Fructose-a double edged sword for cell death versus differentiation and long-term survival of NSC-34 motor neuron like cells in vitro

In various neurological and neurodegenerative diseases (ND), motor neurons (MN) of the spinal cord are affected leading to movement impairments. The ND, Amyotrophic Lateral Sclerosis (ALS), is caused due to MN degeneration. ALS aicts athletes and other major sports personalities, who generally consume fructose enriched sports drinks. Recently, we have reported that high fructose (F5%) impairs the metabolic activity in the NSC-34, MN cell line and reduces the healthspan of C. elegans. But how fructose impacts the MNs either in vitro or in vivo in the long term is not understood. Here we report, to our surprise, that high fructose (F5%) treatment of NSC-34 leads to differentiation of 1-2% of cells with progressive neurite extension. They could be maintained for 80 days in vitro with 5% CO 2 and O 2 at 18.8%. On the contrary, 5% fructose signicantly reduced cell viability by ~85% and inhibited cell proliferation by Day10. Nuclear staining displayed multiple nuclei in the cells indicative of cytokinesis inhibition which led to the lack of cell proliferation. Further, F5% signicantly increased ROS levels (^~34%), the potential cause for reduced viability. In addition, no induction of expression of the master oxidative stress response regulator, the transcription factor, nrf-2, or the downstream effector, sod1, was evident. Despite the adverse effects, in the absence of any, F5% is a potential strategy to maintain at least a small percentage of MNs for a long time, ~45 days in vitro, which also reinforces the Redox-Cell death versus cell survival conundrum.


Introduction
In recent decades, lifestyle and food intake has drastically changed. Though life expectancy has increased, associated chronic non-communicable diseases like metabolic syndrome, hypertension, diabetes mellitus and neurodegenerative diseases caused due to lifestyle modi cations are on the rise 1 .
While fructose occurs naturally in fruits in small quantities, its uncontrolled consumption as the inexpensive and highly addictive 2 re ned sugar/high fructose corn syrup, in large quantities found in confectioneries, baked/processed food and sports drinks have meteorically risen in the past two-three decades. Only recently, it has come to light that fructose plays a key role in a icting most of the world's population with obesity 3 , non-alcoholic fatty liver disease, 4,5 and the like.
Off -late, fructose is implicated to be involved in various neurological conditions, neuro-in ammation 6 and neurodegenerative diseases 7 . It has been known for some time now, that fructose can cross the blood-brain barrier in small quantities, and contribute to energy output 8 . Recent studies also suggest that in the event of a surplus of glucose in the brain, it is converted to fructose through the sorbitol pathway and metabolized 9 . In the Neurodegenerative disease (ND), Amyotrophic Lateral Sclerosis (ALS), characterized by motor neuron (MN) degeneration leading to progressive paralysis and eventually death, which is both sporadic and familial, the preponderance is higher in sports personnel 10 . They generally consume large quantities of sports drinks enriched with glucose and fructose. Also, when ALS is caused by C9orf72 repeats, fructose metabolism seems to be impaired in the astrocytes of these patients 11 . As in ALS, mobility or motor control is impaired in many NDs like Parkinson's disease, and deciphering the mechanism(s) in vivo is a challenge. NSC-34, a mouse spinal cord motor neuron cell-line derived through fusion with neuroblastoma [12][13][14][15] , is an in vitro model widely studied.
NSC-34 cells express the neurotransmitter acetylcholine synthesizing enzyme choline acetyltransferase (ChAT) 16 . Under optimal conditions, like the addition of human umbilical cord-derived stromal cells conditioned medium 17 , retinoic acid, 18 or vascular endothelial growth factor 19 , these cells differentiate to generate neurites as long as 0.6mm in vitro 17 . As primary spinal cord motor neuron cultures are challenging to establish and even harder to maintain in vitro for a long term of more than three weeks 20 , NSC-34 is a good cell line model to understand the general effect of fructose and identify the potential mechanisms that bring about the adverse effects. One of the downstream effectors of nrf-2, the enzyme, Cu/Zn Superoxide dismutase I (SOD1), which dismutates O 2 ▪− and converts it to H 2 O 2 in the cytoplasm and mitochondrial intermembrane space 28 , is also implicated in ALS 16 . Since fructose plays a major role in energy metabolism and the generation of ROS 29,30 , its effects on the oxidative stress response need to be better understood. Also, fructose effects on neuronal differentiation and long-term maintenance are not known.
Hence, in this report, we address the effects of fructose on NSC-34 cell proliferation, differentiation, neurite extension, maintenance for a long term, and the potential contribution of ROS in cell proliferation inhibition and/or induction of differentiation.

