Bioelectronic medicine for restoring autonomic balance in autoimmune diseases

Recently, advances in the neural interface technology led to a development a new therapeutic approach termed the bioelectronic medicine (BEM) that aims to restore the SPB in the autonomic nervous system [ 24 ]. In this section, we will review the autonomic nerves that have been clinically targeted by the BEM. The largest and most dominant parasympathetic nerve is the vagus nerve. The BEM that is targeting the vagus nerve can modulate the activities of multiple internal organs, as the vagus nerve directly innervates the heart, lungs, liver, esophagus, stomach, pancreas, small intestine, and proximal colon, while its innervation of kidneys, spleen, adrenal medulla, and reproductive organs is debated [ 25 ]. At the cervical (neck) level, the vagus nerve can be easily accessed without laparoscopy to allow simple surgical placement of the stimulating electrode cuff [ 26 ]. At the abdominal level, the vagus nerve is located rather deep in the body and requires a laparoscopic surgery for placing the electrode cuffs on the anterior and posterior vagal nerve trunks [ 27 ]. The vagus nerve stimulation (VNS) at the cervical level is Food and Drug Administration (FDA)-approved for treating epilepsy [ 28 ], depression [ 29 ], and stroke [ 30 ]. Clinical trials are ongoing in the US for the RA [ 31 , 32 ], heart failure [ 33 , 34 ], asthma [ 35 ], post-COVID syndrome [ 36 ], and nephrotic syndrome [ 37 ] and in the EU for Crohn’s disease [ 38 ]. The abdominal VNS was previously FDA-approved for treating obesity, but that approval was since reversed [ 39 ], citing a rather its modest effect on weight loss compared to bariatric surgical procedures [ 40 ]. The sacral nerve is nearly as large as the cervical vagus and is most easily accessible (also without laparoscopy) inside the sacral foramina allowing simple surgical placement of a stimulating lead along the nerve [ 41 ]. Unlike the vagus nerve, which is nearly 100% parasympathetic, the sacral nerve at the S2-S4 levels contains the parasympathetic as well as somatic fibers, including both the afferents and efferents [ 42 ]. The sacral nerve stimulation (SNS) can modulate the activities of multiple pelvic organs, including distal colon, anorectum, bladder, urethra, and genitals [ 43 ]. The SNS is FDA-approved for treating overactive bladder [ 44 ] and fecal incontinence [ 45 ] and is approved for constipation in the EU [ 46 ] but not in the US. Clinical trials are ongoing or recently completed for the RA [ 47 ], irritable bowel syndrome [ 48 , 49 ], and endometriosis [ 50 ]. Unlike the parasympathetic nerves that arise from two discrete anatomic locations (cervical brainstem and sacral spinal cord), the sympathetic nerves arise at multiple sympathetic ganglia along the torso [ 51 ], with each sympathetic nerve typically following an artery by splitting into multiple fascicles that surround the artery to form a neurovascular bundle [ 52 ]. These sympathetic fascicles cannot be easily dissected from the artery wall, so a stimulating electrode (e.g. a cuff or a patch) has to be placed around the artery without compromising its pulsatility [ 52 – 54 ]. Delicate arterial cuffs for applying the BEM therapy were recently developed by Galvani Bioelectronics [ 52 , 53 ] and applied at the splenic nerve for treating the RA [ 55 , 56 ]. In addition to stimulating the parasympathetic and sympathetic nerves carrying both the afferent and efferent information, yet another BEM approach is to stimulate the autonomic nerves containing 100% afferent fibers that relay visceral information into the brainstem autonomic centers. Surgically, one of the autonomic afferent nerves, the carotid sinus nerve, is readily accessible by placing a stimulating patch electrode on a surface of the carotid artery at a level of the carotid sinus without compromising its pulsatility [ 54 ]. Such patch-based BEM therapy is FDA-approved for treating the heart failure [ 57 ]. An alternative surgical approach for accessing the carotid sinus nerve is to insert an intra-arterial stent in the carotid artery at a level of the carotid sinus (also without compromising the arterial pulsatility), with the stent transiently expanding during each cardiac systole to apply mechanical pressure on the nerve endings [ 58 ]. Such stent-based BEM therapy is being clinically evaluated in the US for treating the hypertension [ 59 , 60 ]. Yet another BEM approach for activating the vagus nerve is via the somato-vagal reflex induced by electrical skin stimulation on the leg or arm at the locations corresponding to specific acupoints, such as ST36 (Zusanli, on the leg) and PC6 (Neiguan, on the arm) [ 61 – 68 ]. Electrical stimulation at these acupoints was shown to be effective in treating sepsis [ 69 , 70 ], ulcerative colitis [ 71 ], and RA [ 72 ].

