- Email: [email protected]
Changes in the Vipergic, Cholinergic and Adrenergic Innervation of Human Penile Tissue in Diabetic and Non-Diabetic Impotent Males
0022-534 7/87 /1375-1053$02.00/0 Vol. 137, May Printed in U.S.A.
THE JOURNAL OF UROLOGY
Copyright © 1987 by The Williams & Wilkins Co.
CHANGES IN THE VIPERGIC, CHOLINERGIC AND ADRENERGIC INNERVATION OF HUMAN PENILE TISSUE IN DIABETIC AND NONDIABETIC IMPOTENT MALES J. LINCOLN,
R CROWE, P. F_ BLACKLAY, J. P. PRYOR, J. S. P. LUMLEY
AND
G. BURNSTOCK*
From the Department of Anatomy and Embryology and Centre for Neuroscience, University College London, the Department of Surgery, St. Bartholomew's Hospital, and the Institute of Urology, The Shaftesbury Hospital, London, England
ABSTRACT
Penile tissue (consisting of corpus cavernosum and tunica albuginea) was obtained from 19 patients undergoing surgery for the implantation of penile prostheses. The tissue was examined for vasoactive intestinal polypeptide-like immunoreactivity in nerves, acetylcholinesterase-positive staining in nerves and noradrenaline content. Impotence was due to a variety of causes; 11 patients were classified as a 'non-neuropathic' group on the basis of their clinical history which included Peyronie' s disease, vascular disease, hypertension and psychogenic impotence. Vasoactive intestinal polypeptide-like immunoreactive and acetylcholinesterase-positive nerves were present and the pattern and distribution were similar in each patient in this group. The noradrenaline content of the tunica albuginea was significantly lower than the corpus cavernosum (p <0.02), although there was a linear relationship between the noradrenaline contents of the two regions (r = 0_95, p <0.01). By comparison, a complete absence of vasoactive intestinal polypeptide-like immunoreactivity in nerves was observed in a patient with a cauda equina lesion. Five out of six diabetic patients studied revealed a marked reduction in vasoactive intestinal polypeptide-like immunoreactivity in nerves associated with the cavernous smooth muscle, while acetylcholinesterase-positive staining was reduced in three out of five diabetic patients studied. The noradrenaline content of the corpus cavernosum from diabetic patients was significantly lower (p <0.02) than that of the 'nonneuropathic' group. The noradrenaline content of the tunica albuginea, however, was similar in both groups. The results provide evidence that VIPergic, cholinergic and adrenergic nerves in the penis are affected in diabetes mellitus and thus may contribute to the development of impotence in diabetic patients. Penile erection is a vascular phenomenon under nervous control. 1• 2 Initially, the parasympathetic nervous system was thought to be of primary importance in the control of erection, but psychogenic erections have been shown to be mediated via sympathetic pathways in the absence of sacral spinal cordmediated reflexogenic erections. 2 Recently, it has been recognized that it is likely that both portions of the autonomic nervous system function together to produce highly specific vasodilatation and vasoconstriction responses in the appropriate vasculature. 2 In addition, the contractile state of the smooth muscle of the trabeculae of the corpus cavernosum has been implicated to have a role in both erection and detumescence and this is also under autonomic nervous control. 3• 4 Apart from the contribution of adrenergic and cholinergic nerves in the control of penile vessels and erectile tissue, recent studies have provided evidence that vasoactive intestinal polypeptide (VIP) acts as a transmitter in nerves supplying the penile arteries4-6 and cavernous smooth muscle. 4 • 5 • 7 Normal penile erection in man depends on the integration of thoracolumbar and sacral afferent signals, intact autonomic nerves to the penis, normal erectile tissue and an adequate blood supply to the penis. 2 Abnormalities in any one of these areas can lead to erectile dysfunction. The prevalence of impotence in diabetic men has been reported to be as high as 50%. 8 Atherosclerotic changes to blood vessels are a complication of diabetes and several clinical studies have suggested that atherosclerosis is the major aetiol-
ogical factor involved in the development of impotence in diabetes. 9- 11 However, it has also been suggested that impotence in diabetic men is a manifestation of autonomic neuropathy in diabetes. 12- 14 Both the parasympathetic 15 and sympathetic 16• 17 innervation of the penis have been implicated. In a recent study from this laboratory a marked reduction was found in VIPimmunoreactivity in nerves of the penis from streptozotocindiabetic rats, and a similar reduction was found in a preliminary study of one human diabetic patient with impotence. 18 A reduction has also been reported in VIP-immunoreactivity and VIP content of the corpus cavernosum from impotent patients. However, the reduction was apparent in all impotent patients regardless of the cause of impotence. 19 In the present study, human penile tissue was obtained from a series of 19 patients undergoing surgery for the implantation of penile prostheses. The series consisted of patients suffering from impotence due to a variety of causes including diabetes, Peyronie's disease and vascular disease. Biochemical, histochemical and immunohistochemical techniques were used to investigate the adrenergic, cholinergic and VIPergic innervation of erectile tissue. The results in impotence of varying aetiology were compared. MATERIALS AND METHODS
Specimens of human penile tissue (containing both corpus cavernosum and tunica albuginea) were obtained from a series of 19 patients undergoing surgery for the implantation of penile Accepted for publication January 5, 1987. prostheses. The cause of erectile dysfunction was categorized * Requests for reprints: Dept. of Anatomy and Embryology and Centre for Neuroscience, University College London, Gower Street, on the following basis: clinical history, examination, nocturnal penile tumescence tests, the diagnostic use of intracavernosal London WClE 6BT, England. papaverine, psychiatric evaluation and additional evidence of Supported by a grant from the British Diabetic Association. 1053
1054
LINCOLN AND ASSOCIATES
neuropathy. Each tissue specimen was divided into several portions for analysis, which was carried out before clinical details of the patients were provided. Immunofluorescence staining for VIP was achieved using the indirect method. 20 Tissue was rapidly fixed within one hour of surgery in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), pH 7.1 to 7.4 for two hours at 4C, rinsed in PBS containing 7% sucrose and 0.01 % sodium azide, and left overnight in the same buffer. Details of the immunofluorescence histochemical staining for VIP have been described previously.18 Catecholamine fluorescence histochemistry was carried out using the glyoxylic acid technique. 21 The thiocholine method22 was used for the localization of acetylcholinesterase (AChE) activity. Butyrylcholinesterase activity was inhibited by preincubation of the tissue sections with tetraisopropylpyrophosphoramide (ISOMPA 10-5 M) for 0.5 hr. Sections were incubated with acetylthiocholine iodide for 1.5 hr. All the immunohistochemical and histochemical techniques were carried out on cryostat sections (10 µm.) cut using a cryostat at a cabinet temperature of -20C with a blade cooled by dry ice. The entire section was examined for the presence of VIPlike immunoreactivity, catecholamine fluorescence and positive AChE staining. The overall pattern of innervation in the tunica albuginea, cavernous smooth muscle and blood vessels was assessed on the basis of examination of 12 to 20 sections from each patient and the observations were divided into four broad categories: absent, sparse, intermediate and numerous. The tunica albuginea was dissected from the corpus cavernosum and both regions were frozen and stored in liquid nitrogen for the biochemical assay of noradrenaline. Levels of noradrenaline in the tissues were measured using high performance liquid chromatography with electrochemical detection. Samples were homogenized in 500 µI. of 0.1 M perchloric acid containing 0.4 mM sodium bisulphite and 12.5 ng. of an internal standard, dihydroxybenzylamine (DHBA). The extraction procedure, slightly modified by the addition of 0.1 mM ethylenediaminetetra-acetate (EDTA) in the solution used for washing the alumina, was that of Keller et al. 23 Chromatography was carried out at a flow rate of 1.5 ml./min. using a mobile phase consisting of 0.1 M sodium dihydrogen phosphate, 0.1 mM EDTA, 5 mM heptane sulphonate (pH 5.0) containing 10% (v/v) methanol on a µ-Bondapak C-18 reverse phase column. Detection was accomplished with a waxy carbon paste electrode set at a potential of +0.72 V. Quantitation was carried out by measuring the peak heights of noradrenaline and DHBA in samples and external standards. Noradrenaline levels in each sample were corrected for variations in the extraction and analytical procedures using-the measurements made of the internal standard DHBA. 24 RESULTS
Brief clinical details of the patients investigated in the present study are given in table 1. The patients have been divided into three groups according to their clinical history. Evidence of vascular or neurological disease is included, no patient had any endocrine disorder. Group I consists of 11 non-diabetic patients in whom damage to the autonomic nerve supply of the penis was not indicated clinically as a factor in the development of impotence. It includes patients where a non-neuropathic cause of impotence was diagnosed (Peyronie's disease, anti-hypertensive therapy and psychogenic) and patients where the possibility of a neuropathic cause existed but was considered unlikely since there was no evidence of abnormal bladder function (pelvic fracture, spina bifida and laminectomy). Group II consists of two non-diabetic patients. One patient had a cauda equina lesion and was shown to have abnormal bladder function (decreased urinary sensation). The second patient had undergone a cystectomy and therefore could not satisfy the criteria for Group I.