Materials And Methods
Cell Culture Neuroblastoma x mouse spinal cord motor neuron cell line (NSC-34) was maintained using standard protocols as reported earlier 7,16,17 in Dulbecco's modi ed Eagle medium (DMEM) with 4.5mg/ml glucose supplemented with 10% Fetal Bovine Serum (FBS) and an antibiotic -antimycotic solution of Streptomycin and Penicillin and Amphotericin B. The cells were maintained at 37 o C with 5% CO 2 and 18.8% O 2 in the Forma series II CO 2 incubator (Thermo Fischer Scienti c, Waltham, USA) and sub-cultured every 3-4 days. All the reagents were from Gibco -Life Technologies, Gaithersburg, MD, USA unless otherwise mentioned.

Fructose Treatment
Fructose at a nal concentration of 5% (277mM) was added to the cells seeded in T25 asks (2x10 5 cells) in DMEM complete medium and this was maintained at 37 o C with 5% CO 2 and 18.8% O 2 for the long-term assay.

Measurement of cell viability
Thiazolyl Blue Tetrazolium Bromide (MTT) was used to measure cell viability. 5x10 4 cells/well were seeded on 24 well plates and allowed to stabilize overnight. The next day, fructose at a nal concentration of 5% was added to the same high glucose (4.5 mg/ml) DMEM medium and incubated for various days. After removing the medium at the end of exposure time, the cells were treated with MTT for 3 hours. The medium was replaced with 100% DMSO (dimethyl sulphoxide), incubated for 30 min at 37 o C, then colorimetric measurement was done at 570nm. All reagents were from SRL Chemicals, Chennai, India. MTT assay was carried out thrice in duplicates.

Nuclear staining
Acridine orange (AO) is a nucleic acid dye that permeates the nuclear membrane and emits green uorescence when bound to dsDNA 31 . For this assay, 2x10 4 cells were seeded onto 24 well plates and treated with fructose 5% for 48 hours, followed by staining with AO (Sigma Aldrich, Bengaluru, India) at a nal concentration of 50µg/ml for 30 mins at 37 o C. After a couple of 1xPBS washes, the cells were subject to uorescence microscopy.
Measurement of cellular ROS levels ROS levels were measured by using 2′,7′-Dichloro uorescin diacetate (DCF-DA) (Sigma Aldrich, Bengaluru, India). A previously established protocol was used 32 . Roughly, 5x10 4 cells/well were seeded on 24 well plates and allowed to stabilize overnight. Cells were then treated with fructose (F5%) and incubated for 48 hours, following which the medium was changed to the subsequent DMEM medium with F5% containing 2′,7′-Dichloro uorescin diacetate (DCF-DA) (10µm) and incubated for one hour at 37 o C. The uorescence intensity was measured at excitation and emission wavelengths of 485 nm and 535 nm, respectively. DCF-DA background was subtracted, and the graph was plotted. The assay was done thrice in quadruplicates.

Microscopy
The cells were analyzed for nuclear staining in Nikon Eclipse-Ti uorescence microscope, using FITC lter. The nucleus was viewed with 10x and 40x objectives. The same microscope was utilized for viewing and capturing the phase-contrast images. The images were obtained with a CCD camera and Qimaging software.

Statistics
Mean and standard deviation were calculated, and signi cance was estimated by t-tests using GraphPad Prism ver. 8.2.1 (GraphPad Software, La Jolla, California, USA).

Induction of motor neuron differentiation and long-term survival
High fructose (5%) treatment, in the presence of 18.8% O 2 provided through oxygen in ow in addition to the standard cell culture conditions of 37 o C and 5% CO 2 , induced differentiation in 1-2% of the cells in the motor neuron cell line NSC-34 (Fig. 1A). Moreover, extensive neurite extension and branching was evident in the differentiated cells, with the neurites growing into long and thick processes as time progressed ( Fig.1) (3-4weeks DIV). Although fructose is reported to cause cell death 7,33 , high fructose exposure supported the survival of ~0.5% of cells as differentiated neurons till ~80 days ( Fig. 2A-F). This survival was evident only in the presence of 5% fructose with 18.8% O 2 supplementation.

Disruption of cell proliferation through suppression of cell division
Although high fructose treated cells showed visible division of the nucleus (Fig. 4) cytoplasmic division was absent which resulted in multinucleation. This was evident even in some of the differentiated motor neuron-like cells (Fig. 4A). Staining with Acridine orange on day 20 revealed two or more well-divided nuclei (Fig.4B). While normal cell division was imminent in the control cells (Fig. 5B.a) it was lacking in F5% treated cells. The cells seemed to be arrested before cytokinesis resulting in multinucleation ( Fig.   5B.b-d), and thereby inhibition of cell proliferation.

