Hyperplasia is a cellular adaptation to maintain homeostasis, and hyperplastic organs are larger due to cell proliferation [1]. Hyperplastic changes can be induced by physiological stressors (e.g., an increase in the size of breasts during pregnancy) or by pathological stressors (e.g., enlargement of adrenal gland stimulated by adrenocorticotropic hormone released by a pituitary adenoma).

The PVS, reported as an anatomical entity of the acupuncture meridian by Kim, is a novel circulatory tissue composed of primo nodes and primo vessels [2-5]. In particular, Kim [6] reported in 1965 that hemolytic anemia induced by phenylhydrazine (PHZ) or phosphonoacetic acid enlarged the primo nodes and vessels in the vascular system by 2-3 times in the rabbits. Kim reported an increased erythropoietic activity in the enlarged primo nodes. The earlier observation on the erythropoietic changes in the primo nodes were further confirmed by recent studies using animal models of heart failure and acute hemolytic anemia. in 2017, Lim et al. [7] showed in a rat model of heart failure (HF) that hyperplastic changes associated with hematopoiesis were observed in the primo vascular system (PVS) tissue isolated from the organ surface of the abdominal cavity (osPVS). In the study, there were significant increases in the number, size, and proportion of red-colored tissues in the osPVS in HF rats. Later, Shen et al. [8] also found similar morphological changes of osPVS tissue in a rat model of hemolytic anemia (HA) induced by PHZ. The effects of PHZ on red blood cell (RBC) and erythrocyte counts peaked during the first 3-6 days and attenuated starting 10 days after administration. During the same period, reversible hyperplastic changes occurred in the osPVS tissues. Recently, the latter was further confirmed in a micro-computed tomographic study of the osPVS tissue of rats with HA [9].

The initial pathology of HF and HA are different, but it is likely that the common downstream outcome is systemic hypoxia, a low-oxygen state in the tissue. It is known that hypoxia or diseases causing hypoxia are accompanied by an increase in sympathetic activity. In HF, sympathetic nerve activity is elevated in the early symptomatic phase, and the initial elevation plays a compensatory role. However, prolonged hyperactivation blunts the sensitivity of baroreflexes and cardiopulmonary reflexes, aggravates the clinical status, and negatively affects the prognosis [10-12]. An increase in sympathetic activity has also been reported in acute anemia or hemorrhage in dogs [13] and rabbits [14], as well as in sleep apnea, which causes intermittent hypoxia [15,16].

Hypoxia is also known to induce extramedullary hematopoiesis, a form of blood cell hyperplasia outside the bone marrow that occurs in the spleen, liver, and many other organs [17,18]. Erythropoiesis induced by hypoxia in the spleen is well characterized [19-21]. Stress erythropoiesis is a stem cell-based tissue regeneration process [22].

In the rats with HF or HA, systemic hypoxia is accompanied by sympathetic overactivity in rats showing hyperplastic changes in the osPVS tissue [7,8]. Catecholamines also modulate hematopoiesis in the bone marrow [23]. Based on these findings, we hypothesized that sympathetic overactivity is associated with hyperplastic changes in the osPVS tissue of rats with HF and HA. To test this hypothesis, we examined the effects of exercise training (ExT), which normalizes elevated sympathetic activity in HF rats [24], and 6-hydroxydopamine (6-OHDA), which induces chemical sympathectomy by destroying the catecholaminergic nerve endings [25]. In addition, we further identified the adrenergic receptors mediating hyperplastic morphological changes in the PVS system tissue.

1. Animals

Sprague-Dawley rats weighing 120-130 g (male, 5 weeks old; Han Lym Lab. Animal Co., Gyeonggi-do, Korea) were used in this study. The experiments were conducted in accordance with the Guide for Laboratory Animal Care Advisory Committee of Seoul National University and approved by the Institute of Laboratory Animal Resources of Seoul National University (SNU-190530-3).

2. Experiment I: effects of ExT on the osPVS tissue of HF rats

1) HF model

To prepare a rat model of HF, myocardial infarction was induced by ligating the left coronary artery as previously described [26]. Briefly, the rats were intraperitoneally injected with an anesthetic cocktail (alfaxalone, 41.7 mg/kg and xylazine, 16 mg/kg; Alfaxan and Rompun respectively, M&S Korea, Gyeonggi-do, Korea). A mechanical ventilation machine (Harvard Apparatus, Holliston, MA, USA) was applied for ventilation, connected with a 16-gauge catheter. The left coronary artery was ligated by 6-0 sterile silk sutures (Ailee Co., Ltd., Busan, Korea) after the heart was exposed at the third intercostal space. The sham-operated rats had the same process performed except coronary artery ligation. The infarction sizes were determined at the end of the seventh week after cardiac surgery, when the osPVS tissue samples were isolated. All the hearts and lungs were stored in 10% neutral buffered formalin solution for later analysis. Infarction sizes smaller than 30% were excluded from the data analysis.

