from differences in IgG-induced N2A acute neurite retraction occurring in the presence or absence of the mGlu2/3R agonist LY379268, at 5–10 µM concentrations (of LY379268) which alone had no effect on neurite retraction. The mGlu2/3R agonist LY379268 was previously reported to cause positive allosteric modulation of the 5HT2AR protomer affinity at 5HT2AR/mGlu2R heteromers [4]. A one microgram per milliliter concentration (~ 7 nM) of the potent Pt 1, SCZ plasma IgG autoantibodies caused 50% inhibition of N2A neurite outgrowth (Figure 2A). Pre-incubation (for 5 minutes) followed by co-incubation of N2A cells with a 10 micromolar concentration of LY379268 caused a ‘shift to the left’ in the dose-response curve of the Pt 1 plasma IgG-induced neurite retraction: 50% inhibition of N2A neurite outgrowth occurred at substantially lower, i.e.~ 0.5 microgram per milliter concentration (3.5 nM) of Pt 1 IgG autoantibodies (Figure 2A). The “potentiating effect” of 10 µM LY379268 on neurite retraction was more pronounced at low concentrations of SCZ plasma IgG ranging from (3.5–14 nM) (Figure 2A). Presumably, at these lower concentrations a high proportion of unoccupied 5HT2A receptors are available to undergo positive allosteric modulation via mGlu2R protomer binding to LY379268. The dose-response curve for DOI-induced neurite retraction underwent a ‘shift- to-the-left’ in the presence of (10 µM) LY379268: the concentration of DOI needed to evoke 25–30% neurite inhibition decreased from 10 µM (in the absence of LY379268) to 5 uM in the presence of LY379268 (Figure 2B). A saturating concentration of the reversible 5-HT2AR agonist DOI (20 µM) caused 40% peak neurite retraction compared to 50% neurite retraction induced by an ~2000-fold lower concentration (7 nM) of potent SCZ plasma IgG autoantibodies (Figure 3).

A

B

Figure 2. A) Diabetic schizophrenia Pt plasma autoantibodies or B) the hallucinogen DOI was incubated at the indicated concentrations in the presence (orange symbol) or absence (blue symbol) a 10 micromolar concentration of LY379268 in N2A cells. Each point represents mean acute neurite retraction which varied by < 10%. A) Similar result was obtained in the diabetic schizophrenia Pt 2 plasma autoantibodies.

Figure 3. Schizophrenia plasma autoantibodies (Pt 1, 2) was incubated with a micromolar concentration of LY379268 alone (solid bars) or with (open bars) a two-hundred nanomolar concentration of the selective 5-HT2AR antagonist M100907 in N2A cells. Results are mean + SE as described in Materials and Methods.

LY379286-potentiated SCZ plasma IgG autoantibody-induced neurite retraction was significantly prevented by co-incubation with the highly selective 5HT2AR antagonist M100907 (500 nM) suggesting mGlu2/3R agonism enhanced neurite retraction via 5HT2AR-dependent signaling. One possibility is that an mGlu2/3R agonist increased 5HT2AR receptor affinity for SCZ plasma IgG autoantibodies via a heteromeric receptor-receptor interaction as was previously reported for mGlu2/3 agonist action at the 5HT2AR/mGlu2R complex [4].

Potentiation of 5HT2AR- mediated neurite retraction (by a 10 µM concentration of LY379268) was observed in the IgG autoantibodies from twelve patients tested including: chronic paranoid schizophrenia (n=5), Parkinson’s disease (n=5), dementia (n=1) and diabetic nephropathy (n=1). The mean level of neurite retraction (induced by an IgG concentration which alone caused~ 50% retraction) was significantly increased in the presence of 10 µM LY379268 (74 + 10% vs 52 + 7%; P < 0.001; n=12) (Table 4).

Table 4. Mean N2A neurite retraction induced by neurovascular pathologies’ AutoAB in the presence or absence of (7.5–10 µM) LY379268, a selective mGlu2/3R agonist

^ Mean acute N2A neurite retraction (after five minutes’ incubation) in the presence of the indicated mean concentration of plasma autoantibodies (protein-A eluate fraction) with or without a 7.5–10 µM concentration of the mGlu2/3R agonist LY379268.

Other pathologies: n=1 dementia, n=1 diabetic nephropathy, n=1 major depressive disorder

Balanced Gq/11- and Gi/o-coupled signaling in response to SCZ plasma IgG autoantibodies

Second generation anti-psychotic medications, e.g. clozaril, risperodone, exhibit high affinity binding to 5-HT2AR and bias 5-HT2AR signaling in favor of Gi- coupled pathways [11]. They also bind less avidly to the dopamine 2 receptor, D2R [11]. The reversible 5-HT2AR hallucinogenic agonist DOI biases signaling in favor of Gq/11 [12]. Yet the irreversible 5-HT2AR agonist LSD, and serotonin and its psychoactive metabolites promotes a mix of Gi- and Gq/11 coupled signaling [2, 12]. Psychosis can occur in subsets of major depressive disorder and in ~50% of Parkinson’s disease patients. We next investigated whether the ability to evoke survival promotion in N2A cells might differentiate autoantibodies in a subset of psychosis-prone patients and the mechanisms underlying autoantibody-induced N2A survival promotion.

Plasma IgG autoantibodies in Patients 1, 2, 4 having chronic schizophrenia (10 µg/mL) caused dose-dependent increased N2A cell survival compared to an identical concentration of IgG autoantibodies from four patients with either MDD (n=3) or PD (n=1), only one of whom (PD) had experienced visual hallucinations (Figure 4A). The autoantibodies from a patient with the systemic autoimmune condition discoid lupus erythematosus caused significantly greater N2A growth stimulation (144% + 5% vs. 102 + 8%; P < 0.01) compared to mean growth stimulation in an identical concentration of the autoantibodies from three chronic schizophrenia patients (Figure 4A).

A

B

Figure 4. A) Plasma autoantibodies in three patients with schizophrenia (solid line) , four patients with another disorder (dotted line), or a patient with diabetes and discoid lupus (dashed line) were incubated with N2A cells at the indicated concentrations for 24 hours. Each point represents mean N2A cell survival, i.e. quadruplicate determinations which varied by < 10%.

* P< 0.01 compared to solid line or dotted line B) Patient 1 diabetic schizophrenia autoantibodies (60 nM) was incubated alone (solid bars) or with a 200 nanomolar (speckled bar) or a 500 nanomolar concentration (open bar) of the selective 5-HT2AR antagonist M100907 in N2A cells for 24 hours. Results are mean + SE.