High prevalence of AIDs and demonstration of the sympathetic dominance in most AID patients create an opportunity for applying the BEM to restore the SPB as a means of inducing a well-demonstrated beneficial anti-inflammatory effect on the autoimmune activity. The BEM clinical toolbox is already quite extensive and includes cervical and abdominal VNS, SNS, splenic nerve stimulation, and carotid nerve stimulation, with many smaller autonomic nerve branches potentially targetable as well, as the neural interface technology is miniaturizing in the coming years. The BEM therapies have already been applied for treating IBD (both the Crohn’s disease and ulcerative colitis), RA, and lupus. As we learn more about the potential mechanisms of the BEM therapy at the cellular and molecular levels and develop smaller neural interfaces, the range of AIDs treated by the BEM is likely to expand rapidly.

The BEM anti-inflammatory therapies targeting the spleen (i.e. VNS and splenic nerve stimulation) are shown to activate the choline acetyltransferase (ChAT)-and β2 nicotinic acetylcholine receptor (β2 nAChR)-expressing T cells, which in turn activate the anti-inflammatory α7 nAChR-expressing macrophages resident in the spleen; while the BEM anti-inflammatory therapies targeting the intestines (i.e. VNS and SNS) activate the enteric neurons, which in turn activate the anti-inflammatory α7 nAChR-expressing macrophages resident in the intestinal mucosa [ 105 – 107 ]. In both the spleen and intestines, activation of the α7 nAChR-expressing macrophages inhibits the release of pro-inflammatory cytokines [ 107 , 108 ]. Interestingly, while the intestinal mucosa contains both the α7 nAChR-expressing macrophages and the α7 nAChR-expressing dendritic cells, only the former were shown to activated by the VNS [ 109 ]. BEM-initiated activation of the α7 nAChR-expressing macrophages in the intestines restores a healthy balance among resident T cells by reducing prevalence of pro-inflammatory ones, such as the T-helper 17 (Th17) cells, and increasing prevalence of anti-inflammatory ones, such as the regulatory T (Treg) cells [ 107 ]. Additional and less-explored BEM-initiated mechanisms in the intestines involve enhancing the barrier function of the epithelial cells [ 110 , 111 ] and suppressing infiltration of neutrophils and monocytes into the intestinal tissue [ 112 ]. Cellular and molecular mediators of the anti-inflammatory response in the adrenal medulla are not well-studied, with a recent animal study showing that the anti-inflammatory response to a low-level (0.5 mA) somatic afferent stimulation at ST36 uses the vagal supraspinal pathway, which induces NPY+ adrenal chromaffin cells to secrete epinephrine, norepinephrine, and dopamine; while the anti-inflammatory response to a high-level (3 mA) somatic afferent stimulation at ST36 uses the sympathetic spinal (rather than vagal supraspinal) pathway, which induces NPY + splenic neurons to secrete norepinephrine [ 104 , 113 , 114 ].

A low (usually undetectable) level in autoimmunity is normal and even essential for lymphocyte selection and immune system homeostasis, while a moderate level of autoimmunity can be detected as circulating autoantibodies and small tissue infiltrates but generally does not cause clinical symptoms. When a high level of autoimmunity is sustained for a few days or weeks, it becomes pathogenic and starts to generate clinical symptoms of various autoimmune diseases (AIDs). AIDs are caused by a chronic malfunction of the immune system, when it recognizes auto-antigens in our own molecules and cells as harmful to us, initiating a cascade of pro-inflammatory molecular and cellular events leading to destruction of our own tissues. Due to heterogenicity of affected tissues and organs throughout our body, various AIDs are diagnosed by different specialties of physicians, making it difficult to estimate the overall prevalence of AIDs. A recent epidemiological study in Spain [ 1 ] indicated an overall prevalence of about 11%, with the cumulative prevalence of 8.5% (or three quarters of all AIDs) attributed to the thyroiditis, psoriasis, inflammatory bowel disease (IBD), type 1 diabetes, rheumatoid arthritis (RA), and polymyalgia rheumatica ( Table 1 ). The list of prevalent AIDs may soon be extended to include the Post-COVID syndrome, since autoimmunity play a critical role in its development [ 2 ]. As can be seen on Table 1 , the most commonly affected organs are the thyroid gland, skin, small and large intestines, pancreas, skeletal muscles, joints, and kidney. Other organs can also be affected, such as the liver ( e.g . primary biliary cholangitis, primary sclerosing cholangitis, and autoimmune hepatitis) and bladder ( e.g . interstitial cystitis), but with a prevalence below 0.2% [ 3 , 4 ].