2) ExT protocol

To examine the effects of ExT on hyperplastic morphological changes in the rats with HF, a 3-week ExT regimen was applied to the HF rats starting in the fourth week after cardiac surgery during the dark phase (20:00-24:00). The rats in the ExT groups were individually mounted on a treadmill (Exer-3/6 Treadmill, Columbus Instruments, Columbus, OH, USA) [24]. The rats were first subjected to exercise familiarization at a speed of 10 m/min and then increased to 20-23 m/min up a 5% grade. To ensure the effects of ExT, the ExT group rats were trained 5 days per week for 3 weeks (15 training sessions in total). Rats that failed to complete all ExT sessions were excluded. The rats that were not assigned to ExT were allocated to the sedentary (Sed) groups, Sham-Sed and HF-Sed.

3. Experiment II: effects of 6-OHDA on the osPVS tissue in the rats treated with PHZ

To examine whether chemical sympathectomy can affect the hyperplastic morphological changes observed in rats with HA, rats were randomly allocated to the control group and groups treated with 6-OHDA, PHZ, and 6-OHDA+PHZ. The rats in the control and 6-OHDA groups were intraperitoneally administered 0.2 ml of phosphate-buffered saline (control group) and 6-OHDA (100 mg/kg with 0.1% ascorbic acid) for chemical sympathectomy (6-OHDA group) on day 0 [27]. The rats in the PHZ group were intraperitonially administered PHZ for 2 consecutive days at 40 mg/kg from day 0. The rat in the 6-OHDA+PHZ group received both 6-OHDA and PHZ treatment. On day 6 after the first PHZ injection, we isolated the osPVS tissues from the surface of the abdominal organs of rats under anesthesia using a stereomicroscope with ×10 magnification.

4. Experiment III: effects of norepinephrine on the osPVS tissue in rats

1) Effects of norepinephrine of the osPVS tissue

To determine whether sympathetic activity is involved in hyperplastic morphological changes in the osPVS, 4 doses of norepinephrine (NE) were administered to the rats in 4 treatment groups. Rats were randomly allocated to these groups. The NE was dissolved in dimethyl sulfoxide and administered through an osmotic minipump (Alzet model 2001, Durect, Cupertino, CA, USA) inserted into the subcutaneous layer of the nape at the concentrations of 0, 1, 3, and 10 mg/kg/d [27,28]. On the sixth day after insertion, blood samples were drawn into tubes by an ice-chilled syringe with EGTA/glutathione (Sigma-Aldrich, St. Louis, MO, USA) solution. The plasma was separated by being centrifuged at 10,000 × g for 5 min at 4℃ and stored at –70℃ for later NE assays [29]. The plasma NE concentration was determined by using an enzyme-linked immunosorbent assay kit (BA E5200, LDN Immunoassays and Services, Nordhorn, Germany).

2) Effects of NE on osPVS tissue under α- or β-adrenoceptor blockade

The determine the receptor subtype mediating the effects of NE on the morphology of osPVS tissue, we examined the effects of NE in the presence of adrenergic antagonists. Typical non-selective α-antagonist (phentolamine mesylate, 20 mg/kg/d) and β-antagonist (propranolol hydrochloride, 50 mg/kg/d) were administered in drinking water from day 0 to day 6 to the rats treated with NE (3 mg/kg/d, from day1 to day 6) [30]. NE, phentolamine and propranolol were purchased from Sigma-Aldrich, St. Louis, MO, USA). On day 6, the osPVS tissues were isolated from the rats and morphological changes were analyzed.

3) Hematological tests

For the blood cell counts, all rats were anesthetized with isoflurane. In Experiment I, blood samples (0.3-0.5 ml) were collected from the retro-orbital sinus in tubes with K2EDTA (BD Diagnostics, Franklin Lakes, NJ, USA) on the second day of the seventh week after cardiac surgery. In Experiments II and III, blood samples were collected on the sixth day after the initial administration of PHZ, 6-OHDA, or NE. Hematological parameters such as RBC count, white blood cell (WBC) count, hematocrit, and mean cell volume (MCV) were determined using a high-volume hematology analyzer (ADVIA 2120, Siemens Healthineers, Erlangen, Germany).

4) Sampling of the osPVS tissues and staining

The osPVS tissue samples were isolated from the surface of abdominal organs under stereomicroscopy (×10 magnification). In case of bleeding and contamination of the PVS tissue by blood, such tissues were excluded from the analyses. Therefore, the red color in the some PVS tissues analyzed must be from the chromophore from the PVS tissue itself as shown previously [7]. For hematoxylin and eosin staining, the tissues were sectioned in paraffin and stained according to conventional protocols. An osPVS tissue sample was defined as consisting of one or more PNs connected by at least one PV.

5) Morphological analyses

The size of the PNs and the number of tissues with red chromophore were determined from the pictures of osPVS tissue collections in Petri dishes containing phosphate-buffered solution with a black background. We measured PN size as the area of PNs using ImageJ software. All data are expressed as mean ± standard error of the mean. The Student’s t-test or Fisher exact test was used to analyze the data. One-way analysis of variance was also used to determine the significance of the differences in the effects of NE with the post hoc Dunnett test. The threshold for statistical significance was set at p < 0.05.

The results are based on analyses of 37 PNs from 19 rats in Experiment I, 58 PNs from 12 rats in Experiment II, and 132 PNs from 19 rats (III-a) and 238 PNs from 23 rats (III-b) in Experiment III.

1. ExT normalized morphological changes of osPVS tissue in rats with HF

We previously showed that the size and proportion of red-colored tissues increased in a rat model of HF [7]. In Experiment I, we tested whether ExT could also reverse HF-induced morphological changes in the osPVS tissue in rats since it is well-known that ExT can reverse elevated sympathetic activity in HF [24,31].