Mechanism of N2A pro-survival effect in plasma autoantibodies from psychosis-prone subset

Serotonin 2A receptor activation leads to β-arrestin-2- mediated desensitization and receptor internalization. However, β-arrestin-2 can also couple with diverse signaling pathways leading to enhanced cell proliferation and/or survival promotion. The Pt 1, chronic paranoid schizophrenia IgG autoantibodies promoted increased N2A cell proliferation which was significantly blocked by 500 nM concentration of M100907 (Figure 4B) consistent with 5HT2AR-mediated proliferation signaling. LY379268 (5 µM) alone had no effect on N2A survival, but it significantly blocked the Pt 1 and Pt 2, schizophrenia autoantibody-induced pro-survival effect on N2A cells (Figure 5A). Pertussis toxin (100 ng/mL) had no effect alone on N2A cell survival, but significantly blocked the pro-survival effect of Patient 2 SCZ IgG on N2A cells (Figure 5B). Finally, the PI3- kinase inhibitor LY294002 (20 µM) significantly decreased N2A survival-promotion induced by schizophrenia Pt 1, and Pt 3 major depression with psychotic features autoantibodies (Figure 6A, B).

A

B

Figure 5. A) Patient 1 schizophrenia plasma autoantibodies (40 nM) was incubated alone (solid bar) or with (speckled bar) a 5 micromolar concentration of the mGlu2R agonist LY379268 in N2A cells for 24 hours. Similar results were obtained in the plasma autoantibodies from Pt 2 chronic schizophrenia, and a Parkinson’s disease patient who had been experiencing visual hallucinations. B) Patient 2 chronic schizophrenia plasma autoantibodies (40 nM) was incubated alone (solid bar) or with (speckled bar) a 100 ng/mL concentration of pertussis toxin (PTX) in N2A cells for 24 hours. A-B) Results are mean + SE.

A

B

Figure 6. A) Pt 1 schizophrenia plasma autoantibodies (40 nM) or B) Pt 3 major depression with psychotic features autoantibodies was incubated alone (solid bar) or in the presence (speckled bar) of a twenty micromolar concentration of the PI3-kinase inhibitor LY294002 in N2A cells for 24 hours. Results are mean + SE. Similar results were obtained with two additional schizophrenia patient plasma autoantibodies.

Possible association between psychosis and autoantibody-mediated N2A survival promotion

Mean N2A survival promotion in the autoantibodies from seven patients suffering with hallucinations significantly exceeded mean N2A survival in an identical concentration of the plasma autoantibodies from four patients not suffering with hallucinations (112 + 14 vs 72 + 15; P = 0.001) (Figure 7). The patient subgroups (experiencing or not experiencing hallucinations) did not differ significantly in their baseline clinical characteristics (Table 5).

Figure 7. Mean N2A cell survival after 24 hours incubation in the presence of an identical 40 nM concentration of the plasma autoantibodies from schizophrenia (n=5), major depression (n=1), Parkinson’s disease (n=1) patients all suffering with hallucinations (n=7, solid bar) or from patients with major depression (n=3) or diabetes without hallucinations (n=4, open bar). Results are mean + SE.

Table 5. Baseline characteristics in patients suffering or not suffering hallucinations

^ Schizophrenia (n=5), Major depression (n=1), Parkinsons disease (n=1)

* Major depression (n=3), type 2 diabetes without neuropsychiatric disorder (n=1)

Discussion

Here we demonstrate (for the first time) the existence of 5-HT2AR- activating IgG autoantibodies in plasma from a subset of chronic schizophrenia patients. The autoantibodies caused potent, irreversible Gq/11-mediated neurite retraction in N2A cells by a mechanism involving activation of RhoA/ROCK and Gq11- coupled, PLC/IP3R signaling pathways. In patients suffering with severe, recurrent hallucinations and suicidal ideation, the autoantibodies also promoted N2A cell survival and/or proliferation (in part) via apparent recruitment of additional Gi-coupled, PI3-kinase-dependent survival signaling. LY379268 blocked 5HT2AR-dependent, Gi-coupled, N2A survival signaling in autoantibodies from patients suffering with hallucinations in agreement with the report of Gonzalez-Maeso et. al. [4] of similar effects of LY379268 on hallucinogen-specific induction of egr-2.

The 5-HT2A receptor mediates diverse signaling pathway activation in response to different ligands [12], a phenomenon which has been referred to as “functional selectivity” or “biased agonism.” Beta-arrestin-2- directed signaling not only plays a key role in receptor internalization and desensitization, but it acts as a scaffold to organize extracellular regulated-kinase MAPK signaling cascades [13] and components of the PI3-kinase/Src/Akt cell survival pathway [14]. The latter pathway is active in cancer cells [15] including neuroblastoma [16]. Lysergic acid diethylamine (LSD) causes long-lasting 5-HT2AR activation and biased, β-arrestin-2-directed signaling [2] providing a further link between a ‘mix’ of Gq/11-coupled and β-arrestin-2- directed signaling and psychomimetic effects associated with a subset of long-lasting 5-HT2AR agonists, i.e. schizophrenia autoantibodies.

Sensitization of autoantibody-induced neurite retraction by the mGlu2/3R agonist LY379268 is consistent with Gonzalez-Maeso et. al. [4] and suggests positive allosteric modulation of the 5HT2AR protomer affinity at functional 5HT2AR/mGlu2R complexes in N2A neuroblastoma cells as a plausible mechanism. Receptor heteromers have been reported in neuroblastoma cell, e.g. A2R/D2R [17], and both 5HT2AR [9] and mGlu2R [18] are normally expressed in mouse neuroblastoma cells. Yet in our preliminary experiments, the dopamine 2 receptor (D2R) agonist quinpirole (10 µM) did not sensitize N2A cells to schizophrenia plasma autoantibody- (n=4 different patients) or DOI-induced neurite retraction suggesting that targeting of 5-HT2AR by schizophrenia plasma autoantibodies is specific for homomers or 5-HT2AR/mGlu2R heteromers.