TABLE
1. Brief clinical details of the 19 patients investigated in the
present study Recent Drug History
Duration Age Impotence Group Patl· ent (Yrs.) (Yrs.)
II
III
1
64
3
2 3
58 57
3
4 5
53 58
2 8
6
30
7
59
10
8
56
9
9
45
10 11
33 61
20 1 2
12 13
45 52
2 3
NIL NIL
14
60
3
Glibenclamide
15
73
4
16
45
2
17
51
2
18
42
16
19
33
3
5
Mianserin Prochlorperazine NIL Hydrocortisone lndomethacin Azathioprine NIL NIL
NIL Never Potent
Clinical Details Peyronie's disease. Peyronie's disease. Peyronie's disease, rheumatoid arthritis. Peyronie's disease. Pelvic fracture, complicated urethral surgery, bladder function normal. Spina bifida, bladder and bowel function normal. Hypertension for 10 years.
Atenolol Lorazepam Cyclopenthiazide NIL Some evidence of vascular disease. Psychogenic. NIL NIL No firm diagnosis. NIL No firm diagnosis, laminectomy, bladder function normal. Cystectomy. Cauda equina lesion, reduced urinary sensation, reduced sensation left leg.
Non-insulin-dependent diabetes, no evidence of neuropathy. Glibenclamide Non-insulin-dependent diabetes, no evidence of neuropathy. Chlorpropamide Non-insulin-dependent diabetes, absent ankle and knee reflexes. Insulin Insulin-dependent diabetes, no evidence of neuropathy. Insulin Insulin-dependent diabetes, decreased peripheral sensation, postural hypotension. Insulin Insulin-dependent diabetes, postural hypotension, gustatory _s:l'\".e_ating,_diabetic_ diarrhoea, sensory neuropathy.
Group III consists of all the patients suffering from diabetes mellitus (n = 6; three non-insulin-dependent, three insulindependent) whether or not they showed clinical evidence of peripheral and/or autonomic neuropathy. The mean ages of the three groups were similar: Group I 52 ± 3.4 years (± SEM, n = 11), Group II 49 ± 3.5 years (n = 2), Group III 51 ± 5.8 years (n = 6). A summary of the results obtained from the immunohistochemical, histochemical and biochemical investigation of tissue from each patient is given in table 2. In certain cases insufficient tissue was available to perform all the techniques described. Immunohistochemical study of VIP. The pattern of distribution of VIP-immunoreactive nerves in the corpus cavernosum (CC), tunica albuginea (TA) and blood vessels (BV) within the erectile tissue from the present series of patients has been summarized in table 2. There was some variation in the pattern of innervation in different sections from individual patients. The results given in table 2 are the average assessment on examination of 12 to 20 sections from each patient. Micro-
AUTONOMIC INNERVATION OF PENIS IN DIABETIC IMPOTENT MALES
1055
TABLE 2. A summary of the results obtained from the immunohistochemical, histochemical and biochemical studies of 19 impotent patients
Patient
VIP-Like Immunoreactive Nerves
cc
TA
+++ +++ +++
+ + +
++ ++ ++
+++ +++ ++ +++
+ + + +
++ ++ ++ +
+
++
11
+++ ++
12 13
0
0
0
+++
+
+++
1 2 3 4 5 6 7 8 9 10
14 15 16 17 18 19
BV
0
0
0
+ +
0
+
+
0
0
0 0
+
+
AChE-Positive Nerves
Noradrenaline Content µ,g./gm.
tissue
cc
TA
BV
cc
+++
+
++
+++ ++ +++ +++ +++ +++ +++
+ + + + + + +
++ + ++ ++ ++ + +
0.34 0.52 0.45 0.41
0.33
0.37 0.21
0.19 0.03
0.22 0.33
0.04 0.17
+++
0
+++
+
++
0.19 0.20
0.09 1.35
+++ ++ + + +
++
+++ ++ + + +++
0.33 0.27 0.12
0.12 0.08 0.15
0.12 0.03
0.14 0.22
+ + 0
TA 0.33
Numbered according to the chart in table 1. 0 = absent, + = sparse, ++ = intermediate, +++ = numerous VIP-like immunoreactive or AChE-positive nerves (CC, corpus cavernosum; TA, tunica albuginea; BV, blood vessels).
graphs illustrating the pattern of innervation were selected on the basis that they were representative of the average assessment made. In Group I patients the tunica albuginea revealed only sparse innervation by VIP-like immunoreactive nerves which were localized mainly in nerve bundles (fig. IA). Moderate to sparse patterns of VIP-like immunoreactive nerves were also observed on the adventitial-medial border of some blood vessels within the erectile tissue from Group I patients. The highest density of VIP-like immunoreactive nerves was found in association with the cavernous smooth muscle (fig. lB). The patient with a cauda equina lesion was the only Group II patient investigated for VIP-like immunoreactivity. In this patient no VIP-like immunoreactivity was observed in nerves associated with the cavernous smooth muscle, tunica albuginea or blood vessels. Five of the diabetic Group III patients, three insulin-dependent and two non-insulin-dependent, revealed a marked reduction in VIP-like immunoreactive nerves when compared with Group I patients. The reduction was observed in the tunica albuginea and blood vessels but was most marked in the smooth muscle of the corpus cavernosum (fig. lC). The VIP-like immunoreactivity in the corpus cavernosum and tunica albuginea from one non-insulin-dependent diabetic, however, was indistinguishable from the Group I patients and additionally revealed the most dense innervation of blood vessels by VIP-like immunoreactive nerves of any of the patients studied. Histochemical study of AChE. In general, the pattern of distribution of positive AChE staining was similar to that described for VIP-like immunoreactivity (see table 2). Thus in Group I there were numerous AChE-positive nerve fibres in the cavernous smooth muscle (fig. 2A), while the tunica albuginea revealed sparse AChE-positive staining. The blood vessels of the corpus cavernosum revealed sparse to moderate innervation by AChE-positive nerve fibres. The Group II patient who had undergone a cystectomy gave similar results to those from Group I patients. Three Group III diabetic patients revealed a marked reduction in AChE-positive nerve fibres in the corpus cavernosum (fig. 2B). The AChE-positive staining of the corpus cavernosum
Fm. 1. VIP immunohistochemistry. A, arrow indicates fluorescent nerve fibres in nerve bundle within tunica albuginea from Group I patient. B, arrows indicate fluorescent nerve fibres in corpus cavernosum from Group I patient. C, marked reduction in VIP-like immunoreactivity in corpus cavernosum from Group III diabetic patient. Calibration bars = 80 µm.