Fig. 1A shows a typical osPVS tissue sample on the surface of abdominal organs in situ. The sample has 2 PNs (arrowheads) connected by a PV (arrow), and the upper PN measures 0.63 × 1.16 mm (short × long axis), with an area of 0.62 mm². The samples from each rat from 4 treatment groups (Sham-Sed, Sham-ExT, HF-Sed, HF-ExT) were placed on a Petri dish filled with phosphate-buffered solution as shown in Fig. 1B. The size and number of samples, as well as the proportion of tissues with red chromophore, were determined from those images. Some PNs were larger in the HF-Sed group (B3). Some samples had red chromophore (B2-B4), which was observed in both the usual (B2 and B4) and enlarged tissues (B3). Fig. 1C is a hematoxylin and eosin–stained image of the PNs showing 3 types of cells: erythrocytes, small-round cells with little cytoplasm, and large cells with granules. In the red-colored PNs, erythrocytes were the major cells (C2). These observations are consistent with previous reports [7,8], and indicate that the red chromophore is not from blood contamination that occurred during surgical procedures.

Figure 1. Effects of exercise training (ExT) on the morphology of osPVS tissue in rats with heart failure (HF). (A) Representative sample of osPVS tissue composed of 2 PNs (arrow heads) and a PV (arrow) on the surface of the small intestine from a rat in the Sham-Sed group. Scale bars: 2 mm. (B) Collections of isolated intact PVS tissue samples in phosphate-buffered saline solution from one rat from each of 4 treatment groups; Sham-Sed (B1), Sham-ExT (B2), HF-Sed (B3) and HF-ExT (B4). Scale bars: 5 mm. Note that the intensity of disseminated red chromophore in the PNs from Sham-ExT and HF-ExT rats. (C) Hematoxylin and eosin-stained cells in the PN sections without (C1) and with red chromophore (C2). Note 3 types of cells: RBCs (1), small round cells with large nuclei (2), and large cells with dark granules (3). Scale bar: 20 μm. (D-F) Summary bar graphs showing the number of osPVS samples (D), the size of the PN (E), and the proportion of osPVS samples containing red chromophore (F) in the rats of Sham-Sed (n = 24 from 5 rats), Sham-ExT (n = 31 from 5 rats), HF-Sed (n = 41 from 5 rats), and HF-ExT groups (n = 41 from 4 rats). Values are means ± standard error. *p < 0.05; **p < 0.01; ***p < 0.001 by the Student’s t-test or the Fisher exact test.

The results of these experiments are summarized in Fig. 1D-F. The mean number of isolated osPVS per rat was higher in the HF condition, resulting in a greater total number from 4 rats in the HF-Sed group (n = 32) compared to the Sham-Sed group (n = 20). ExT did not affect the number of osPVS in either the sham or the HF condition (Fig. 1D). The size and color of PNs were also different in the HF group, as the size of the PNs was more than 2-fold larger in the HF-Sed group than in the Sham-Sed group (1.40 ± 0.18 vs. 0.46 ± 0.09 mm², p < 0.01, n = 4), suggesting that HF increased the size of PN. As shown in Fig. 1E, ExT reduced the PN size in HF to a size comparable to that in the Sham-Sed group (0.56 ± 0.08 mm², p < 0.01), but did not affect the PN size in the sham condition (p = 0.38). As a result, the mean PN size in HF-ExT rats reversed to the sham level (Sham-Sed and Sham-ExT). Fig. 1F shows that the proportion of red-colored osPVS was higher in the HF-Sed group, and ExT reduced this elevation to a level comparable to that of the Sham-Sed group. It is of note that ExT increased the proportion of red-colored osPVS in the sham condition. The results indicate that the effect of ExT on the proportion of red-colored PNs is dependent on the physiological state of the rats (i.e., basal sympathetic tone). HF and/or ExT did not affect blood cell counts, including RBCs. There was no indication of anemia in the HF group (Table 1). Collectively, the results indicate that ExT reversed the HF-induced increase in size and proportion of red-colored tissue to its control levels.

Blood was sampled from 12-week-old rats at the seventh week after cardiac surgery. N is the number of the rats tested in each group. The reticulocyte number is presented as % of RBCs. Values are means ± standard error of the mean. *p-values were obtained by one-way ANOVA. Note that there was no significant difference among the blood cell counts of the 4 treatment groups. Sham-ExT and Sham-Sed, rats with sham surgery with (Sham-ExT) and without exercise training (Sham-Sed); HF-ExT and HF-Sed, HF rats with (HF-ExT) and without exercise training (HF-Sed). RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell..

&md=tbl&idx=1' data-target="#file-modal"">Table 1

Blood cell counts in heart failure (HF) rats with or without exercise training (ExT).