Targeting mGlu2R agonism as a mechanism to suppress presynaptic glutamate release (via mGlu2 autoreceptors) reduced certain psychomimetic drug effects in animals, but did not lead to overall reduction in both negative and positive symptoms in schizophrenia patients [19]. Sensitization of 5-HT2AR-mediated neurite withdrawal by an mGlu2/3 agonist demonstrated here (in vitro) suggests a possible mechanism in which mGlu2R agonism may promote a ‘negative symptom’ in schizophrenia. For example, major depression autoantibodies from patients suffering with anhedonia, a negative symptom in schizophrenia, not only caused robust Gq/11-mediated neurite retraction in N2A neuroblastoma cells [9], but also led to decreased sucrose preference, the behavioral equivalent of anhedonia, in mice following autoantibody intracerebroventicular infusion [20]. More study is needed to determine whether Gq/11-mediated neurite retraction might a useful biomarker in drug discovery aimed at reducing combined negative and positive symptoms in schizophrenia.

Excessive glutamatergic signaling in the cortex and striatum is a feature of 5-HT2AR activation which may underly (in part) delusions and hallucinations occurring in schizophrenia [21] and in a subset of Parkinson’s disease. Serotonin 2A receptor agonism promotes neuronal glutamate release [22] which in turn may enhance 5-HT2AR protomer affinity at 5HT2AR/mGlu2R heteromers in the prefrontal cortex. Since constitutive activation of Gq11-coupled GPCR signaling was reported to recruit additional Gi-coupled downstream signaling in independent GPCRs [23], we cannot rule out the possibility that some ‘cross-talk’ between 5HT2AR and mGlu2R agonism may occur (in part) via mechanisms occurring downstream of direct receptor-receptor interaction(s).

Autoantibodies in a subset of psychosis-prone individuals might possess structural characteristics which bias 5-HT2AR signaling in favor of dual Gq/11 and Gi-coupled, or β-arrestin-2-directed pathways. One possibility is circulating immune complexes which were reported to increase in patients with schizophrenia [24]. Systemic autoimmunity, characterized by a high prevalence of immune complexes, is a risk factor for schizophrenia identified in prior studies [6]. Circulating immune complexes in chronic lymphocytic leukemia (CLL) promoted increased B cell survival through activation of anti-apoptotic and pro-survival signaling pathways [25] and in a prior report, the autoantibodies in a Parkinson’s disease patient with CLL strongly promoted increased N2A proliferation [10]. Taken together, autoantibody-induced cancer cell proliferation and survival might suggest immune complexes which appear to promote biased signaling in favor of dual Gq/11-mediated and β-arrestin-2-directed pathway activation.

A limitation of our study is that it is cross-sectional. More study is needed to determine whether autoantibodies are present in acute-onset, or drug-naive schizophrenia or in other psychotic disorders. Long-standing type 2 diabetes mellitus was associated with 5-HT2AR-activating autoantibodies in patients having diffuse microvascular injury and chronic inflammation, e.g. diabetic kidney disease [9,10]. Yet type 2 diabetes per se is unlikely to have accounted for autoantibodies in the present subset of schizophrenia patients who were free of significant microvascular (i.e. renal or retinal) complications. Schizophrenia is thought to arise through complex gene-environment interactions leading to abnormal neurodevelopment. Brain-reactive antibodies occurring as a result of inflammation, infection or systemic autoimmunity is one potential environmental mechanism in schizophrenia causation [6].

In summary, chronic schizophrenia plasma IgG autoantibodies appeared to activate a Gq11/phospholipase C/inositol triphosphate receptor pathway and RhoA/Rho kinase signaling to cause acute N2A neurite retraction. Autoantibody-induced neurite retraction was positively modulated by the mGlu2/3R agonist LY379268 consistent with reported mGlu2R-driven, positive modulation of 5-HT2AR protomer affinity at 5HT2AR/mGlu2R heteromers [4] found in cortical brain regions underlying perception. Biased agonism at 5-HT2AR mediated by the hallucinogen LSD [2] or by certain autoantibodies from patients suffering delusions and halllucinations involves recruitment of Gi-coupled, PI3-kinase-dependent mechanisms (which for the autoantibodies) was associated with enhanced survival in N2A neuroblastoma cells.

Acknowledgement

We thank Dr. S. Varia and Dr. J. Alder (Dept of Neuroscience, Rutgers-RWJ Med School) for generously providing N2A mouse neuroblastoma cells. Supported in part by a grant from the Veterans Biomedical Research Institute (East Orange, NJ) to MBZ.

The authors report no multiplicity of interest affecting the objectivity of the study results. Presented (in part) at the 2019 Experimental Biology Meeting, Orlando, Florida Supported in part by a grant from the Veterans Biomedical Research Institute, East Orange, New Jersey to MBZ.