from two non-insulin-dependent diabetics without clinical evidence of neuropathy was similar to that of Group I patients. There was no marked difference in the histological appearance of the tissues from any of the patients studied, as judged by routine haematoxylin and eosin staining. Examples of the tunica albuginea and corpus cavernosum are given in figs. 2C and 2D, respectively. Biochemical assay of noradrenaline. Adrenergic nerves could not be visualized using the glyoxylic acid technique in any of the patients studied due to high background levels of autofluorescence in the tissue specimens. A similar finding has been reported previously. 17• 25 The adrenergic innervation was therefore investigated by biochemical measurement of noradrenaline levels in the corpus cavernosum and tunica albuginea and the results have been given in table 2. The mean noradrenaline content of the corpus cavernosum from Group I patients was 0.36 ± 0.038 µg./gm. tissue (± SEM, n = 8) which was significantly higher (p <0.02, Student's t test) than the noradrenaline content of the tunica albuginea (0.18 ± 0.054 µg./gm. tissue, n = 6). There was some patient to patient variability in the noradrenaline content of both regions; how-
1056
LINCOLN AND ASSOCIATES
"
Fm. 2. A, arrows indicate AChE-positive nerve fibres in corpus cavernosum from Group I patient. B, marked reduction in AChE-positive nerve fibres in corpus cavernosum from Group III diabetic patient. Arrow indicates blood vessel that is negative for AChE staining. C, haemotoxylin and eosin staining of tunica albuginea. D, haematoxylin and eosin staining of corpus cavernosum. Calibration bars = 80 µm.
ever, it was observed that where the noradrenaline content of the tunica albuginea was low the level in the corpus cavernosum also tended to be low. In six Group I patients sufficient material was available for both regions to be assayed for noradrenaline. When the noradrenaline content of the corpus cavernosum was plotted against the noradrenaline content of the tunica albuginea from these six patients, the relationship was found to be linear by regression line analysis (r = 0.95, n = 6, p <0.01) and the equation for the line was calculated. The results have been shown in fig. 3 where each symbol represents one patient and is numbered according to the chart in table 1. Confidence limits for the line at the 95% level have also been indicated on the figure. The Group II patient who had undergone a cystectomy gave noradrenaline values which were marginally outside the 95% confidence limits of the line. The patient with a cauda equina
lesion had the highest noradrenaline content in the tunica albuginea of any of the patients studied. The mean noradrenaline content of the corpus cavernosum from Group III diabetic patients was 0.17 ± 0.055 µg.jgm. tissue (n = 5) which was significantly lower (p <0.02) than that of the Group I patients. There was no significant difference, however, between the mean noradrenaline level of the tunica albuginea from Group I patients and that from Group III diabetic patients (0.14 ± 0.023 µg.jgm. tissue; n = 5). The results from the Group III diabetic patients were also assessed individually by the correlation between the noradrenaline content of the corpus cavernosum and of the tunica albuginea (fig. 3). In contrast to Group I, three Group III diabetic patients with clinical evidence of neuropathy had lower levels of noradrenaline in the corpus cavernosum than the tunica albuginea and thus the values were markedly outside the
1057 (1)
:, E :J
"'
(J)
06
values
0
C:
'>ro© 0-5 rn
01
:i.U
t:
Q)
which other groups could be com-
(j)
(/)
::,
c0 a.0 OU
o4 0-4
06
z
10/ 03 15 ..
0-2
7
13@
.. ·;.,12
17 AA16 0-1
0-1
0-2
0-3
0-4
If-
1-3
1-4
NA Content µg/g tissue Tunica Albuginea
FIG. 3. Relationship between noradrenaline content of corpus cavemosum and tunica albuginea. Each symbol represents single patient and is numbered according to chart in table 1. 0 = Group I patients. Relationship between noradrenaline contents of two regions was linear (r = 0.95, p <0.01). Dotted lines indicate 95% confidence intervals of line. it = Group II patients. A = Group HI, diabetic patients.
95% confidence limits of the line (fig. 3). The noradrenaline levels from two non-insulin-dependent diabetics without evidence of neuropathy, however, were within the range predicted from Group I patients. DISCUSSION
In the present study all the patients investigated suffered from impotence, and it was possible to compare the pattern of autonomic innervation of the penis in impotence due to a variety of causes. Studies of this kind are faced with considerable difficulties with regard to the availability of both suitable control material and large patient numbers. In ~~.Mn,rn,,~ studies, control penile tissue was obtained from transsexuals and carcinoma patients undergoing surgery. 16 · 17• 25 · 26 However, on occasion, patients with Peyronie's disease have been used as controls in combination with patients tomy. 19 · 26 It has been suggested that the fibrous tissue, in Peyronie's disease, interferes wc,u,.o.un,c,uy with the normal function of the cu,uu,u,;v system. 27 In an electron m,~P,,O~An,~ abnormalities were observed in the within the fibrous plaque while the unmyehnated axons of the surrounding erectile tissue appeared unchanged, thus there was no ultrastructural evidence to suggest that the autonomic nervation of the penis was affected in Peyronie's disease. 28 In the present investigation, four patients with Peyronie's disease were combined with seven other patients in whom neuropathy was not present clinically nor was there any disturbance in bladder or bowel function. It was realized that the presence of normal bladder function, in cases where doubt occurred, did not totally exclude the possibility that neuropathic changes had occurred in the penis. However, since the penis and bladder share a common nerve supply, 12 • 13 • 29 this was considered the most appropriate criterion for establishing Group I. This view was strengthened by the consistent results obtained for Group I patients despite the differing causes of impotence and was particularly encouraging in the biochemical measurements of noradrenaline levels where more precise quantitation could be achieved. The results from Group I patients were therefore considered to form a baseline of 'non-
A detailed discussion on the autonomic innervation of the penis in Group II patients is beyond the scope of the present the main aim of which was to investigate changes occurring in impotent patients with diabetes mellitus. A recent immunohistochemical of the cat spinal cord revealed that VIP is localized primarily in the sacral segments that supply the urinary tract, large intestine and sex organs" 30 In the present series of patients, the patient with a cauda equina lesion revealed a complete absence of VIP-like immunoreactivity in nerves of the cavernous smooth muscle, tunica albuginea and blood vessels" Nerve damage due to cystectomy has been recognized as an iatrogenic cause of impotence" 29 While the noradrenaline levels of the cystectomy patient were marginally outside Group I values, the pattern of AChE-positive nerves was indistinguishable from Group L At this stage it is not possible to conclude whether impotence in this patient was due to damage to nerves, blood vessels or psychological factors following radical surgery" A linear relationship was shown to exist between the noradrenaline levels of the corpus cavernosum and tunica albuginea from Group I patients, which has not been published before, and represents a new and useful approach for future studies of this kind. The noradrenaline levels in control penile tissue, reported in a previous study, 16 were very variable, possibly due to regional variation in tissue samples obtained from penectomies. The presence of large errors for mean control noradren aline content can prevent the detection of subtle changes in other groups. While patient to patient variability was less marked in the present study, it was still advantageous to examine the relationship between the two regions. It was thus possible to compare the individual results obtained for each Group II and Group III patient with the entire range of values in Group I, as well as to analyze the differences between the mean values for each group, The justification for examining the relationship between the two regions lies in the fact that it has been shown anatomically that the nerve supply to the corpus cavernosum crosses the tunica albuginea to innervate the cavernous smooth muscle. 29 Thus the nerve bundles, largely localized in the tunica albuginea, and the single nerve fibres of the corpus cavernosum could be regarded as different regions of the same nerve supply. If damage occurs to the sympathetic nerves then it would be predicted that a decrease in noradrenaline content would occur at the site of the terminal varicosities in the corpus cavernosum rather than in the nerve trunks of the tunica albuginea. Therefore - .. of the -~,."'-""!,'),'::c~c'::.!v between the noradrenaline levels of also have the advantage of an unu~u.Hvu nerve damage as opposed to In the present one of the and had received a (3antagonist has been shown to decrease sym pathetic nerve activity. 31 The noradrenaline levels of this patient were the lowest of any of the Group I patients; however, the relationship between the two regions was maintained. In contrast, three diabetic patients revealed a significant, preferential decrease in the noradrenaline content of the corpus cavernosum relative to the tunica albuginea which suggests that damage to the sympathetic nerves of the erectile tissue had probably occurred in these patients. A significant reduction in noradrenaline levels in erectile tissue in diabetes-associated impotence has been reported previously although levels in the tunica albuginea were not measured, 16· 17 Unlike the present study, the reduction was not found in non-insulin-dependent diabetics. Results from three noninsulin-dependent diabetics have been reported here. The patient who presented with clinical symptoms of neuropathy gave values which were significantly outside the range for Group I while those of the two patients with no clinical indication of - q ~w