	Sham-Sed (n = 6)	Sham-ExT (n = 7)	HF-Sed (n = 10)	HF-ExT (n = 5)	p*
RBC (106 cells/ml)	8.17 ± 0.19	8.38 ± 0.10	8.62 ± 0.26	8.94 ± 0.06	0.155
Hemoglobin (g/dL)	15.6 ± 0.24	15.7 ± 0.15	16.0 ± 0.46	16.8 ± 0.14	0.119
Hematocrit (%)	45.5 ± 0.54	46.5 ± 0.48	47.4 ± 1.34	49.8 ± 0.39	0.087
MCV (fL)	55.8 ± 0.91	55.5 ± 0.54	55.1 ± 0.60	55.8 ± 0.64	0.879
MCH (pg)	18.8 ± 0.30	18.7 ± 0.09	18.6 ± 0.14	18.7 ± 0.05	0.865
MCHC (g/dL)	33.7 ± 0.12	33.7 ± 0.05	33.6 ± 0.03	33.6 ± 0.10	0.931
Reticulocytes (%)	2.78 ± 0.48	2.38 ± 0.21	2.78 ± 0.33	2.05 ± 0.16	0.428
WBC (103 cells/ml)	13.0 ± 1.63	11.2 ± 1.09	12.1 ± 0.67	13.3 ± 1.49	0.742

Blood was sampled from 12-week-old rats at the seventh week after cardiac surgery. N is the number of the rats tested in each group. The reticulocyte number is presented as % of RBCs. Values are means ± standard error of the mean. *p-values were obtained by one-way ANOVA. Note that there was no significant difference among the blood cell counts of the 4 treatment groups. Sham-ExT and Sham-Sed, rats with sham surgery with (Sham-ExT) and without exercise training (Sham-Sed); HF-ExT and HF-Sed, HF rats with (HF-ExT) and without exercise training (HF-Sed). RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell..

2. 6-OHDA, an agent for chemical sympathectomy, normalized PHZ-induced changes in osPVS morphology and blood cell counts

We previously found that ExT can ameliorate elevated sympathetic tone by reducing the electrical excitability of pre-sympathetic neurons [24]. Therefore, we hypothesized that the ExT-induced reduction in sympathetic tone is the mechanism underlying the ExT-induced reversal of morphological changes in the HF rats results shown in Fig. 1. In Experiment II, we examined the effects of depleting sympathetic transmission on the similar hypertrophic changes of the osPVS in the rats with acute HA induced by PHZ administration [8]. We chose a rat model of chemical sympathectomy using 6-OHDA [25,32], in which sympathetic activity can be reversibly removed.

Fig. 2 illustrates the effects of 6-OHDA on the morphology of osPVS tissue in rats with HA. Fig. 2A present collections of osPVS tissue samples from the untreated rats (control) and those treated with PHZ and 6-OHDA+PHZ. The red-colored tissues were more frequent in the rats of PHZ group. As shown in Fig. 2B, the mean number of isolated osPVS samples per rat was greater in the rats in the PHZ group, resulting in a higher total number from the 3 animals in the PHZ-injected group (n = 25) than in the control group (n = 12). Sympathectomy with 6-OHDA did not affect the number of osPVS in the PHZ-injected groups. However, surprisingly, we were not able to sample a single osPVS tissue from the 6 rats treated with 6-OHDA. Fig. 2C and 2D illustrate that PHZ increased the PN size (p < 0.05, n = 3) and the proportion of red-colored PVS tissue samples (p < 0.001, n = 3). These results are similar to those induced by ExT in Fig. 1. In contrast, in the rats treated with 6-OHDA, those effects of PHZ were significantly inhibited (Fig. 2C and 2D). Administration of 6-OHDA significantly reduced PN size (p < 0.05, n = 3, Fig. 2C) in the 6-OHDA+PHZ group to a level larger than that of the control group. Similarly, 6-OHDA blocked the increase in the proportion of red-colored osPVS tissue in the rats treated with PHZ (p < 0.001, n = 3, Fig. 2D) to a level comparable to the control. Collectively, the results indicate that the chemical sympathectomy with 6-OHDA reversed the morphological changes in the rats with PHZ-induced HA.

Figure 2. Effects of chemical sympathectomy (6-OHDA) on the morphology of osPVS tissue in rats with acute HA. (A) Collections of isolated intact PVS tissue samples in PBS solution from one rat from each of 3 groups. A1, Control group; A2, PHZ group treated with phenylhydrazine; A3, group treated with 6-6OH dopamine and phenylhydrazine. Scale bars: 4 mm. Note that there no osPVS was found in the 6-OHDA-treated rats (6 rats). (B-D) Summary bar graphs showing the number of osPVS samples per rat (B), the mean size of the PN (C), and the proportion of osPVS samples containing red chromophore (D). Control, n = 12 from 3 rats; PHZ, n = 25 from 3 rats; 6-OHDA + PHZ, n = 21 from 3 rats. Values are means ± standard error. *p < 0.05, ***p < 0.001, by the Student’s t-test or the Fisher exact test. PHZ = phenylhydrazine; 6-OHDA = 6-hydroxydopamine.

Table 2 summarizes the results of blood cell counts in the rats of 4 experimental groups. The rats treated with PHZ showed blood cell counts typical to acute HA: significant decrease in erythrocytes, hematocrit and hemoglobin, and significant increase in reticulocytes [8]. It is noteworthy that the WBC count increased in the PHZ groups, which is also consistent with previous research [33,34] indicating activation of the immune response. To our surprise, 6-OHDA blocked the PHZ-induced decrease in the RBC count and hemoglobin, and the PHZ-induced increase in reticulocytes and the WBC count.

