**RNA-Binding Proteins HuB, HuC, and HuD are Distinctly Regulated in Dorsal Root Ganglia Neurons from STZ-Sensitive Compared to STZ-Resistant Diabetic Mice**

**Cosmin Cătălin Mustăciosu 1,**†**, Adela Banciu 2,**†**, Călin Mircea Rusu 1,3, Daniel Dumitru Banciu 2, Diana Savu 1, Mihai Radu 1,\* and Beatrice Mihaela Radu 3,4**


Received: 17 March 2019; Accepted: 19 April 2019; Published: 22 April 2019

**Abstract:** The neuron-specific *Elav*-like Hu RNA-binding proteins were described to play an important role in neuronal differentiation and plasticity by ensuring the post-transcriptional control of RNAs encoding for various proteins. Although *Elav*-like Hu proteins alterations were reported in diabetes or neuropathy, little is known about the regulation of neuron-specific *Elav*-like Hu RNA-binding proteins in sensory neurons of dorsal root ganglia (DRG) due to the diabetic condition. The goal of our study was to analyze the gene and protein expression of HuB, HuC, and HuD in DRG sensory neurons in diabetes. The diabetic condition was induced in CD-1 adult male mice with single-intraperitoneal injection of streptozotocin (STZ, 150 mg/kg), and 8-weeks (advanced diabetes) after induction was quantified the *Elav*-like proteins expression. Based on the glycemia values, we identified two types of responses to STZ, and mice were classified in STZ-resistant (diabetic resistant, glycemia < 260 mg/dL) and STZ-sensitive (diabetic, glycemia > 260 mg/dL). Body weight measurements indicated that 8-weeks after STZ-induction of diabetes, control mice have a higher increase in body weight compared to the diabetic and diabetic resistant mice. Moreover, after 8-weeks, diabetic mice (19.52 ± 3.52 s) have longer paw withdrawal latencies in the hot-plate test than diabetic resistant (11.36 ± 1.92 s) and control (11.03 ± 1.97 s) mice, that correlates with the installation of warm hypoalgesia due to the diabetic condition. Further on, we evidenced the decrease of *Elav*-like gene expression in DRG neurons of diabetic mice (*Elavl2*, 0.68 ± 0.05 fold; *Elavl3*, 0.65 ± 0.01 fold; *Elavl4*, 0.53 ± 0.07 fold) and diabetic resistant mice *(Ealvl2*, 0.56 ± 0.07 fold; *Elavl3*, 0.32 ± 0.09 fold) compared to control mice. Interestingly, *Elav*-like genes have a more accentuated downregulation in diabetic resistant than in diabetic mice, although hypoalgesia was evidenced only in diabetic mice. The *Elav*-like gene expression changes do not always correlate with the Hu protein expression changes. To detail, HuB is upregulated and HuD is downregulated in diabetic mice, while HuB, HuC, and HuD are downregulated in diabetic resistant mice compared to control mice. To resume, we demonstrated HuD downregulation and HuB upregulation in DRG sensory neurons induced by diabetes, which might be correlated with altered post-transcriptional control of RNAs involved in the regulation of thermal hypoalgesia condition caused by the advanced diabetic neuropathy.

**Keywords:** *Elav*-like; Hu proteins; diabetes; streptozotocin; thermal response; hypoalgesia; dorsal root ganglia neurons

#### **1. Introduction**

Hu proteins are members of the RNA-binding proteins (RBP) superfamily and are encoded by Embryonic Lethal, Abnormal Vision, and Drosophila (*ELAV*) genes. The Hu proteins family has four members HuB (encoded by *ELAV-like 2* gene), HuC (encoded by *ELAV-like 3* gene), HuD (encoded by *ELAV-like 4* gene), and HuR or HuA (encoded by *ELAV-like 1* gene). Three of these proteins have identified as neuronal specific (i.e., HuB, HuC, and HuD), while the fourth is ubiquitary (HuR).

RBP are well known for the post-transcriptional control of RNAs encoding multiple proteins [1]. In particular, RBPs play essential roles in the nervous system, such as alternative splicing of neuronal proteins (i.e., neurotransmitters, membrane receptors, cell adhesion molecules, and components of signal transduction proteins), protection of the mRNAs for long-distance transport and guidance of the protein localization [2–5].

Neuronal-enriched *ELAV*-like (*nELAVL*) Hu proteins were described to play essential roles in neuronal development and plasticity [6,7] in the central and peripheral nervous system. *nELAVL* Hu proteins are binding to the adenylate-uridylate-rich (ARE) RNA elements in the 3 untranslated regions (3 -UTR) of target proteins, including growth associated protein 43 (GAP-43) [8,9], c-myc and vascular endothelial growth factor (VEGF) [10], and neprilysin (a potent amyloid β degrading enzyme) [11] stabilizing them. Moreover, *nELAVL* Hu proteins autoregulate themselves [12] or interact/stabilize other neuronal RBPs, e.g., Musashi-1 [13] and NOVA1 [14]. In the central nervous system, *nELAVL* Hu-proteins have been involved in regulating neuronal excitability by controlling the glutamate synthesis pathway and their gene deletion induces spontaneous epileptic seizure activity [15], by binding to the mRNA encoding Kv1.1 voltage-gated potassium channels [16]. In the peripheral nervous system, *nELAVL* Hu proteins are localized in the dorsal root ganglia (DRG) neurons [17–20]. This anatomical localisation of *nELAVL* Hu-proteins is correlated with their functional role of binding mRNA encoding proteins (i.e., brain-derived neurotrophic factor, GAP-43) involved in peripheral nerve regeneration upon lesion [8,21,22], being upregulated in the early stages of nerve recovery.