As described in the previous section, various BEM approaches and nerve branches are available for modulating the activities of the parasympathetic and sympathetic nerves. Therefore, it is important to determine which of these might be best for treating individual AIDs. Twenty years ago, Dr. Kevin Tracey suggested that VNS can be applied to suppress autoimmune activation in multiple immuno-competent organs, such as the spleen, lymph nodes of the small intestine, and adrenal medulla [ 73 ]. The anti-inflammatory effect of VNS on the spleen attracted a lot of attention, once clinical efficacy was demonstrated for several spleen-mediated AIDs, such as the RA [ 74 , 75 ] and lupus [ 76 ]. The anti-inflammatory effect of VNS on the small intestine was demonstrated in the Crohn’s disease [ 77 , 78 ]. The animal studies suggest that the anti-inflammatory effects of VNS on two immuno-competent organs, spleen and small intestine, are mediated via different abdominal branches of the vagus [ 79 – 82 ]. The same abdominal branch of the vagus controls both the spleen and the adrenal gland: it passes through the sympathetic celiac ganglion [ 83 ], where it makes no synaptic connections with the splenic nerve [ 84 ], so the VNS effect on the spleen is likely mediated by direct vagal synapses on the spleen [ 85 ], while the VNS effect on the adrenal medulla is likely indirectly mediated by the synaptic connections between the vagus and the adrenal nerve (the vago-sympathetic reflex) [ 86 , 87 ]. The VNS also has an anti-inflammatory effect on other organs, such as the liver, lungs, and upper genital tract, with preclinical studies demonstrating its efficacy in the rodent models of pulmonary arterial hypertension [ 88 ], hepatitis [ 89 ], and endometriosis [ 90 ]. Therefore, future clinical studies may be designed to apply VNS at the organ-specific abdominal branches in order to achieve better efficacy and avoid the side effects seen in the cervical VNS, particularly on the heart (bradycardia) and lungs (dyspnea and bronchospasms) [ 91 , 92 ]. The distal colon is controlled by a branch of the sacral nerve (the pelvic splanchnic nerve) rather than by the vagus [ 93 ]. Accordingly, the anti-inflammatory effect of SNS on the colon and bladder have been observed in the clinical studies for treating ulcerative colitis [ 94 ] and interstitial cystitis [ 95 ], while the anti-inflammatory effect of SNS on other pelvic organs (e.g. ovaries and prostate) has not been clinically evaluated. Since the vagus and sacral nerves contain both the afferent and efferent parasympathetic fibers, the anti-inflammatory effect of VNS and SNS could be due to either a direct activation of these parasympathetic efferents and/or the activation of the parasympathetic afferents projecting to the brainstem parasympathetic control centers and resulting in activation of the vagal and sacral efferents [ 96 , 97 ]. Moreover, the vago-sympathetic reflex has been proposed, where the abdominal vagal efferents in the celiac ganglion activate the sympathetic efferents in the greater splanchnic nerve branches, including its splenic nerve [ 98 – 103 ] and adrenal nerve [ 104 ]. With that rationale, Galvani Bioelectronics recently initiated clinical trials in the US and UK, where a direct splenic nerve stimulation (rather than VNS) is being used to treat the RA [ 55 , 56 ]. No other sympathetic nerves have been targeted to date in clinical studies aiming to provide anti-inflammatory therapy to patients with AIDs.

Despite a heterogenicity of autoimmune responses in AIDs, one emerging common theme is the involvement of the autonomic nervous system with two main components, the sympathetic and parasympathetic branches, jointly regulating the immune functions throughout the body [ 5 ]. The degree of sympatho-parasympathetic balance (SPB) can be readily measured by calculating the spectral parameters of heart rate variability (HRV) in the electrocardiogram [ 6 , 7 ]. The HRV assessment of autonomic activity gained popularity due to its accuracy and reproducibility, when the assessment is performed at specific physiological conditions ( e.g ., physical activity or post-prandial state) [ 8 ]. The HRV assessment is also simpler while providing a similar accuracy compared with the traditional autonomic tests, such as deep breathing, Valsalva maneuver, and orthostatic stress (tilt table) test [ 9 , 10 ]. Using the HRV measure of SPB, a clinical study in thyroiditis patients uncovered considerable sympathetic dominance before the therapy and a recovery of SPB after a 6-month thyroid hormone replacement therapy [ 11 ]. Similarly, considerable sympathetic dominance was observed in patients with psoriasis [ 12 ], IBD [ 13 – 16 ], type 1 diabetes [ 17 – 19 ], RA [ 20 – 22 ], and polymyalgia rheumatica [ 23 ]. Furthermore, the HRV measure of SPB was shown to correlate with severity of the autoimmune response and to be predictive of the AID progression, as seen in the IBD [ 13 , 14 ], type 1 diabetes [ 17 – 19 ], and RA [ 20 – 22 ].