Blood was sampled from 6-week-old rats on the sixth day after administration of PHZ alone, 6-OHDA alone, or PHZ with 6-OHDA. N is the number of rats tested in each group. Values are means ± standard error of the mean. *p-value was obtained by one-way ANOVA and statistical significance determined by post hoc Dunnett t-test compared to control (marked by ‘^a’) and by Bonferroni multiple comparison (marked by ‘^#’). †The analyzer was saturated, probably due to many immature reticulocytes in this group. The manually counted number of reticulocytes was 54.5 ± 6.46%. PHZ = phenylhydrazine; 6-OHDA = 6-hydroxydopamine; RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell..

&md=tbl&idx=2' data-target="#file-modal"">Table 2

Blood cell counts in rats treated with 6-OHDA, PHZ, and PHZ with 6-OHDA.

	Control (n = 4)	6-OHDA (n = 5)	PHZ (n = 4)	PHZ+6-OHDA (n = 5)	p*
RBC (106 cells/ml)	6.00 ± 0.21	5.55 ± 0.15	3.78 ± 0.14a,#	5.58 ± 0.14#	< 0.001
Hemoglobin (g/dL)	12.9 ± 0.43	11.9 ± 0.30	9.5 ± 0.24a,#	11.8 ± 0.14a,#	< 0.001
Hematocrit (%)	38.0 ± 1.19	36.1 ± 0.89	33.6 ± 1.01a	36.0 ± 0.40	0.036
MCV (fL)	63.4 ± 0.47	65.1 ± 0.55	99.8 ± 1.79a	64.6 ± 1.14	< 0.001
MCH (pg)	21.5 ± 0.25	21.4 ± 0.18	28.2 ± 0.60a	21.2± 0.33	< 0.001
MCHC (g/dL)	34.0 ± 0.26	32.8 ± 0.16a	28.3 ± 0.19a	32.8 ± 0.13a	< 0.001
Reticulocytes (%)	11.7 ± 0.48	11.8 ± 0.39	†	11.5 ± 0.31	0.222
WBC (103 cells/ml)	7.69 ± 0.59	9.38 ± 0.32	17.73 ± 1.55a,#	11.73 ± 1.05a,#	< 0.001

Blood was sampled from 6-week-old rats on the sixth day after administration of PHZ alone, 6-OHDA alone, or PHZ with 6-OHDA. N is the number of rats tested in each group. Values are means ± standard error of the mean. *p-value was obtained by one-way ANOVA and statistical significance determined by post hoc Dunnett t-test compared to control (marked by ‘^a’) and by Bonferroni multiple comparison (marked by ‘^#’). †The analyzer was saturated, probably due to many immature reticulocytes in this group. The manually counted number of reticulocytes was 54.5 ± 6.46%. PHZ = phenylhydrazine; 6-OHDA = 6-hydroxydopamine; RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell..

3. NE increased the PN size but not the proportion of red-colored osPVS tissue

The results in Experiment II indicate that the normalization of the PHZ-induced hyperplastic changes in the osPVS tissue could be due to the 6-OHDA-induced blockade of hemolytic changes in erythrocytes in the rats treated with PHZ. To further probe the correlation between chemical sympathectomy and normalization of hyperplastic changes in the osPVS tissue, we examined whether NE administration mimicked the hyperplastic changes of the osPVS tissue in normal rats. Fig. 3A shows osPVS tissue samples collected from 1 rat from 4 treatment groups treated with 0, 1, 3, and 10 mg/kg/d (NE0, NE1, NE3, NE10 groups). The number of tissues sampled per rat is larger in the treated groups (A2-A4). The PN size in A3 was larger than in the control group (NE0). The mean number of isolated osPVS tissue samples per rat was greater in the NE-treated rats, resulting in a higher total number in the NE-treated groups than in the vehicle control (n = 22, Fig. 3B). The size of PNs was enlarged most in the rats in the NE3 group (p < 0.05, by one-way ANOVA with the Dunnett multiple comparison test, Fig. 3C). It is noteworthy that the PNs were not enlarged in the NE10 group, wherein PN size was comparable to that of the control group. Interestingly, the proportion of red-colored osPVS tissue was not altered by NE administration (Fig. 3D).

Figure 3. Dose-dependent effects of norepinephrine on the morphology of osPVS tissues in rats. (A) Collection of isolated intact PVS tissue samples in phosphate-buffered saline solution isolated from one rat from each of 4 groups. A1-A4, groups treated with 0, 1, 3, and 10 mg/kg/d; subcutaneous mini-osmotic pump) Scale bars = 4 mm. (B-D) Summary bar graphs showing the number of osPVS samples per rat (B, p = 0.031, by one-way ANOVA), the size of PNs (C, p = 0.036, by one-way ANOVA), and the proportion of osPVS samples containing red chromophore (D). The number of rats in the NE0, NE1, NE3, and NE10 groups was 5, 5, 5, and 4, respectively. Values are mean ± standard error. *p < 0.05, by one-way ANOVA with the Dunnett multiple comparison test.