References

1. Xu T, Pandey SC (2000) Cellular localization of serotonin(2A) (5HT(2A)) receptors in the rat brain. Brain Res Bull 51: 499–505. [crossref]
2. Wacker D, Wang S, McCorvy JD (2017) Crystal Structure of an LSD-Bound Human Serotonin Receptor. Cell 168: 377–389
3. Moreno JL, Holloway T, Albizu L, Sealfon SC, Gonzalez-Maeso J (2011) Metabotropic glutamate mGlu2 receptor is necessary for the pharmacological and behavioral effects induced by hallucinogenic 5-HT2A receptor agonists. Neurosci Lett 493: 76–79.
4. González-Maeso J, Ang RL, Yuen T, Chan P, Weisstaub NV, et al. (2008) Sealfon SC Identification of a serotonin/glutamate receptor complex implicated in psychosis. Nature 452: 93–97.
5. Huang A, Amos TB, Joshi K, Wang L, Nash A (2018) Understanding healthcare burden and treatment patterns among young adults with schizophrenia. J Med Econ 21: 1026–1035.
6. Benros ME, Mortensen PB, Eaton WW (2012) Autoimmune diseases and infections as risk factors for schizophrenia. Ann N Y Acad Sci 1262: 56–66. [crossref]
7. Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014) Biological insights from 108 schizophrenia-associated genetic loci. Nature 511: 421–427.
8. Zimering MB, Mirkovic N, Pandya M, Zimering JH, Behnke JA, et al. (2016) Toxic Immunoglobulin Light Chain Autoantibodies are Associated with a Cluster of Severe Complications in Older Adult Type 2 Diabetes. J Endocrinol Diab 3: 1–14.
9. Zimering MB (2017) Diabetes Autoantibodies Mediate Neural- and Endothelial Cell- Inhibitory Effects Via 5-Hydroxytryptamine- 2 Receptor Coupled to Phospholipase C/Inositol Triphosphate/Ca2+ Pathway. J Endocrinol Diab 4: 1–10.
10. Zimering MB (2018) Circulating Neurotoxic 5-HT2A Receptor Agonist Autoantibodies in Adult Type 2 Diabetes with Parkinson’s Disease. J Endocrinol Diabetes 5: 10.
11. Weiner DM, Burstein ES, Nash N, Croston GE, Currier EA, et al. (2001) 5-hydroxytryptamine2A receptor inverse agonists as antipsychotics. J Pharmacol Exp Ther 299: 268–276. [crossref]
12. Schmid CL, Raehal KM, Bohn LM (2008) Agonist-directed signaling of the serotonin 2A receptor depends on beta-arrestin-2 interactions in vivo. Proc Natl Acad Sci U S A 105: 1079–1084. [crossref]
13. Luttrell LM, Roudabush FL, Choy EW (2001) Activation and targeting of extracellular signal-regulated kinases by beta-arrestin scaffolds. Proc Natl Acad Sci U S A 98: 2449–2454.
14. Schmid CL, Bohn LM (2010) Serotonin, but not N-methyltryptamines, activates the serotonin 2A receptor via a ß-arrestin2/Src/Akt signaling complex in vivo. J Neurosci 30: 13513–13524.
15. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB (2005) Exploiting the PI3K/AKT pathway for cancer drug discovery. Nat Rev Drug Discov 4: 988–1004. [crossref]
16. Johnsen JI, Segerström L, Orrego A, Elfman L, Henriksson M, et al. (2008) Inhibitors of mammalian target of rapamycin downregulate MYCN protein expression and inhibit neuroblastoma growth in vitro and in vivo. Oncogene 27: 2910–2922. [crossref]
17. Johnsen JI, Segerström L, Orrego A, Elfman L, Henriksson M, et al. (2008) Inhibitors of mammalian target of rapamycin downregulate MYCN protein expression and inhibit neuroblastoma growth in vitro and in vivo. Oncogene 27: 2910–2922. [crossref]
18. Fuxe K, Agnati LF, Jacobsen K (2003) Receptor heteromerization in adenosine A2A receptor signaling: relevance for striatal function and Parkinson’s disease. Neurology 61: 19–23.
19. Wang J, Wang Z, Liu R (2018) Metabotropic glutamate receptor subtype 2 is a cellular receptor for rabies virus. PLoS Pathogens 14.
20. Muguruza C, Meana JJ, Callado LF (2016) Group II Metabotropic Glutamate Receptors as Targets for Novel Antipsychotic Drugs. Front Pharmacol 7: 130.
21. Zimering MB, Behnke JA, Thakker-Varia S, Alder J (2015) Autoantibodies in human diabetic depression inhibit adult neural progenitor cells in vitro and induce depressive-like behavior in rodents. J Endocrinol Diab 2: 11.
22. Aghajanian GK, Marek GJ (2000) Serotonin model of schizophrenia: emerging role of glutamate mechanisms. Brain Res Brain Res Rev 31: 302–312.
23. Ansah TA, Ferguson MC, Nayyar T (2011) The 5-HT2A receptor antagonist M100907 produces antiparkinsonian effects and decreases striatal glutamate Front Systems. Neurosci 2011.
24. Baker RA, Casarosa P, Timmerman H (2004) Constitutively active Gq/11-coupled Receptors Enable Signaling by Co-expressed Gi/o-coupled Receptors. J Biol Chem 279: 5152–5161.
25. Mailian KR, Boiadzhian AS, Sogoian AF, Sim RB, Manukian LA (2005) Concentration and protein composition of circulating immune complexes in the blood of patients with schizophrenia and subjects with positive familial history of disease. Zh. Nevrol. Psikhiatr. Im. S. S. Korsakova 105: 55–60
26. Balakrishnan K, Peluso M, Fu M (2015) The phosphoinositide-3-kinase (PI3K)-delta and gamma inhibitor, IPI-145 (Duvelisib), overcomes signals from the PI3K/AKT/S6 pathway and promotes apoptosis in CLL. Leukemia 29: 1811–1822.

DOI: 10.31038/EDMJ.2019343

Abstract

Di-2-Ethylhexyl Phthalate (DEHP) is the widely used to convey flexibility and transparency to plastic products made of polyvinyl chloride and also in the manufacture of medical devices. DEHP disrupts reproductive tract development in an antiandrogenic manner and also may induce neurobehavioral changes. In previous works, we demonstrated that chronic postnatal exposure to DEHP alters the neuroendocrine regulation of the testicular axis, modifying the hypothalamic concentration of excitatory neurotransmitters and therefore induces an anxiogenic effect. In the Elevated Plus Maze (EPM) test, dizocilpine (MK-801) induces a decrease in anxiety-related behaviors throughout NMDA receptor blockade. The objective of this work was to investigate whether the blockade of NMDA receptors of glutamate by the non-competitive antagonist MK-801 could modify the anxiety-like behavior induced by chronic postnatal exposure to DEHP (30 mg/kg body weight/day, orally from birth) in young adult male rats in the EPM test. The results show that NMDA receptor blockade by MK-801 (0.1 mg/kg body weight, i.p.) in DEHP exposed animals is able to produce a significant decrease in time spent in closed arms (TSC) and in Freezing Time (FT) as well as an increase in time spent in open arms (TSO) in the EPM test, indicating an anxiolytic effect. In conclusion, our results suggest: 1) NMDA receptor blockade by MK-801 can reverse anxiety-like behavior induced by exposure to DEHP during the early period of life. 2) The glutamatergic system is involved in the anxiogenic effect of phthalate, which is probably triggered by its known antiandrogenic action.

Keywords

DEHP; Endocrine Disruptor; Dizocilpine (MK-801); Glutamate Receptors; Anxiety like-behavior

Introduction

N-Methyl-D-Aspartate (NMDA) receptor-mediated glutamate transmission is one of the most significant mechanisms during multiple stages of brain development and has been implicated in cognitive functions and emotional responses [1]. It has been suggested that during the early neurodevelopmental period the treatment with non-competitive antagonist of NMDA receptors such as dizocilpine (MK-801), can cause long-term effects in the anatomical, neurochemical, neurophysiological and behavioral features of rodents [2,3]. Indeed, there are evidences that MK-801 has an anxiolytic potential [4–6]. During the critical period of brain maturation, MK-801 induces locomotor hyperactivity and decreases anxiety levels in adolescent rats.⁶ In the elevated plus maze (EPM) test a decrease was observed in anxiety-related behaviors caused in adult male mice by MK-801 in the early developmental period [7]. NMDA receptor blockade could reduce neuronal activity in pathways which lead to the release of GABA or monoamines, and which have themselves been implicated in anxiety-related processes [8].