The role of RBPs in diabetes and its complications was extensively documented [23]. To detail, it was described the regulation of beta-pancreatic cell function by various RBPs [24], including neuronal-enriched RBPs [25]. The altered regulatory function exerted by the ubiquitary HuR protein in diabetes was often described [25–28]. On the other hand, although the role of *nELAVL* Hu proteins was described in diabetes [29–31], yet no attention was paid so far to their expression changes in DRG sensory neurons associated with the diabetic condition.

We aimed to elucidate the role played by Hu proteins expressed by the DRG sensory neurons in diabetes. To this purpose, we have employed the streptozotocin (STZ)-induced model of diabetes in CD-1 adult male mice. We have explored the gene and protein expression for *nELAVL* Hu proteins in DRG neurons between diabetic and control mice and we correlated them with the changes in animal glycemia, body weight, or their response to hot thermal stimulation. We also analyzed the distinct changes between animals sensitive or resistant to the STZ-induction of diabetes.

#### **2. Results**

### *2.1. Diabetic Mice Have Changes in Glycemia and Body Weight Compared to Diabetic Resistant or Control Mice*

We started our experimental protocol with two CD-1 mice groups: citrate buffer-injected group (*N* = 20) and STZ-injected group (*N* = 20). In the STZ-injected group, seven out of 20 animals died quickly. We measured the glycemia weekly for 7 weeks. Hyperglycemia was considered above 260 mg/dL, as previously described [32,33]. Considering the hyperglycemia threshold, at the end of 7 weeks we

separated the surviving animals of the STZ-injected group (*N* = 13) into two subgroups: STZ-sensitive group (*N* = 7, glycemia > 260 mg/dL) and STZ-resistant group (*N* = 6, glycemia < 260 mg/dL), that will be further called diabetic group and diabetic resistant group, respectively. Then, we plotted the glycemia variation for the diabetic, diabetic resistant, and control mice groups (Figure 1).

**Figure 1.** Blood glucose values (in mg/dL) were represented as mean ± SD for control, diabetic resistant, and diabetic mice. Statistical significance was indicated \*\*\* *p* < 0.001, \*\* *p* < 0.01, \* *p* < 0.05.

Only the diabetic group had an increase in glycemia (from 144.16 ± 17.29 mg/dL to 615.50 ± 45.05 mg/dL, *N* = 7), while the diabetic resistant group (from 106.85 ± 18.73 mg/dL to 152.14 ± 33.55 mg/dL, *N* = 6) and the control group (from 127.89 ± 26.63 mg/dL to 107.89 ± 36.63 mg/dL, *N* = 20) had no significant changes. The two-way ANOVA analysis indicated statistical significance of the glycemia, for the diabetic condition, and for their interaction (Table S1). The one-way ANOVA weekly comparison between the animal groups indicated that diabetic mice had higher glycemia compared to diabetic resistant and control mice, starting from the first week after STZ-induction of diabetes and with this difference accentuating to fifth–seventh week (Table S2) and is indicated in Figure 1. The weekly comparison in the diabetic group showed the increase of glycemia up to the fifh week, followed by a plateau-like evolution up to the eighth week (Table S3). Meanwhile, the weekly comparison of the glycemia values for the control and diabetic groups did not indicate significant changes.

We have also measured the body weight for the diabetic, diabetic resistant and control CD-1 mice groups weekly, for 8 weeks, after the STZ-diabetes induction (Figure 2). All animal groups had an overall increase of the body weight, but the increase rate was higher for the control group (from 21.75 ± 2.75 g to 34.19 ± 2.66 g, *N* = 20) compared to the diabetic group (from 22.74 ± 2.68 g to 30.08 ± 2.66 g, *N* = 7) and diabetic-resistant group (from 22.74 ± 2.18 g to 28.65 ± 2.78 g, *N* = 6). The two-way ANOVA analysis indicated statistical significance of the body weight, of the diabetic condition, and of their interaction (Table S4). The one-way ANOVA comparison between the animal groups indicated that control mice were heavier than diabetic mice and diabetic resistant mice, starting from the fourth wk after STZ-induction of diabetes (Table S5). The one-way ANOVA weekly comparison in each animal group showed a continuous body weight increase, for the whole duration of the protocol (for 8 weeks), with statistical significance in all three animal groups (Table S6).

**Figure 2.** Body weight (in g) monitored for 8 weeks after streptozotocin (STZ)-induction of diabetes. Body weight values were represented as mean ± SD for control, diabetic resistant, and diabetic mice. Statistical significance is indicated \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.