Table 3 summarizes the effects of subcutaneous NE on body weight gain and blood cell counts. Subcutaneous administration of NE via mini-osmotic pump (1, 3, and 10 mg/kg/d) increased plasma NE level by about 1.5-, 4.3-, and 7.3-fold relative to the control level (0.3 ng/ml). In the rats in the NE3 group, body weight gain and the blood cell count parameters were not significantly different from those in the control group. However, in the rats in the NE10 group, body weight gain was significantly lower. Erythrocyte and hemoglobin values were higher, whereas those of MCV and reticulocytes decreased significantly. Collectively, NE at a lower dose (≤ 3 mg/kg/d) exerted a dose-dependent trophic effect on the osPVS tissue without significant change in the body weight gain and various parameters of blood cell counts. However, at a higher dose (≥ 10 mg/kg/d), the trophic effect of NE on the PVS tissue was not evident, the body weight gain was reduced, and the hematological parameters were significantly altered. Taken together, the results indicate that the dose of 3 mg/kg/d is appropriate for analyzing the trophic effect of NE on the osPVS. It is also noteworthy that the proportion of red-colored PVS tissue was not dependent on the dose of NE administration.

The rats in the control, NE1, NE3, and NE10 groups were infused via a subcutaneous mini-osmotic pump at the nape inserted at the rates of 0, 1, 3, and 10 mg/kg/day, respectively. Body weight before NE administration = 123 ± 0.70 g (n = 20, 5-week-old rats). Blood was sampled after 6 days of NE administration. NE = norepinephrine; RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell. N is the number of rats tested. †N = 4. Values are means ± standard error of the mean. *p-value was obtained by one-way ANOVA and statistical significance compared to the control group was determined by the post hoc Dunnett t-test and marked by ‘^a’ for each individual parameter..

&md=tbl&idx=3' data-target="#file-modal"">Table 3

Effects of NE infusion on body weight gain, plasma NE levels, and blood cell counts in rats.

	Control (n = 5)	NE 1 (n = 5)	NE 3 (n = 5)	NE 10 (n = 5)	p*
Plasma NE (ng/ml)	0.30 ± 0.033	0.46 ± 0.900	1.30 ± 0.169	2.20 ± 0.232a,†	< 0.001
Body weight gain (g) during NE infusion	54.7 ± 13.84	59.6 ± 11.28	54.3 ± 1.68	27.7 ± 13.42a	< 0.001
RBC (106 cells/ml)	6.0 ± 0.13	5.9 ± 0.14	6.3 ± 0.05	6.6 ± 0.16a	0.007
Hemoglobin (g/dL)	12.2 ± 0.25	12.0 ± 0.24	12.9 ± 0.06	13.0 ± 0.19a	0.006
Hematocrit (%)	42.3 ± 0.75	42.6 ± 0.65	44.2 ± 0.29	43.7 ± 0.49	0.083
MCV (fL)	70.3 ± 0.54	72.5 ± 0.87	70.5 ± 0.53	66.7 ± 1.28a	0.002
MCH (pg)	20.3 ± 0.18	20.6 ± 0.22	20.6 ± 0.11	19.9 ± 0.28	0.088
MCHC (g/dL)	28.9 ± 0.08	28.4 ± 0.38	29.2 ± 0.26	29.8 ± 0.25	0.011
Reticulocytes (%)	14.4 ± 0.56	15.9 ± 0.86	12.8 ± 0.33	9.1 ± 0.95a	< 0.001
WBC (103 cells/ml)	9.5 ± 0.92	8.0 ± 1.60	11.7 ± 2.09	9.5 ± 1.43	0.442

The rats in the control, NE1, NE3, and NE10 groups were infused via a subcutaneous mini-osmotic pump at the nape inserted at the rates of 0, 1, 3, and 10 mg/kg/day, respectively. Body weight before NE administration = 123 ± 0.70 g (n = 20, 5-week-old rats). Blood was sampled after 6 days of NE administration. NE = norepinephrine; RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration; WBC = white blood cell. N is the number of rats tested. †N = 4. Values are means ± standard error of the mean. *p-value was obtained by one-way ANOVA and statistical significance compared to the control group was determined by the post hoc Dunnett t-test and marked by ‘^a’ for each individual parameter..

4. Both α- and β-adrenoceptors are involved in NE-induced hyperplastic morphological changes in osPVS tissue

The results presented in Fig. 1-3 showed that the morphological changes of osPVS tissue can be reversed by exercise training in HF rats, blocked by chemical sympathectomy in anemia rats, and mimicked by NE in healthy rats. These findings indicate that the size and red chromophore of the osPVS tissue are under positive control of sympathetic activity. Next, we examined which receptors mediate this noradrenergic regulation of the size and red chromophore of the osPVS in healthy rats. Fig. 4A shows the collective images of osPVS tissue isolated from 1 rat in 4 treatment groups: rats without any treatment (control), rats treated with NE alone (3 mg/kg/d, for 6 days; NE), rats treated with NE and an α-adrenoceptor blocker (phentolamine, 20 mg/kg/d in drinking water for 7 days; Phen-NE), or a β-adrenoceptor blocker (propranolol, 50 mg/kg/d in drinking water for 7 days; Prop-NE). As illustrated in Fig. 4A, the number of samples per rat is reduced in Prop-NE group (A4) and the PN size is increased in the NE and Phen-NE group (A2 and A3). The mean number of samples per rat was significantly smaller in the Prop-NE group than those in the other 3 groups (Fig. 4B). NE administration alone or with pretreatment with the α-adrenoreceptor blocker phentolamine significantly increased the PN size (p < 0.001; Fig. 4C). In contrast, NE administration with pretreatment of a β-adrenoceptor blocker (propranolol) did not induce such an increase in the PN size and the differences in PN size were significant (p < 0.05 and p < 0.01). It is noteworthy that in the Phen+NE group, the magnitude of the PN size increase was greater than that induced by the administration of NE alone (p < 0.05). The proportion of red-colored samples was not affected by NE administration. The proportion significantly increased in the Phen-NE group, whereas it decreased in the Prop-NE group. Collectively, these results indicate that β-adrenoceptors play a major role in mediating NE-induced hyperplastic morphological changes of the osPVS tissue, while α-adrenoceptors mediate a tonic suppressive effect on the hyperplastic changes of the osPVS tissue (Fig. 4C).