Di-2(ethyl-hexyl phthalate) (DEHP) is an endocrine disruptor with antiandrogenic action, which is used as a plasticizer in many products, especially in medical devices and in manufacturing a wide variety of consumer products made with Polyvinyl Chloride (PVC) [9]. The brain has been determined to be at risk of DEHP exposure. Gestational and postnatal exposure to DEHP can affect neurodevelopment and lead to anomalies by disrupting normal brain development and function [10]. In previous works, we demonstrated that chronic postnatal exposure to DEHP alters the neuroendocrine regulation of the testicular axis, modifying the hypothalamic concentration of excitatory and inhibitory neurotransmitters [11,12]. DEHP could act on anxiety in different ways depending on the treatment, age, and sex of the animals [13–16]. It has been reported that postnatal exposure to DEHP can induce anxiogenic effect in pre and peripubertal male rats but not in females in the same stages of sexual maturation. Moreover, it has been proposed that the decrease in testosterone levels induced by postnatal exposure to DEHP could be one possible mechanism underlying DEHP anxiogenic-like behavior in immature male rats [17]. Recently, we have demonstrated that GABA agonists, muscimol and baclofen, can reverse DEHP neuroendocrine effects as well as its anxiogenic action in young adult male rats, supporting the notion that GABAergic system may be one of the neurotransmitter systems involved in the effects produced by DEHP exposure during early periods of neurodevelopment [18].

On these bases we hypothesized those DEHP-induced modifications in the concentration of excitatory amino acids in the brain of immature rats [11,12] could also have a modulatory role in the anxiogenic behavior induced by exposure to this endocrine disruptor during early periods of neurodevelopment. Therefore, the objective of this study was to investigate whether the blockade of NMDA receptors of glutamate by the non-competitive antagonist MK-801 could modify the anxiety-like behavior induced by chronic postnatal exposure to DEHP in young adult male rats in the elevated plus maze (EPM) test.

Materials and Method

All animal procedures were performed following the protocols of the National Institute of Health—Guide for the Care and Use of Laboratory Animals. The approval to conduct the study was granted by the Animal Care and Ethics Committee of the School of Medicine, Universidad de Buenos Aires (UBA; CICUAL).

Animals

Wistar rats used were provided by the Department of Physiology, School of Medicine, UBA, Argentina. Animals were kept under a controlled environment (temperature 22–24C; lights on from 7.00 h to 19.00 h) and they had free access to food and filtered water, until the time of killing. All animals were fed with balanced food for laboratory rodents (Cooperation, ACA-16014007, Asociación de Cooperativas Argentinas-División de Nutrición Animal). The diet contained 15% of soy, but as the food used and the quantity of food intake by control and DEHP-treated groups were similar, we assumed that all animals were exposed to equivalent levels of food-borne phytoestrogens. Moreover, the same lots of diet were provided to animals from all groups at the same time, during the course of the study, to control across groups for possible variation in the diet content. We used ultrapure-filtered water (obtained from EDS-Pack, Millipore Merck, installed in the Milli-Q water system) that was presumed to be free of phthalates and other Endocrine Disruptors (EDs). To minimize additional exposures to substances that may act as EDs, rats were housed in stainless steel cages with wood beddings and water was supplied in glass bottles.

Drugs and Doses

DEHP was purchased (1 g/ml>99% pure; Aldrich Chemical Company, Inc., Milwaukee, Wisconsin, USA) and the final solution administered to the animals was made up fresh daily by adding 200 ml of DEHP to 1 l of filtered water to reach a concentration of 0.2 mg/ml and sonicating for 30 min to ensure a permanent and homogenized solution. An oral route in DEHP administration was chosen intending to mimic best the most common route of human exposure to the ED. The estimated average DEHP dose of exposure was 30 mg/kg body weight (BW)/day, based on the daily intake of DEHP solution and the weight of the related animal. Liquid consumption was measured calculating the difference between the amount of liquid placed in the bottle every day and the remaining amount on the following day to assess the intake. It was assumed that all DEHP solution missing in the bottle had been consumed by the animals. Assessments did not contemplate possible leakage or evaporation of the solution or potential loss of DEHP activity during the 24-h period. No significant differences between the amount of liquid consumed by dams and pups receiving DEHP and those which did not receive this chemical were found. The dose was chosen based on prior studies published by us, in which we demonstrated alterations in the reproductive axis and in the behavior in immature male rats exposed to DEHP at a dose of 30 mg/kg BW/day but not in animals exposed to a lower dose (3 mg/kg BW/day) [11,12,17,18].

On the other hand, MK- 801 ([(+)-3 (2-carboxypiperazine-4-yl) propyl-1-phosphonic acid]; Research Biochemicals Inc., Natick, Mass, USA) was dissolved in saline and injected at dose of 0.1 mg/kg BW, i.p., 1 h before performing the behavioral test. MK-801 can be administered intraperitoneally because it crosses easily the blood-brain barrier. The chosen dose does not produce toxic effects [19].

Experimental Design

Pregnant dams were placed individually in metallic cages and upon delivery, pups were sexed; male pups were separated and distributed with one surrogate dam (n= eight male pups per dam). On postnatal day (PND) 1, surrogate dams with their male pups were randomly assigned into control (C) and DEHP exposure groups. Dam’s exposure to DEHP began on PND 1 and continued until weaning. On PND 21, pups of each group continued receiving the same treatment until PND 60. On this day, control and DEHP-exposed male pups were randomly assigned (n= 8 animals per group) to the following treatments: (1) W + S: controls that were given water and injected with saline; (2) DEHP + S: animals that were given DEHP and injected with saline; (3) W+ MK-801: animals that were given water and injected with MK-801; (4) DEHP + MK-801: animals that were given DEHP and injected with MK-801. After receiving the MK-801 or the vehicle, animals were submitted to the behavioral test.