1058
LINCOLN AND ASSOCIATES
neuropathy were similar to the 'non-neuropathic' group. Clinical details of the patients in the previous studies were not given. Therefore, it cannot be excluded that non-insulin-dependent diabetics with neuropathy may be as susceptible to sympathetic nerve dysfunction in the penis as insulin-dependent diabetics. The changes in histochemical staining for AChE in erectile tissue in diabetes, observed in the present study, have not been reported previously. It should be noted that AChE staining is not entirely specific for cholinergic nerves. For example, AChE has been shown to hydrolyze substance P and, in the dorsal horn of the spinal cord, the distribution of AChE staining correlates well with substance P-like immunoreactivity. 32 Thus the reduction in AChE staining in the diabetic group of patients cannot provide conclusive evidence of a reduction in cholinergic nerves alone. Attempts were made to provide additional information on the cholinergic innervation by the biochemical measurement of choline acetyltransferase activity. Levels of activity proved to be too low to be Il).easured (unpublished observations), a finding which has also been reported previously. 25 The involvement of the parasympathetic innervation of the penis in diabetic impotence, however, has been suggested in clinical studies 13• 15• 33 and the results reported here would tend to support this view. The pattern and distribution of VIP-like immunoreactive nerves resembled that of AChE-positive nerves in the erectile tissue specimens investigated in the present study. It has been suggested that both VIP and AChE may be localized in the same nerve fibres in the human male genital tract. 7 However, it is also recognized that coexistence cannot be assumed from finding similar distributions at the light microscopical level alone. At this stage it is not possible to conclude whether the observation of similar reductions in both VIP-like immunoreactivity and AChE-positive staining in certain diabetic patients signifies changes in a single or two different nerve types. A reduction in VIP-like immunoreactive nerves in penile tissue was reported in a preliminary study from this laboratory on one diabetic patient when compared with three patients with Peyronie's disease. 18 The results of the present investigation revealed a marked reduction in VIP-like immunoreactivity in the corpus cavernosum from five out of six impotent diabetic patients when compared with nine impotent patients who had been classified as 'non-neuropathic'. In contrast, 1.t has been reported that levels of VIP and VIP-like immunoreactivity were similar in impotent patients regardless of the cause of impotence and that they were all reduced when compared with control material from transsexual patients undergoing penectomy.111Thepresent series of patients provides no information on the autonomic innervation of the penis in normal potent males, due to the obvious lack of available tissue, although certain patients (with Peyronie's disease and psychogenic impotence) retained some erectile function. Some caution may also be necessary before assuming that levels in transsexuals represent normal values due to the intensive drug therapy these patients undergo prior to surgery. The aim of the present study was to investigate whether different causes of impotence could be distinguished by the changes in the pattern of autonomic innervation that the erectile tissue exhibited. The results reported here indicate that this was still possible despite the lack of unequivocal control material. The present study does not exclude the possibility that vascular changes also contribute to the development of impotence in diabetes. Furthermore, psychological factors in impotence are as important in diabetics as in the non-diabetic population. Several clinical studies have investigated whether the neuropathic or the vascular changes associated with diabetes are of greater significance to the development of impotence, with conflicting results. Vascular occlusion has been indicated as the major contributing factor by some workers9- 11 while others have found a greater association with neuropathy. 8 • 12• 14
In the single study performed on post mortem tissue, morphological changes were observed in nerves in the corpus cavernosum from impotent diabetics while no evidence was found of microangiopathy.12 The conflict may in part be explained by the variety of clinical procedures used to assess the neurological status of patients and the fact that few of these procedures can reliably indicate the presence or lack of local involvement of the autonomic nerves of the penis in diabetes, which requires more direct study. This point has been demonstrated in a recent clinical study in which the conduction velocity of the dorsal nerve of the penis was shown to be significantly lower than controls in diabetics with impotence while the bulbocavernous reflex and pudendal evoked potential were within the normal range. 34 It is of relevance to this argument that, in the present study, a reduction in VIP-like immunoreactivity and AChEpositive staining in nerves was observed in penile tissue from an insulin-dependent diabetic whose neurological status had been assessed as normal. The autonomic nervous control of erection is complex. It has been reported that adrenergic, 2• 3 • 26 cholinergic 2• 35 and VIPergic4- 7 nerves in the penile vessels and cavernous smooth muscle, may all have significant roles in the process of erection and detumescence. Few studies have attempted direct examination of penile tissue in diabetic patients. The present study consists of an immunohistochemical, histochemical and biochemical investigation of the VIPergic, cholinergic and adrenergic innervation in penile tissue from patients with impotence due to a variety of causes. The results provide evidence that all three nerve types may be affected in diabetes mellitus and thus may contribute to the development of impotence in diabetic males. REFERENCES
1. de Groat, W. C. and Booth, A. M.: Physiology of male sexual function. Ann. Intern. Med., 92: 329, 1980. 2. Weiss, H. D.: The physiology of human penile erection. Ann. Intern. Med., 76: 793, 1972. 3. Benson, G. S., McConnell, J., Lipschultz, L. I., Corriere, J. N. Jr. and Wood, J.: Neuromorphology and neuropharmacology of the human penis. J. Clin. Invest., 65: 506, 1980. 4. Willis, E., Ottesen, B., Wagner, G., Sundler, F. and Fahrenkrug, J.: Vasoactive intestinal polypeptide (VIP) as a possible neurotransmitter involved in penile erection. Acta Physiol. Scand., 113: 545, 1981. 5. Polak, J.M., Mina, S., Gu, J. and Bloom, S. R.: VIPergic nerves-in the penis. Lancet, 2: 217, 1981. 6. Andersson,-P.-0.,-Bfoom,-S. R. andMellandex,S.:Haemodynamics of pelvic nerve induced penile erection in the dog: possible mediation by vasoactive intestinal polypeptide. J. Physiol. (Lond.), 350: 209, 1984. 7. Gu, J., Polak, J. M., Probert, L., Islam, K. N., Marangos, P. J., Mina, S., Adrian, T. E., McGregor, G. P., O'Shaughnessy, D. J. and Bloom, S. R.: Peptidergic innervation of the human male genital tract. J. Urol., 130: 386, 1983. 8. Kolodny, R. C., Kahn, C. B., Goldstein, H. H. and Barrett, D. M.: Sexual dysfunction in diabetic men. Diabetes, 23: 306, 1974. 9. Herman, A., Adar, R. and Rubinstein, Z.: Vascular lesions associated with impotence in diabetic and non-diabetic arterial occlusive disease. Diabetes, 27: 975, 1978. 10. Jevtich, M. J., Edson, M., Jarman, W. D. and Herrera, H. H.: Vascular factor in erectile failure among diabetics. Urology, 19: 163, 1982. 11. Lehman, T. P. and Jacobs, J. A.: Etiology of diabetic impotence. J. Urol., 129: 291, 1982. 12. Faerman, I., Glocer, L., Fox, D., Jadzinsky, M. N. and Rapaport, M.: Impotence and diabetes: histological studies of autonomic nervous fibers of the corpora cavernosa in impotent diabetic males. Diabetes, 23: 971, 1974. 13. Ellenberg, M.: Sexual function in diabetic patients. Ann. Intern. Med., 92: 331, 1980. 14. Karacan, I.: Diagnosis of erectile impotence in diabetes mellitus. Ann. Intern. Med., 92: 334, 1980.
AlJT01'10Iv'.iIC IfTNEl'-lVATIOJ\f Ol? P:El\IIS IN Dil-\BE'TIC Ih-'.e:POTEI\J'T lVIALES
15. 16.
17. 18.
19.
20. 21.
22. 23. 24.