Figure 4. Effects of α- and β-adrenoceptor blockers on the morphology of osPVS tissue in the rats infused with norepinephrine. (A) Collection of isolated osPVS tissue samples from one rat in each of 4 groups in PBS solution. A1, Control; A2, NE (3 mg/kg/d, subcutaneous mini-osmotic pump); A3, Phen+NE (NE 3 mg/kg/d + phentolamine [α-adrenoceptor blocker], 20 mg/kg in drinking water); A4, Prop + NE (NE 3 mg/kg/d + propranolol [β -adrenoceptor blocker], 50 mg/kg in drinking water). Scale bar = 1 mm. (B-D) Summary bar graphs showing the mean number of osPVS samples per rat (B; t-test), the mean size of primo nodes (C; t-test), and the proportion of red-colored PNs (D; T-test). The number of rats in the control, NE, NE+Phen, and NE+Prop groups as 5, 7, 5, and 6, respectively. NE = norepinephrine; Phen = phentolamine; Prop = propranolol.

The morphological data obtained in this study indicate that the hyperplastic changes of osPVS tissue is under sympathetic control, which is mediated mostly by β-adrenoceptors in rats. Sympatholytic treatments such as ExT and 6-OHDA normalized the increased size and red-colored sample proportion of the osPVS tissues in the rats with HF (Fig. 1) and hemolytic anemia (Fig. 2). Administration of NE induced hyperplastic changes in the rats pretreated with a nonselective α-adrenoceptor blocker (phentolamine) but not in the rats pretreated with a non-selective β-adrenoceptor (propranolol) (Fig. 3 and 4). These findings provide novel experimental evidence for sympathetic regulation of the PVS.

The morphological changes of the osPVS tissue in the rat model of HF and HA are consistent with previous reports on the same tissue. The PN size increases more than 2-fold in HF (Fig. 1E of this study vs. Fig. 53.1e of [7]) and in HA (Fig. 2C of this study vs. Fig. 3k of [8]). The increase in the size of the osPVS tissue was associated with an increase in the number of erythrocytes [7,8,35]. Considering that the osPVS tissue is an organ of extra-marrow hematopoiesis [36], the red chromophore is an indication of erythropoiesis. These findings collectively indicate that the morphological changes in the osPVS tissue in this study constitute a good example of extramedullary hematopoietic hyperplasia in the spleen and liver in HF and HA [18,37].

The most salient finding in this study is that activation of the sympathetic nervous system causes/induces erythropoietic hyperplasia in osPVS tissue. ExT, which is known to reduce sympathetic tone [24,38], also normalized hyperplastic changes in the osPVS tissue in the rats with HF. This observation was further confirmed in the rat model of HA by using chemical sympathectomy with 6-OHDA. Further confirmation was obtained by mimicking similar hyperplastic changes of the osPVS tissue in the normal rats by administering NE through a mini-osmotic pump. The results are consistent with recent studies. Adrenergic activity mediates splenic myelopoiesis in mice under chronic social stress [39] and with diabetes [40]. In addition, splenic erythropoiesis is also considered as a stem cell-based tissue regeneration response [22].

In the present study, the NE-induced hyperplastic changes in the osPVS tissue were reduced to the control level under β-adrenoceptor blockade; however, the changes in response to α-adrenoceptor blockade were even larger than those induced by NE alone (Fig. 4). These results indicate that β-adrenoceptor is positively coupled to the hyperplastic changes, whereas α-adrenoceptor is negatively coupled. At an NE dose of 3 mg/kg/d (1.3 ng/ml in plasma), β-adrenoceptor-mediated action is predominant. This observation is in line with 2 reports showing that 1) proliferation of granulocyte macrophage progenitors and myelopoiesis were mediated by β2 adrenoceptors [40] and 2) social stress increased erythroid progenitor cells in the spleen, which was mediated by β-adrenoceptor activation [39].