EPM test

The elevated plus maze (EPM) test is a widely used behavioral assay for rodents, and it has been validated to assess the antianxiety effects of pharmacological agents and steroid hormones and to define mechanisms underlying anxiety-related behavior [20]. The EPM apparatus consists of two open arms (10 x 50 cm), alternating in right angles with two closed arms (10 x 50 x 10 cm), delimiting a central area. The maze was elevated 50 cm above the floor. Before starting the test, animals were individually placed in a rectangular plastic glass area (40 × 40 cm) for 5 min in order to habituate them to the test environment. After that, the rats were placed in the central area of the maze, facing one of the closed arms, and were allowed to explore it for 5 min. The maze’s arms were equally illuminated so that the animals did not perceive lighting differences. Behavioral tests were performed from 12: 00 h to 14: 00 h. We used 10% ethanol to clean each arm of the maze and to remove olfactory cues every time between trials. Each rat was tested only once. The animal’s behavior was videotaped and the number of entries and the time spent in both, open and closed arms, were measured by an observer. The parameters measured were: Time Spent in Open (TSO) and Closed (TSC) arms, Total number of Entries (TE) and Time of Freezing (FT). These parameters were measured following a four-paw criterion; entry into the arm of the EPM was defined as the animal placing all four paws in that particular part of the maze. It is considered that anxiety-like behavior is characterized by a decrease in TSO and an increase in TSC. On the other hand, TE provides a built-in control measure for general hyperactivity or sedation. TSO, TSC and FT were expressed in seconds.

Statistical Analysis

Results are presented as means ± SEM. The statistical analysis was based on individual offspring numbers. The surrogate dam was not used as the unit. Therefore, we did not consider the potential effects of the surrogate dam in the statistical analysis, as pups were exposed to this chemical through her milk. All data were checked for normality by the Kolmogorov–Smirnov test and then, they were analyzed using Kruskal-Wallis test (non-parametric ANOVA); posthoc Mann Whitney (Bonferroni correction) to compare unpaired groups. The statistical software used was GraphPad InStat 3 and a difference was considered statistically significant when p < 0.05.

Results

Figure 1 shows the effect of MK-801 on TSO in adult male rats exposed to DEHP from birth. Significant differences between treatments were found (K ₍₃₎ =25.860, p < 0.0001). When compared with the control group, a significant decrease in TSO were observed in rats exposed to DEHP (DEHP+S vs W+S, p < 0.01) while MK-801 treatment increased this parameter (W+MK vs W+S, p <0.001). MK-801 also produced an increase in TSO with respect to DEHP-exposed group (W+MK vs DHEP+S, p < 0.001) and was able to reverse the effect of the endocrine disruptor (DEHP+MK vs DEHP+S, p<0.001). No significant differences were found between W+MK and DEHP+MK groups, p=0.7984).

Figure 1. Effect of acute administration of MK-801 on the time spent in open arms in adult male rats exposed postnatally to DEHP. W: water; S: saline; DEHP: di-2(ethyl-hexyl phthalate; MK: MK-801. ** p<0.01 vs W+S; *** p<0.001 vs W+S; ### p<0.001 vs DEHP+S. Each value represents the mean ± SEM of eight animals per group.

As can be seen in Figure 2, there were significant differences in TSC between treatments (K ₍₃₎ = 14.809, p<0.01). DEHP significantly increased the TSC compared to the control group (DEHP+S vs W+S, p < 0.05) while MK-801 produced a significant decrease (W+MK vs W+S, p < 0.05) and reversed the effect of DEHP (DEHP+MK vs DEHP+S, p<0.01).

Figure 2. Effect of acute administration of MK-801 on the time spent in closed arms in adult male rats exposed postnatally to DEHP. W: water; S: saline; DEHP: di-2(ethyl-hexyl phthalate; MK: MK-801. * p<0.05 vs W+S; ## p<0.01 vs DEHP+S; ### p<0.001 vs DEHP+S. Each value represents the mean ± SEM of eight animals per group.

Figure 3 shows the number of total entries in both arms of the maze. Significant differences were detected between treatments (K ₍₃₎ = 23.642, p<0.001). As expected, there were no significant differences in DEHP-treated animals with respect to the control group (DEHP+S vs W+S, p =0.2627). MK-801 produced a significant increase in TE (W+MK vs W+S, p < 0.001). The same effect was observed with DEHP + MK (DEHP+MK vs DEHP+S, p<0.001). No differences were found between both groups treated with MK-801 (DEHP+MK vs W+MK, p=0.2475).

Figure 3. Effect of acute administration of MK-801 on the number of total entries in adult male rats exposed postnatally to DEHP. W: water; S: saline; DEHP: di-2(ethyl-hexyl phthalate; MK: MK-801. *** p<0.001 vs W+S; ### p<0.001 vs DEHP+S. Each value represents the mean ± SEM of eight animals per group.

Time spent in freezing behavior (Figure 4) showed significant differences between groups (K₍₃₎ = 26.042, p<0.001). There was a significant increment in DEHP-exposed rats (DEHP+S vs W+S, p<0.001) and a significant decrease in animals injected with MK-801 (W+MK vs W+S, p<0.001). In addition, MK-801 antagonized the effect of DEHP ((DEHP+MK vs DEHP+S, p<0.001).

Figure 4. Effect of acute administration of MK-801 on freezing time in adult male rats exposed postnatally to DEHP. W: water; S: saline; DEHP: di-2(ethyl-hexyl phthalate; MK: MK-801. *** p<0.001 vs W+S; ### p<0.001 vs DEHP+S. Each value represents the mean ± SEM of eight animals per group.

Discussion

In order to investigate if the NMDA receptors of glutamate are involved in the changes in the anxiety-like behavior induced by early exposure to endocrine disruptor DEHP, we compared the performance in the Elevated plus Maze of adult male rats exposed to DEHP postnatally and treated or not with MK-801, a non-competitive antagonist of glutamate receptors. Our results showed that NMDA receptor blockade by MK-801 was able to reverse anxiety-like behavior induced by exposure to DEHP during the early period of life.

As we expected, our results showed an anxiogenic effect of DEHP, which was evidenced by a significant increase in TSC and a decrease in TSO in the EPM test. These results were consistent with the findings reported in immature rats and mice exposed perinatally to DEHP (10, 30, 50, and 200 mg/kg) [17,21] as well as in young adult male rats chronically exposed to DEHP (30 mg/kg) from birth [18].