M.: Impotence in diabetes: the neuroiogic factor. Ann. Intern. 75~ 213, 1971. Melman, A., Henry, D. P. and Felton, D. L.: Catecholamine content of the peniie corpora in patients with diabetes-associated impotence. Surg. Forum, 29: 634, 1978. Melman, A., Henry, D. P., Felten, D. L. and O'Connol', B.: Effect of diabetes upon penile sympathetic nerves in impotent patients. South. Med. J., 73: 307, 1980. Crowe, R., Lincoln, J., Blacklay, P. F., Pryor, J. P., Lumley, J. 8. P. and Burnstock, G.: Vasoactive intestinal polypeptidelike immunoreactive nerves in diabetic penis. Diabetes, 32: 1075, 1983. Gu, J., Polak, J. M., Lazarides, M., Morgan, R., Pryor, J. P., Marangos, P. J., Blank, M. A. and Bloom, S. R.: Decrease of vasoactive intestinal polypeptide (VIP) in the penises from impotent men. Lancet, 2: 315, 1984. Coons, A. H.: Fluorescent antibody methods. In: General Cytochemical Methods. Edited by J. F. Danielle. New York: Academic Press, pp. 399-422, 1958. Lindvall, 0. and Bjorklund, A.: The glyoxylic acid fluorescence histochemical method: a detailed account of the methodology for the visualisation of central catecholamine neurons. Histochemistry, 39: 97, 1974. Karnovsky, M. J. and Roots, L.: A 'direct coloring' thiocholine method for cholinesterases. J. Histochem. Cytochem., 12: 219, 1964. Keller, R., Oke, A., Mefford, I. and Adams, R. N.: Liquid chromatographic analysis of catecholamines: routine assay for :regional brain mapping. Life Sci., 19: 995, 1976. Moyer, To P. and Jiang, N.-S.: Optimized isocratic conditions for analysis of catecholamines by high-performance reversed-phase paired-ion chromatography with amperometric detection. J.
25. 1\!Ielman, LI!."""' Felten) D. L. and B. L.: Alteration of the penile corpora in patients with impo1980. tence. Invest. Urol., 17: 26. Meiman, A. and Hemy, D.: possible role of the catecholamines of the corpora in penile erection. J. Urol., 121: 1979. 27. Metz, P., Ebbehoj, J., Uhrenholdt, A. and Wagner, · Peyronie's disease and erectile failure. J. Urol., 130: 1103, 1983. 28. Vande Berg, J. S., Devine, C. J., Horton, C. E., Somers, K Do, Wright, G. L Jr., Leffel!, M. S., Dawson, D. M., G!eischman, S. H. and Rowe, M. J.: Peyronie's disease: an electron microscopic study. J. Urol., 126: 333, 1981. 29. Lue, T. F., Zeineh, S. J., Schmidt, R. A. and Tanagho, E. A.: Neuroanatomy of penile erection: its :relevance to iatrogenic impotence. J. Urol., 131: 273, 1984. 30. Kawatani, M., Lowe, I. P., Nadelhaft, I., Morgan, C. and de Groat, W. C.: Vasoactive intestinal polypeptide in visceral afferent pathways to the sacral spinal cord of the cat. Neurosci. Lett., 42: 311, 1983. 31. Chernow, B., Zaloga, G. P., Lake, C R., Coleman, M. D. and Ziegler, M. G.: Effect of antihypertensive therapy on sympathetic nervous system activity in patients with essential hype1tension. Fed. Proc., 43: 72, 1984. 32. Chubb, L W., Hodgson, A. J. and White, G. H.: Acetylcholinesterase hydrolyzes substance P. Neuroscience, 5: 2065, 1980. 33. Clarke, B. F., Ewing, D. J. and Campbell, I. W.: Diabetic autonomic neuropathy. Diabetologia, 17: 195, 1979. 34. Lin, J. T. and Bradley, W. E.: Penile neuropathy in insulindependent diabetes mellitus. J. 133: 213, 1985. 35. McConnell, J., Benson, G. S. and J.: Autonomic innervation of the mammalian penis: a histochemical and physiological study. J. Neural Transm., 45: 227, 1979.
THE JOURNAL OF UROLOGY
Copyright © 1987 by The Williams & Wilkins Co.
CHANGES IN THE VIPERGIC, CHOLINERGIC AND ADRENERGIC INNERVATION OF HUMAN PENILE TISSUE IN DIABETIC AND NONDIABETIC IMPOTENT MALES J. LINCOLN,
R CROWE, P. F_ BLACKLAY, J. P. PRYOR, J. S. P. LUMLEY
AND
G. BURNSTOCK*
From the Department of Anatomy and Embryology and Centre for Neuroscience, University College London, the Department of Surgery, St. Bartholomew's Hospital, and the Institute of Urology, The Shaftesbury Hospital, London, England
ABSTRACT
Penile tissue (consisting of corpus cavernosum and tunica albuginea) was obtained from 19 patients undergoing surgery for the implantation of penile prostheses. The tissue was examined for vasoactive intestinal polypeptide-like immunoreactivity in nerves, acetylcholinesterase-positive staining in nerves and noradrenaline content. Impotence was due to a variety of causes; 11 patients were classified as a 'non-neuropathic' group on the basis of their clinical history which included Peyronie' s disease, vascular disease, hypertension and psychogenic impotence. Vasoactive intestinal polypeptide-like immunoreactive and acetylcholinesterase-positive nerves were present and the pattern and distribution were similar in each patient in this group. The noradrenaline content of the tunica albuginea was significantly lower than the corpus cavernosum (p <0.02), although there was a linear relationship between the noradrenaline contents of the two regions (r = 0_95, p <0.01). By comparison, a complete absence of vasoactive intestinal polypeptide-like immunoreactivity in nerves was observed in a patient with a cauda equina lesion. Five out of six diabetic patients studied revealed a marked reduction in vasoactive intestinal polypeptide-like immunoreactivity in nerves associated with the cavernous smooth muscle, while acetylcholinesterase-positive staining was reduced in three out of five diabetic patients studied. The noradrenaline content of the corpus cavernosum from diabetic patients was significantly lower (p <0.02) than that of the 'nonneuropathic' group. The noradrenaline content of the tunica albuginea, however, was similar in both groups. The results provide evidence that VIPergic, cholinergic and adrenergic nerves in the penis are affected in diabetes mellitus and thus may contribute to the development of impotence in diabetic patients. Penile erection is a vascular phenomenon under nervous control. 1• 2 Initially, the parasympathetic nervous system was thought to be of primary importance in the control of erection, but psychogenic erections have been shown to be mediated via sympathetic pathways in the absence of sacral spinal cordmediated reflexogenic erections. 2 Recently, it has been recognized that it is likely that both portions of the autonomic nervous system function together to produce highly specific vasodilatation and vasoconstriction responses in the appropriate vasculature. 2 In addition, the contractile state of the smooth muscle of the trabeculae of the corpus cavernosum has been implicated to have a role in both erection and detumescence and this is also under autonomic nervous control. 3• 4 Apart from the contribution of adrenergic and cholinergic nerves in the control of penile vessels and erectile tissue, recent studies have provided evidence that vasoactive intestinal polypeptide (VIP) acts as a transmitter in nerves supplying the penile arteries4-6 and cavernous smooth muscle. 4 • 5 • 7 Normal penile erection in man depends on the integration of thoracolumbar and sacral afferent signals, intact autonomic nerves to the penis, normal erectile tissue and an adequate blood supply to the penis. 2 Abnormalities in any one of these areas can lead to erectile dysfunction. The prevalence of impotence in diabetic men has been reported to be as high as 50%. 8 Atherosclerotic changes to blood vessels are a complication of diabetes and several clinical studies have suggested that atherosclerosis is the major aetiol-
ogical factor involved in the development of impotence in diabetes. 9- 11 However, it has also been suggested that impotence in diabetic men is a manifestation of autonomic neuropathy in diabetes. 12- 14 Both the parasympathetic 15 and sympathetic 16• 17 innervation of the penis have been implicated. In a recent study from this laboratory a marked reduction was found in VIPimmunoreactivity in nerves of the penis from streptozotocindiabetic rats, and a similar reduction was found in a preliminary study of one human diabetic patient with impotence. 18 A reduction has also been reported in VIP-immunoreactivity and VIP content of the corpus cavernosum from impotent patients. However, the reduction was apparent in all impotent patients regardless of the cause of impotence. 19 In the present study, human penile tissue was obtained from a series of 19 patients undergoing surgery for the implantation of penile prostheses. The series consisted of patients suffering from impotence due to a variety of causes including diabetes, Peyronie's disease and vascular disease. Biochemical, histochemical and immunohistochemical techniques were used to investigate the adrenergic, cholinergic and VIPergic innervation of erectile tissue. The results in impotence of varying aetiology were compared. MATERIALS AND METHODS