In the bone marrow, adrenergic signals are mainly conveyed by β-adrenoceptors, while α1-adrenoceptors mediate inhibition of myelopoiesis [23]. Chronic stress has been found to increase the expression of myeloid transcription factors by a decline in lymphocyte precursors (myeloid bias) [41]. α1-agonists and β-blockers are known to have therapeutic potential for normalizing myeloid bias [23]. In Fig. 3 of this study, NE induced dose-dependent hyperplastic changes in the rats treated with 1 and 3 mg/kg/d, but those changes did not increase further at the dose of 10 mg/kg/d. At this high dose, the erythrocyte number and hemoglobin levels increased, whereas MCV and reticulocyte numbers decreased (Table 3). At present, it is difficult to interpret this observation. However, considering that reticulocytes are an index of erythropoiesis, decreased erythrocyte number is likely to be the result of inhibition of erythropoiesis by high NE in this group. A complex dose-response relationship is also reported in the bone marrow erythropoiesis. Penn et al. [27] showed that a normal level of NE is necessary for erythropoiesis, but supraphysiologic doses of exogenous NE inhibited normal erythropoiesis, which may explain the persistent anemia after severe traumatic injury [42]. Collectively, the adrenergic modulation of the osPVS tissue appears like that of myelopoiesis in the lymphoid tissues, the spleen and the bone marrow in general. The relative roles of α- and β-adrenoceptors in regulating hyperplastic changes in the osPVS tissue remain to be studied further.

The red chromophore in the osPVS tissue has been reported in previous studies [7,8], and it is known to originate from mature and immature erythrocytes in the tissue. The osPVS tissue contains hematopoietic progenitor cells and pluripotent stem cells that can differentiate into erythrocytes and other blood cells [36]. In this aspect, the proportion of red-colored tissue can be an index of erythroid lineage in hematopoiesis in osPVS tissue. In this study, the proportion of red-colored tissue increased in the rats with HF (Fig. 1) or HA (Fig. 2), but not in the healthy rats treated with NE (Fig. 3). It is likely that production of erythropoietin, an important factor for the erythroid lineage of hematopoiesis [43] is stimulated in hypoxic rats with HF or HA, but not in normoxic rats treated with NE [44]. Interestingly, the proportion of red-colored tissue was larger in the normoxic rats treated with NE plus phentolamine (an α-adrenoceptor blocker) than in the normoxic rats treated with NE alone (Fig. 4C). These observations also imply that erythropoiesis in the normoxic rats is under dual adrenergic control: stimulation via β-adrenoceptor activation and inhibition via α-adrenoceptor activation. These findings are very similar to the adrenergic regulation of hematopoiesis in the bone marrow discussed earlier [23].

In this study, 6-OHDA treatment reversed the PHZ-induced changes in erythrocytes, hemoglobin, MCV, MCHC, reticulocyte and WBC levels on day 6 after the treatment in the rats (Table 2). However, the PN size of the osPVS tissue in the 6-OHDA-PHZ group was significantly larger than that in the control group (Fig. 2C). This discrepancy suggests that although 6-OHDA ameliorated the effects of PHZ on the hematological parameters measured on day 6, the effects of PHZ were not fully prevented in the rats. Oxidative toxicity is the common mechanism of PHZ-induced hemolysis and phagocytosis [34,45,46] and 6-OHDA-induced sympathectomy [47]. The effects of PHZ appear in a day, peak during 3-6 days, and recover on day 10 and later after treatment [8], whereas the effects of 6-OHDA on blood catecholamine levels and blood cell count become significant at 7 days and 14 days, respectively [32]. We were not able to find any report stating that the oxidative toxicity was reduced by the interactions of 2 agents in the living body. Interestingly, it has been reported that catecholamines can reduce erythrocyte apoptosis [48] by blocking Ca²⁺-permeable non-selective cation channels in erythrocytes. Epinephrine (a catecholamine) has also been reported to reduce auto-oxidation and precipitation of hemoglobin chains via a mechanism that does not involve oxygen [49]. Therefore, considering that 6-OHDA is also a catecholamine, 6-OHDA is likely to counteract the effects of PHZ on erythrocytes via mechanisms other than oxidative toxicity.

Hyperplastic changes occurred in the osPVS tissue of normoxic and hypoxic animals (Fig. 1-4). This finding is in good agreement with those of Kim [3,7] on the morphology of the PVS tissue in the lymphatic vessel. Kim [3] reported 4 stages of morphological dynamics based on long-term observations of the PVS tissue. In the first 2 stages, the PN is opaque and milky-white, its boundary is not well distinguished, and cellular elements are not developed yet. In the subsequent 2 stages, the PNs present with the PV, the PN contour is well distinguished, and many lymphoid cells are found inside and outside of the PN of osPVS tissue, as reported previously in the osPVS tissue [9]. Collectively, the results suggest that hyperplastic morphological changes could be a common feature among different types of PVS tissues present in the lymphatic and blood vessels, organ surface, and nervous system in rats.

In conclusion, the results in this study showed that hyperplastic changes in the osPVS tissue are stimulated by β-adrenergic receptor activation but inhibited by α-adrenergic receptor activation in rats. The former effect is dominant at low to high levels of NE, but the latter effect becomes apparent at a very high level of NE. These findings indicate that the morphology of the osPVS tissue is under dual control of the sympathetic nervous system, and that the PVS is another organ capable of hyperplastic change, like the spleen, bone marrow, adrenal gland, and breasts, in physiological and pathological states of the body.

This study was supported by a grant from the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology to PDR (2018R1D1A1B07043448).

Conceptualization: Ryu PD; Data curation: Shen Y, Ryu PD; Formal analysis: Shen Y, Ryu PD; Funding acquisition: Ryu PD; Investigation: Shen Y, Kim YJ; Methodology: Shen Y, Kim YJ; Project administration: Ryu PD; Resources: Ryu PD; Supervision/Validation: Ryu PD; Writing (Original/review/editing): Ryu PD.

The authors declare no conflict of interest.