However, the mechanism underlying the enhanced anxiety-like behavior induced by DEHP is not clear. In one way DEHP may influence anxiety-like behavior by disturbing the hypothalamic-pituitary-adrenal axis (HPA). In fact, exposure of male rats to DEHP leads to an age-dependent activation of the HPA ‘in vivo’ and of adrenocortical steroidogenesis ‘ex vivo’, showing a specific susceptibility of immature animals to this phthalate [22]. Perturbations of hormonal homeostasis of the HPA axis were found in adolescent female rats exposed to DEHP during lactation [23]. Direct and transgenerational effects of perinatal DEHP exposure on social behaviors and anxiety-like behavior in mice, as well as transgenerational effect of DEHP on corticosterone levels were reported by other authors [24,25]. Another way in which DEHP could influence anxiety is by altering the hypothalamic-pituitary-gonadal axis. It has been reported that gestational and lactational DEHP exposure is able to disrupt the neuroendocrine control of the gonadal axis during sexual maturation in rats [11,12]. The changes induced by DEHP in the hypothalamic concentration of GABA, the inhibitory neurotransmitter involved in the neuroendocrine control of the gonadal axis as well as in the development of anxiety, support the idea that the GABAergic system could participate in the behavioral effects of the DEHP [18]. Moreover, in the adulthood perinatal DEHP-exposed males displayed anxiogenic status, which correlate with a decrease in testosterone levels [7]. Given that the treatment with testosterone was able to reverse the disruptive effect of chronic postnatal exposure to DEHP on the testicular axis of rats, it was suggested that the antiandrogenic action of this chemical could be a probable mechanism underlying its anxiogenic effect [17]. In addition, Xu et al [21]. have suggested that the inhibition of the mitogen-activated protein kinase/extracellular signal-regulated kinases (MAPK/ERKs) activity, which mediates the non-genomic actions of testosterone in the brain cells, as well as the down-regulation of androgen receptors expression in hippocampus could be involved in the anxiogenic-like effect produced by perinatal DEHP exposure during puberty and even after in adult life.

In agreement with other authors [17,18], in the animals exposed to DEHP we did not detect changes in TE in both the open and closed arms. Considering that TE provides an approximate measure for the locomotor activity, these results could indicate that the increase in the anxiety-like behavior induced by DEHP was not influenced by probable changes in locomotor activity. Furthermore, we observed a significant increase in freezing time in DEHP-exposed rats. It is known that freezing response is a behavioral index of fear and that the deficit in testosterone may have a significant effect of increasing freezing response [26]. On this basis, we hypothesized that the increase in freezing time induced by DEHP could be due to the decrease in circulating blood testosterone levels previously found in immature and adult male rats exposed postnatally to this endocrine disruptor at the same dose used in the present work [17,18]. Therefore, it should be considered the possibility that the increase in freezing time by DEHP may be due to either direct or indirect alterations in the neuroendocrine gonadal and adrenal axes modulators, which play a critical role in the regulation of stress responses.

It is known that glutamate is the main excitatory neurotransmitter in the human Central Nervous System. A growing body of evidence suggests that glutamatergic neurotransmission may be involved in the biological mechanisms underlying stress response and anxiety-related disorders [27]. Glutamate mediates its effects via stimulation of ionotropic and metabotropic receptors [28]. During development in the rat the brain is highly sensitive to the effects of the glutamate ionotropic receptor NMDA modulation [29]. The non-competitive antagonists of glutamate receptors have been extensively studied for their anxiolytic action. In particular, in rodent models it has shown that MK-801, a non-competitive antagonist of NMDA receptors, may have anxiolytic like effects on the EPM test [30].

In the present work, we observed that MK-801-treated rats showed lower anxiety level in the EPM and increased spontaneous locomotor activity in comparison with the control group. In fact, consistent with the reported anxiolytic potential of MK-801[5,6,29] we observed that this NMDA receptor antagonist was able to produce a significant decrease in TSC as well as an important increase in TSO, suggesting that NMDA glutamate receptors are involved in the development of anxiety-like behavior. In addition, MK-801 treated animals showed a significant increase in TE, indicating that the locomotor activity can be induced by MK-801 [23].

When young adult male rats exposed to DEHP from birth were treated with the NMDA receptor antagonist, they showed an increase in TSO which was correlated with a decrease in TSC. The number of total entries in both arms of the labyrinth was increased, showing greater locomotor activity compared to animals treated only with DEHP. Freezing response in DEHP-exposed animals and treated with MK-801 was significantly lower than in the control and DEHP groups and showed no difference with the animals treated with MK-801 alone. These findings indicated that MK-801 was able to reverse the anxiogenic action of DEHP, leading the freezing time to values even lower than those observed in the control group.

However, it is difficult to explain the probable mechanisms that could mediate the anxiogenic effect of DEHP through an interaction with NMDA glutamate receptors. It has been reported that blockade of NMDA receptors causes a reduction of neuronal activity in pathways that lead to the release of GABA or monoamines, which themselves have been implicated in anxiety-related processes [8]. In this sense, we demonstrated the participation of the GABAergic system in the anxiety-like behavior induced by chronic postnatal exposure to DEHP in young adult male rats [17]. On the other hand, androgens modulate the structure and functions of the hippocampus related with behavior and exert profound effects on the up-regulation of NMDA receptors in hippocampus of males. DEHP markedly down-regulates the expression of androgen receptors which, in turn, are involved in the regulation of behavior mediated by the hippocampus of pubertal males. Therefore, it may be possible that chronic exposure to DEHP during critical periods of neurodevelopment is capable of inducing anxiogenic behavior in adult male rats, decreasing androgen levels and the expression of their receptors as well as interacting with NMDA glutamate receptors.

In conclusion, our results suggest: 1) NMDA receptor blockade by MK-801 can reverse anxiety-like behavior induced by exposure to DEHP during the early period of life. 2) The glutamatergic system is involved in the anxiogenic effect of phthalate, which is probably triggered by its known antiandrogenic action.

Highlights

• Chronic postnatal exposure to DHEP induces anxiogenic effect in adult male rats
• MK-801 reverses anxiety-like behavior induced by chronic postnatal exposure to DEHP
• The glutamatergic system is involved in the anxiogenic effect of phthalate

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Universidad Buenos Aires under grants UBACYT20020130100439BA (OJP) and UBACYT 20020170100151BA (RAC and SC).