Specimens of human penile tissue (containing both corpus cavernosum and tunica albuginea) were obtained from a series of 19 patients undergoing surgery for the implantation of penile Accepted for publication January 5, 1987. prostheses. The cause of erectile dysfunction was categorized * Requests for reprints: Dept. of Anatomy and Embryology and Centre for Neuroscience, University College London, Gower Street, on the following basis: clinical history, examination, nocturnal penile tumescence tests, the diagnostic use of intracavernosal London WClE 6BT, England. papaverine, psychiatric evaluation and additional evidence of Supported by a grant from the British Diabetic Association. 1053
1054
LINCOLN AND ASSOCIATES
neuropathy. Each tissue specimen was divided into several portions for analysis, which was carried out before clinical details of the patients were provided. Immunofluorescence staining for VIP was achieved using the indirect method. 20 Tissue was rapidly fixed within one hour of surgery in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), pH 7.1 to 7.4 for two hours at 4C, rinsed in PBS containing 7% sucrose and 0.01 % sodium azide, and left overnight in the same buffer. Details of the immunofluorescence histochemical staining for VIP have been described previously.18 Catecholamine fluorescence histochemistry was carried out using the glyoxylic acid technique. 21 The thiocholine method22 was used for the localization of acetylcholinesterase (AChE) activity. Butyrylcholinesterase activity was inhibited by preincubation of the tissue sections with tetraisopropylpyrophosphoramide (ISOMPA 10-5 M) for 0.5 hr. Sections were incubated with acetylthiocholine iodide for 1.5 hr. All the immunohistochemical and histochemical techniques were carried out on cryostat sections (10 µm.) cut using a cryostat at a cabinet temperature of -20C with a blade cooled by dry ice. The entire section was examined for the presence of VIPlike immunoreactivity, catecholamine fluorescence and positive AChE staining. The overall pattern of innervation in the tunica albuginea, cavernous smooth muscle and blood vessels was assessed on the basis of examination of 12 to 20 sections from each patient and the observations were divided into four broad categories: absent, sparse, intermediate and numerous. The tunica albuginea was dissected from the corpus cavernosum and both regions were frozen and stored in liquid nitrogen for the biochemical assay of noradrenaline. Levels of noradrenaline in the tissues were measured using high performance liquid chromatography with electrochemical detection. Samples were homogenized in 500 µI. of 0.1 M perchloric acid containing 0.4 mM sodium bisulphite and 12.5 ng. of an internal standard, dihydroxybenzylamine (DHBA). The extraction procedure, slightly modified by the addition of 0.1 mM ethylenediaminetetra-acetate (EDTA) in the solution used for washing the alumina, was that of Keller et al. 23 Chromatography was carried out at a flow rate of 1.5 ml./min. using a mobile phase consisting of 0.1 M sodium dihydrogen phosphate, 0.1 mM EDTA, 5 mM heptane sulphonate (pH 5.0) containing 10% (v/v) methanol on a µ-Bondapak C-18 reverse phase column. Detection was accomplished with a waxy carbon paste electrode set at a potential of +0.72 V. Quantitation was carried out by measuring the peak heights of noradrenaline and DHBA in samples and external standards. Noradrenaline levels in each sample were corrected for variations in the extraction and analytical procedures using-the measurements made of the internal standard DHBA. 24 RESULTS
Brief clinical details of the patients investigated in the present study are given in table 1. The patients have been divided into three groups according to their clinical history. Evidence of vascular or neurological disease is included, no patient had any endocrine disorder. Group I consists of 11 non-diabetic patients in whom damage to the autonomic nerve supply of the penis was not indicated clinically as a factor in the development of impotence. It includes patients where a non-neuropathic cause of impotence was diagnosed (Peyronie's disease, anti-hypertensive therapy and psychogenic) and patients where the possibility of a neuropathic cause existed but was considered unlikely since there was no evidence of abnormal bladder function (pelvic fracture, spina bifida and laminectomy). Group II consists of two non-diabetic patients. One patient had a cauda equina lesion and was shown to have abnormal bladder function (decreased urinary sensation). The second patient had undergone a cystectomy and therefore could not satisfy the criteria for Group I.
TABLE
1. Brief clinical details of the 19 patients investigated in the
present study Recent Drug History
Duration Age Impotence Group Patl· ent (Yrs.) (Yrs.)
II
III
1
64
3
2 3
58 57
3
4 5
53 58
2 8
6
30
7
59
10
8
56
9
9
45
10 11
33 61
20 1 2
12 13
45 52
2 3
NIL NIL
14
60
3
Glibenclamide
15
73
4
16
45
2
17
51
2
18
42
16
19
33
3
5
Mianserin Prochlorperazine NIL Hydrocortisone lndomethacin Azathioprine NIL NIL
NIL Never Potent
Clinical Details Peyronie's disease. Peyronie's disease. Peyronie's disease, rheumatoid arthritis. Peyronie's disease. Pelvic fracture, complicated urethral surgery, bladder function normal. Spina bifida, bladder and bowel function normal. Hypertension for 10 years.
Atenolol Lorazepam Cyclopenthiazide NIL Some evidence of vascular disease. Psychogenic. NIL NIL No firm diagnosis. NIL No firm diagnosis, laminectomy, bladder function normal. Cystectomy. Cauda equina lesion, reduced urinary sensation, reduced sensation left leg.
Non-insulin-dependent diabetes, no evidence of neuropathy. Glibenclamide Non-insulin-dependent diabetes, no evidence of neuropathy. Chlorpropamide Non-insulin-dependent diabetes, absent ankle and knee reflexes. Insulin Insulin-dependent diabetes, no evidence of neuropathy. Insulin Insulin-dependent diabetes, decreased peripheral sensation, postural hypotension. Insulin Insulin-dependent diabetes, postural hypotension, gustatory _s:l'\".e_ating,_diabetic_ diarrhoea, sensory neuropathy.
Group III consists of all the patients suffering from diabetes mellitus (n = 6; three non-insulin-dependent, three insulindependent) whether or not they showed clinical evidence of peripheral and/or autonomic neuropathy. The mean ages of the three groups were similar: Group I 52 ± 3.4 years (± SEM, n = 11), Group II 49 ± 3.5 years (n = 2), Group III 51 ± 5.8 years (n = 6). A summary of the results obtained from the immunohistochemical, histochemical and biochemical investigation of tissue from each patient is given in table 2. In certain cases insufficient tissue was available to perform all the techniques described. Immunohistochemical study of VIP. The pattern of distribution of VIP-immunoreactive nerves in the corpus cavernosum (CC), tunica albuginea (TA) and blood vessels (BV) within the erectile tissue from the present series of patients has been summarized in table 2. There was some variation in the pattern of innervation in different sections from individual patients. The results given in table 2 are the average assessment on examination of 12 to 20 sections from each patient. Micro-
AUTONOMIC INNERVATION OF PENIS IN DIABETIC IMPOTENT MALES
1055
TABLE 2. A summary of the results obtained from the immunohistochemical, histochemical and biochemical studies of 19 impotent patients
Patient
VIP-Like Immunoreactive Nerves
cc
TA
+++ +++ +++
+ + +
++ ++ ++
+++ +++ ++ +++
+ + + +
++ ++ ++ +
+
++
11
+++ ++
12 13
0
0
0
+++
+
+++
1 2 3 4 5 6 7 8 9 10
14 15 16 17 18 19
BV
0
0
0
+ +
0
+
+
0
0
0 0
+
+
AChE-Positive Nerves
Noradrenaline Content µ,g./gm.
tissue
cc
TA
BV
cc
+++
+
++
+++ ++ +++ +++ +++ +++ +++
+ + + + + + +
++ + ++ ++ ++ + +
0.34 0.52 0.45 0.41
0.33
0.37 0.21
0.19 0.03
0.22 0.33
0.04 0.17
+++
0
+++
+
++
0.19 0.20
0.09 1.35
+++ ++ + + +
++
+++ ++ + + +++
0.33 0.27 0.12
0.12 0.08 0.15
0.12 0.03
0.14 0.22
+ + 0
TA 0.33
Numbered according to the chart in table 1. 0 = absent, + = sparse, ++ = intermediate, +++ = numerous VIP-like immunoreactive or AChE-positive nerves (CC, corpus cavernosum; TA, tunica albuginea; BV, blood vessels).