High fructose induces ROS generation
Fructose is known to increase ROS levels in various cell types 29,30,34 . To identify fructose mediated ROS generation and its impact on NSC-34 cells, rst, ROS levels were measured using the uorescent dye indicator, DCF-DA. F5% signi cantly increased DCF-DA uorescence intensity by 34% upon 48hrs exposure (Fig. 5A), showing a signi cant increase in ROS levels [***-P=0.0002]. The neuronal cell body exhibited extensive vacuolation (Fig. 5B) indicative of ROS mediated cellular stress and toxicity.

Absence of oxidative stress response
Generally, ROS level increase is accompanied by induction of oxidative stress response. As F5% increased ROS levels, an increase in the expression of the oxidative stress response genes was investigated. Nrf-2, the major, master inducer of expression of a multitude of oxidative stress response pathway genes 27 including sod-1, an enzyme responsible for conversion of superoxide into less toxic hydrogen peroxide, and oxygen 28 was determined through RT-PCR. Though both nrf-2 and sod1 were expressed upon F5% treatment, the expression level was the same as in control (Fig. 5C). This lack of induction of expression will result in an abysmal oxidative stress response to protect against increased ROS due to high fructose treatment.

Discussion
In the past few decades, diet and lifestyle have changed drastically. One of the major changes is the addition of fructose, like re ned sugar or high fructose corn syrup, being used widely and indiscriminately in processed/baked confectioneries. Fructose usage is more than that of glucose or sucrose, due to its ready availability. Reports of fructose's role in major diseases like diabetes mellitus 35 , obesity 3,36 , nonalcoholic fatty liver disease 4,5,30 , cancer 37 , and even in neurological/ neurodegenerative diseases including ALS 38 is unsettling. ALS is more prevalent among sports personnel. Energy beverages consumed by them as an immediate source of muscular energy are supplemented with fructose. But fructose's effect on motor neuron degeneration which brings about ALS in sportspeople is unclear. Hence, we set out to address fructose impact on the NSC-34 MN line in vitro. Further, oxygen uptake is quite different in physically active people (sports), with high VO 2max (maximal oxygen uptake) 39 . To closely mimic this, unlike normal mammalian cell culture, the oxygen level was maintained at 18.8%, with exclusive O 2 supply in addition to 5% CO 2 . In vivo O 2 level is below 6.5% in most of the tissues including brain and muscle 40 . But, that of the spinal cord is not known.
Earlier, we reported that similar to high glucose being toxic 41 , exposure of NSC-34 cells to 5% fructose for 48 hrs caused ~60% cell death 42 even without oxygen (18.8%) supplementation. Presently, F5% showed ~85% cell death on day 10 (Fig. 3). Unexpectedly, around 1-2% of the cells differentiated with long, branched neurites and could be maintained in vitro up to 80 days (Fig.1&2) only with O 2 supplementation. The caveat here is that routine CO 2 incubator oxygen level is supposed to be around 18.5 % (present in the air) even without O 2 supplementation 410 . High fructose treatment combined with oxygen supplementation provides a strategy for long-term maintenance of differentiated spinal cord motor neurons which is hard to achieve. A further detailed study will provide novel insights into the motor neuron function and abnormalities.
Earlier we reported that high fructose exposure for a short term (6-48hrs) disrupted mitochondrial function and metabolic activity 42 in NSC-34 cells and shortened the healthspan of C. elegans. Here, we noticed NSC-34 cells with multiple nuclei in the presence of F5% (Fig. 4), while well-divided cells are evident (Fig.4B) in the control. Somehow, high fructose inhibits only cytokinesis, thereby hindering cell proliferation leading to multinucleated cells. One possibility is the ine cient synthesis of the cellular components needed due to reduced or lack of ATP production 43 upon F5% treatment.
Generally, endogenously produced ROS is regulated to activate signaling cascades to induce a multitude of protective functions. But at higher doses, ROS induces cellular toxicity. Fructose is known for increasing ROS production in many cellular models: hepatocytes 30 , adipocytes 43 , myocytes 44 , and cardiomyocytes 45 . Similarly, here high fructose increased ROS levels by 35% (Fig. 5A). Strangely the oxidative stress response machinery through nrf-2 was not induced (Fig. 5C). Besides, multiple vacuoles were observed (Fig.6B) in some of the cells as seen with superoxide/hydrogen peroxide exposure 46  Overall, high fructose is detrimental, and caution is required against its unfettered intake as part of the diet. At the same time, it provides a strategy or direction, in combination with oxygen supplementation, to differentiate and maintain motor neurons for a long term in vitro which can open new avenues on multiple fronts given the severity and lack of treatment for several neurological, neurodegenerative and neuromuscular diseases.