References

1. Pryce C, Mohammed A, Feldon J (2002) Environmental manipulations in rodents and primates: insights into pharmacology, biochemistry and behaviour. Pharmacol Biochem Behav 73: 1–5.
2. Nemeth H, Varga H, Farkas T, et al. (2002) Long-term effects of neonatal MK-801 treatment on spatial learning and cortical plasticity in adult rats. Psychopharmacology 160: 1–8.
3. Latysheva NV and Rayevsky KS (2003) Chronic neonatal N-Methyl-D-Aspartate receptor blockade induces learning deficits and transient hypoactivity in young rats. Prog Neuro- Psychopharmacol Biol Psychiatry 27: 787–794.
4. Sharma AC and Kulkarni SK (1993) Evidence for benzodiazepine receptor interaction with MK-801 in anxiety related behaviour in rats. Indian J Exp Biol 31: 191–193.
5. Poleszak E, Wlaz P, Wróbel A, et al. (2008) NMDA/glutamate mechanism of magnesium induced anxiolytic like behavior in mice. Pharmacol Rep 60: 655–663.
6. Kocahan S, Akillioglu K, Binokay S, et al. (2013) the effects of N-Methyl-D-Aspartate receptor blockade during the early neurodevelopmental period on emotional behaviors and cognitive functions of adolescent Wistar rats. Neurochem Res 38: 989–96.
7. Pınar N, Akillioglu K, Sefil F, et al. (2015) Effect of clozapine on locomotor activity and anxiety-related behavior in the neonatal mice administered MK-801. Bosn J Basic Med Sci 15: 74–9.
8. Fraser CM, Fisher A, Cooke MJ, et al (1997) The involvement of adenosine receptors in the effect of dizocilpine on mice in the elevated plus-maze. Eur Neuropsychopharmacol 7: 267–273.
9. Rowdhwal SS and Chen J (2018) Toxic Effects of Di-2-ethylhexyl Phthalate: An Overview. Biomed Res Int Article ID 1750368, 10.
10. 10.Lin H, Yuan K, Li L, et al. (2015) In Utero Exposure to Diethylhexyl Phthalate Affects Rat Brain Development: A Behavioral and Genomic Approach. Int J Environ Res Public Health 12: 13696–710.
11. Carbone S, Szwarcfarb B, Ponzo O, et al. (2010) Impact of gestational and lactational phthalate exposure on hypothalamic content of amino acid neurotransmitters and FSH secretion in peripubertal male rats. Neurotoxicology 31: 747–751.
12. Carbone S, Samaniego YA, Cutrera R, et al. (2012) Different effects by sex on hypothalamic-pituitary axis of prepubertal offspring rats produced by in utero and lactational exposure to di-(2-ethylhexyl) phthalate (DEHP). Neurotoxicology 33: 78–84.
13. Liang T, Ouyang J, Yi L, et al. (2013) Behavioral changes of rats after short-term exposure to di-(2-ethyl hexyl) phthalate. J South Med Univ 33: 401–405.
14. Park S, Cheong JH, Cho SC, et al. (2015) Di-(2-Ethylhexyl) phthalate exposure is negatively correlated with trait anxiety in girls but not with trait anxiety in boys or anxiety-like behavior in male mice. J Child Neurol 30: 48–52.
15. Quinnies KM, Doyle TJ, Kim KH, et al. (2015) Transgenerational effects of di-(2-ethylhexyl) phthalate (DEHP) on stress hormones and behavior. Endocrinology 156: 3077–3083.
16. Wang R, Xu X, Weng H, et al. (2016) Effects of early pubertal exposure to di-(2-ethylhexyl) phthalate on social behavior of mice. Horm Behav 80: 117–124.
17. Carbone S, Ponzo OJ, Gobetto N, et al. (2013) Antiandrogenic effect of perinatal exposure to the endocrine disruptor di-(2-ethylhexyl) phthalate increases anxiety-like behavior in male rats during sexual maturation. Horm Behav 63: 692–69.
18. Carbone S, Ponzo OJ, Gobetto N, et al. (2018) Effect of di(2-ethylhexyl) phthalate on the neuroendocrine regulation of reproduction in adult male rats and its relationship to anxiogenic behavior: Participation of GABAergic system. Human and Experimental Toxicology 38: 25–35.
19. Wozniak DF, Olney JW, Kettinger L et al. (1990) Behavioral effects of MK-801 in the rat. Psychopharmacology (Berl) 101: 47–56.
20. Walfi AA and Frye ChA (2007) The use of the elevated plus maze as an assay of anxiety-related behavior in rodents. Nat Protoc 2: 323–328.
21. Xu X, Yang Y, Wang R, et al. (2015) Perinatal exposure to di-(2-ethylhexyl) phthalate affects anxiety- and depression-like behaviors in mice. Chemosphere 124: 22–31.
22. Supornsilchai V, Söder O, Svechnikov K (2007) Stimulation of the pituitary-adrenal axis and of adrenocortical steroidogenesis ex vivo by administration of di-2-ethylhexyl phthalate to prepubertal male rats. J Endocrinol 192: 33–9.
23. Wang DC, Chen TJ, Lin ML, et al. (2014) Exercise prevents the increased anxiety-like behavior in lactational di-(2-ethylhexyl) phthalate-exposed female rats in late adolescence by improving the regulation of hypothalamus-pituitary-adrenal axis. Horm Behav 66: 674–84.
24. Quinnies KM, Harris EP, Snyder RW, et al. (2017) Direct and transgenerational effects of low doses of perinatal di-(2-ethylhexyl) phthalate (DEHP) on social behaviors in mice. PLoS One 12: 0171977.
25. Quinnies KM, Doyle TJ, Kim KH, et al. (2015) Transgenerational Effects of Di-(2-Ethylhexyl) Phthalate (DEHP) on Stress Hormones and Behavior. Endocrinology 156: 3077–83.
26. King JA, De Oliveira WL, Patel N (2005) Deficits in testosterone facilitate enhanced fear response. Psychoneuroendocrinology 30: 333–40.
27. Riaza Bermudo-Soriano C, Perez-Rodriguez MM, Vaquero-Lorenzo C, et al. (2012) New perspectives in glutamate and anxiety. Pharmacol Biochem Behav 100: 752–74.
28. Kew JN, Kemp JA (2005) Ionotropic and metabotropic glutamate receptor structure and pharmacology. Psychopharmacology (Berl) 179: 4–29.
29. Haberny KA, Paule MG, Scallet AC, et al. (2002) Ontogeny of the N-methyl-D aspartate (NMDA) receptor system and susceptibility to neurotoxicity. Toxicol Sci 68: 9–17.
30. Kocahan S, Akillioglu K (2013) Effects of NMDA receptor blockade during the early development period on the retest performance of adult Wistar rats in the elevated plus maze. Neurochem Res 38: 1496–500.