graphs illustrating the pattern of innervation were selected on the basis that they were representative of the average assessment made. In Group I patients the tunica albuginea revealed only sparse innervation by VIP-like immunoreactive nerves which were localized mainly in nerve bundles (fig. IA). Moderate to sparse patterns of VIP-like immunoreactive nerves were also observed on the adventitial-medial border of some blood vessels within the erectile tissue from Group I patients. The highest density of VIP-like immunoreactive nerves was found in association with the cavernous smooth muscle (fig. lB). The patient with a cauda equina lesion was the only Group II patient investigated for VIP-like immunoreactivity. In this patient no VIP-like immunoreactivity was observed in nerves associated with the cavernous smooth muscle, tunica albuginea or blood vessels. Five of the diabetic Group III patients, three insulin-dependent and two non-insulin-dependent, revealed a marked reduction in VIP-like immunoreactive nerves when compared with Group I patients. The reduction was observed in the tunica albuginea and blood vessels but was most marked in the smooth muscle of the corpus cavernosum (fig. lC). The VIP-like immunoreactivity in the corpus cavernosum and tunica albuginea from one non-insulin-dependent diabetic, however, was indistinguishable from the Group I patients and additionally revealed the most dense innervation of blood vessels by VIP-like immunoreactive nerves of any of the patients studied. Histochemical study of AChE. In general, the pattern of distribution of positive AChE staining was similar to that described for VIP-like immunoreactivity (see table 2). Thus in Group I there were numerous AChE-positive nerve fibres in the cavernous smooth muscle (fig. 2A), while the tunica albuginea revealed sparse AChE-positive staining. The blood vessels of the corpus cavernosum revealed sparse to moderate innervation by AChE-positive nerve fibres. The Group II patient who had undergone a cystectomy gave similar results to those from Group I patients. Three Group III diabetic patients revealed a marked reduction in AChE-positive nerve fibres in the corpus cavernosum (fig. 2B). The AChE-positive staining of the corpus cavernosum
Fm. 1. VIP immunohistochemistry. A, arrow indicates fluorescent nerve fibres in nerve bundle within tunica albuginea from Group I patient. B, arrows indicate fluorescent nerve fibres in corpus cavernosum from Group I patient. C, marked reduction in VIP-like immunoreactivity in corpus cavernosum from Group III diabetic patient. Calibration bars = 80 µm.
from two non-insulin-dependent diabetics without clinical evidence of neuropathy was similar to that of Group I patients. There was no marked difference in the histological appearance of the tissues from any of the patients studied, as judged by routine haematoxylin and eosin staining. Examples of the tunica albuginea and corpus cavernosum are given in figs. 2C and 2D, respectively. Biochemical assay of noradrenaline. Adrenergic nerves could not be visualized using the glyoxylic acid technique in any of the patients studied due to high background levels of autofluorescence in the tissue specimens. A similar finding has been reported previously. 17• 25 The adrenergic innervation was therefore investigated by biochemical measurement of noradrenaline levels in the corpus cavernosum and tunica albuginea and the results have been given in table 2. The mean noradrenaline content of the corpus cavernosum from Group I patients was 0.36 ± 0.038 µg./gm. tissue (± SEM, n = 8) which was significantly higher (p <0.02, Student's t test) than the noradrenaline content of the tunica albuginea (0.18 ± 0.054 µg./gm. tissue, n = 6). There was some patient to patient variability in the noradrenaline content of both regions; how-
1056
LINCOLN AND ASSOCIATES
"
Fm. 2. A, arrows indicate AChE-positive nerve fibres in corpus cavernosum from Group I patient. B, marked reduction in AChE-positive nerve fibres in corpus cavernosum from Group III diabetic patient. Arrow indicates blood vessel that is negative for AChE staining. C, haemotoxylin and eosin staining of tunica albuginea. D, haematoxylin and eosin staining of corpus cavernosum. Calibration bars = 80 µm.
ever, it was observed that where the noradrenaline content of the tunica albuginea was low the level in the corpus cavernosum also tended to be low. In six Group I patients sufficient material was available for both regions to be assayed for noradrenaline. When the noradrenaline content of the corpus cavernosum was plotted against the noradrenaline content of the tunica albuginea from these six patients, the relationship was found to be linear by regression line analysis (r = 0.95, n = 6, p <0.01) and the equation for the line was calculated. The results have been shown in fig. 3 where each symbol represents one patient and is numbered according to the chart in table 1. Confidence limits for the line at the 95% level have also been indicated on the figure. The Group II patient who had undergone a cystectomy gave noradrenaline values which were marginally outside the 95% confidence limits of the line. The patient with a cauda equina
lesion had the highest noradrenaline content in the tunica albuginea of any of the patients studied. The mean noradrenaline content of the corpus cavernosum from Group III diabetic patients was 0.17 ± 0.055 µg.jgm. tissue (n = 5) which was significantly lower (p <0.02) than that of the Group I patients. There was no significant difference, however, between the mean noradrenaline level of the tunica albuginea from Group I patients and that from Group III diabetic patients (0.14 ± 0.023 µg.jgm. tissue; n = 5). The results from the Group III diabetic patients were also assessed individually by the correlation between the noradrenaline content of the corpus cavernosum and of the tunica albuginea (fig. 3). In contrast to Group I, three Group III diabetic patients with clinical evidence of neuropathy had lower levels of noradrenaline in the corpus cavernosum than the tunica albuginea and thus the values were markedly outside the
1057 (1)
:, E :J
"'
(J)
06
values
0
C:
'>ro© 0-5 rn
01
:i.U
t:
Q)
which other groups could be com-
(j)
(/)
::,
c0 a.0 OU
o4 0-4
06
z
10/ 03 15 ..
0-2
7
13@
.. ·;.,12
17 AA16 0-1
0-1
0-2
0-3
0-4
If-
1-3
1-4
NA Content µg/g tissue Tunica Albuginea
FIG. 3. Relationship between noradrenaline content of corpus cavemosum and tunica albuginea. Each symbol represents single patient and is numbered according to chart in table 1. 0 = Group I patients. Relationship between noradrenaline contents of two regions was linear (r = 0.95, p <0.01). Dotted lines indicate 95% confidence intervals of line. it = Group II patients. A = Group HI, diabetic patients.
95% confidence limits of the line (fig. 3). The noradrenaline levels from two non-insulin-dependent diabetics without evidence of neuropathy, however, were within the range predicted from Group I patients. DISCUSSION
In the present study all the patients investigated suffered from impotence, and it was possible to compare the pattern of autonomic innervation of the penis in impotence due to a variety of causes. Studies of this kind are faced with considerable difficulties with regard to the availability of both suitable control material and large patient numbers. In ~~.Mn,rn,,~ studies, control penile tissue was obtained from transsexuals and carcinoma patients undergoing surgery. 16 · 17• 25 · 26 However, on occasion, patients with Peyronie's disease have been used as controls in combination with patients tomy. 19 · 26 It has been suggested that the fibrous tissue, in Peyronie's disease, interferes wc,u,.o.un,c,uy with the normal function of the cu,uu,u,;v system. 27 In an electron m,~P,,O~An,~ abnormalities were observed in the within the fibrous plaque while the unmyehnated axons of the surrounding erectile tissue appeared unchanged, thus there was no ultrastructural evidence to suggest that the autonomic nervation of the penis was affected in Peyronie's disease. 28 In the present investigation, four patients with Peyronie's disease were combined with seven other patients in whom neuropathy was not present clinically nor was there any disturbance in bladder or bowel function. It was realized that the presence of normal bladder function, in cases where doubt occurred, did not totally exclude the possibility that neuropathic changes had occurred in the penis. However, since the penis and bladder share a common nerve supply, 12 • 13 • 29 this was considered the most appropriate criterion for establishing Group I. This view was strengthened by the consistent results obtained for Group I patients despite the differing causes of impotence and was particularly encouraging in the biochemical measurements of noradrenaline levels where more precise quantitation could be achieved. The results from Group I patients were therefore considered to form a baseline of 'non-
A detailed discussion on the autonomic innervation of the penis in Group II patients is beyond the scope of the present the main aim of which was to investigate changes occurring in impotent patients with diabetes mellitus. A recent immunohistochemical of the cat spinal cord revealed that VIP is localized primarily in the sacral segments that supply the urinary tract, large intestine and sex organs" 30 In the present series of patients, the patient with a cauda equina lesion revealed a complete absence of VIP-like immunoreactivity in nerves of the cavernous smooth muscle, tunica albuginea and blood vessels" Nerve damage due to cystectomy has been recognized as an iatrogenic cause of impotence" 29 While the noradrenaline levels of the cystectomy patient were marginally outside Group I values, the pattern of AChE-positive nerves was indistinguishable from Group L At this stage it is not possible to conclude whether impotence in this patient was due to damage to nerves, blood vessels or psychological factors following radical surgery" A linear relationship was shown to exist between the noradrenaline levels of the corpus cavernosum and tunica albuginea from Group I patients, which has not been published before, and represents a new and useful approach for future studies of this kind. The noradrenaline levels in control penile tissue, reported in a previous study, 16 were very variable, possibly due to regional variation in tissue samples obtained from penectomies. The presence of large errors for mean control noradren aline content can prevent the detection of subtle changes in other groups. While patient to patient variability was less marked in the present study, it was still advantageous to examine the relationship between the two regions. It was thus possible to compare the individual results obtained for each Group II and Group III patient with the entire range of values in Group I, as well as to analyze the differences between the mean values for each group, The justification for examining the relationship between the two regions lies in the fact that it has been shown anatomically that the nerve supply to the corpus cavernosum crosses the tunica albuginea to innervate the cavernous smooth muscle. 29 Thus the nerve bundles, largely localized in the tunica albuginea, and the single nerve fibres of the corpus cavernosum could be regarded as different regions of the same nerve supply. If damage occurs to the sympathetic nerves then it would be predicted that a decrease in noradrenaline content would occur at the site of the terminal varicosities in the corpus cavernosum rather than in the nerve trunks of the tunica albuginea. Therefore - .. of the -~,."'-""!,'),'::c~c'::.!v between the noradrenaline levels of also have the advantage of an unu~u.Hvu nerve damage as opposed to In the present one of the and had received a (3antagonist has been shown to decrease sym pathetic nerve activity. 31 The noradrenaline levels of this patient were the lowest of any of the Group I patients; however, the relationship between the two regions was maintained. In contrast, three diabetic patients revealed a significant, preferential decrease in the noradrenaline content of the corpus cavernosum relative to the tunica albuginea which suggests that damage to the sympathetic nerves of the erectile tissue had probably occurred in these patients. A significant reduction in noradrenaline levels in erectile tissue in diabetes-associated impotence has been reported previously although levels in the tunica albuginea were not measured, 16· 17 Unlike the present study, the reduction was not found in non-insulin-dependent diabetics. Results from three noninsulin-dependent diabetics have been reported here. The patient who presented with clinical symptoms of neuropathy gave values which were significantly outside the range for Group I while those of the two patients with no clinical indication of - q ~w
1058
LINCOLN AND ASSOCIATES
neuropathy were similar to the 'non-neuropathic' group. Clinical details of the patients in the previous studies were not given. Therefore, it cannot be excluded that non-insulin-dependent diabetics with neuropathy may be as susceptible to sympathetic nerve dysfunction in the penis as insulin-dependent diabetics. The changes in histochemical staining for AChE in erectile tissue in diabetes, observed in the present study, have not been reported previously. It should be noted that AChE staining is not entirely specific for cholinergic nerves. For example, AChE has been shown to hydrolyze substance P and, in the dorsal horn of the spinal cord, the distribution of AChE staining correlates well with substance P-like immunoreactivity. 32 Thus the reduction in AChE staining in the diabetic group of patients cannot provide conclusive evidence of a reduction in cholinergic nerves alone. Attempts were made to provide additional information on the cholinergic innervation by the biochemical measurement of choline acetyltransferase activity. Levels of activity proved to be too low to be Il).easured (unpublished observations), a finding which has also been reported previously. 25 The involvement of the parasympathetic innervation of the penis in diabetic impotence, however, has been suggested in clinical studies 13• 15• 33 and the results reported here would tend to support this view. The pattern and distribution of VIP-like immunoreactive nerves resembled that of AChE-positive nerves in the erectile tissue specimens investigated in the present study. It has been suggested that both VIP and AChE may be localized in the same nerve fibres in the human male genital tract. 7 However, it is also recognized that coexistence cannot be assumed from finding similar distributions at the light microscopical level alone. At this stage it is not possible to conclude whether the observation of similar reductions in both VIP-like immunoreactivity and AChE-positive staining in certain diabetic patients signifies changes in a single or two different nerve types. A reduction in VIP-like immunoreactive nerves in penile tissue was reported in a preliminary study from this laboratory on one diabetic patient when compared with three patients with Peyronie's disease. 18 The results of the present investigation revealed a marked reduction in VIP-like immunoreactivity in the corpus cavernosum from five out of six impotent diabetic patients when compared with nine impotent patients who had been classified as 'non-neuropathic'. In contrast, 1.t has been reported that levels of VIP and VIP-like immunoreactivity were similar in impotent patients regardless of the cause of impotence and that they were all reduced when compared with control material from transsexual patients undergoing penectomy.111Thepresent series of patients provides no information on the autonomic innervation of the penis in normal potent males, due to the obvious lack of available tissue, although certain patients (with Peyronie's disease and psychogenic impotence) retained some erectile function. Some caution may also be necessary before assuming that levels in transsexuals represent normal values due to the intensive drug therapy these patients undergo prior to surgery. The aim of the present study was to investigate whether different causes of impotence could be distinguished by the changes in the pattern of autonomic innervation that the erectile tissue exhibited. The results reported here indicate that this was still possible despite the lack of unequivocal control material. The present study does not exclude the possibility that vascular changes also contribute to the development of impotence in diabetes. Furthermore, psychological factors in impotence are as important in diabetics as in the non-diabetic population. Several clinical studies have investigated whether the neuropathic or the vascular changes associated with diabetes are of greater significance to the development of impotence, with conflicting results. Vascular occlusion has been indicated as the major contributing factor by some workers9- 11 while others have found a greater association with neuropathy. 8 • 12• 14
In the single study performed on post mortem tissue, morphological changes were observed in nerves in the corpus cavernosum from impotent diabetics while no evidence was found of microangiopathy.12 The conflict may in part be explained by the variety of clinical procedures used to assess the neurological status of patients and the fact that few of these procedures can reliably indicate the presence or lack of local involvement of the autonomic nerves of the penis in diabetes, which requires more direct study. This point has been demonstrated in a recent clinical study in which the conduction velocity of the dorsal nerve of the penis was shown to be significantly lower than controls in diabetics with impotence while the bulbocavernous reflex and pudendal evoked potential were within the normal range. 34 It is of relevance to this argument that, in the present study, a reduction in VIP-like immunoreactivity and AChEpositive staining in nerves was observed in penile tissue from an insulin-dependent diabetic whose neurological status had been assessed as normal. The autonomic nervous control of erection is complex. It has been reported that adrenergic, 2• 3 • 26 cholinergic 2• 35 and VIPergic4- 7 nerves in the penile vessels and cavernous smooth muscle, may all have significant roles in the process of erection and detumescence. Few studies have attempted direct examination of penile tissue in diabetic patients. The present study consists of an immunohistochemical, histochemical and biochemical investigation of the VIPergic, cholinergic and adrenergic innervation in penile tissue from patients with impotence due to a variety of causes. The results provide evidence that all three nerve types may be affected in diabetes mellitus and thus may contribute to the development of impotence in diabetic males. REFERENCES
1. de Groat, W. C. and Booth, A. M.: Physiology of male sexual function. Ann. Intern. Med., 92: 329, 1980. 2. Weiss, H. D.: The physiology of human penile erection. Ann. Intern. Med., 76: 793, 1972. 3. Benson, G. S., McConnell, J., Lipschultz, L. I., Corriere, J. N. Jr. and Wood, J.: Neuromorphology and neuropharmacology of the human penis. J. Clin. Invest., 65: 506, 1980. 4. Willis, E., Ottesen, B., Wagner, G., Sundler, F. and Fahrenkrug, J.: Vasoactive intestinal polypeptide (VIP) as a possible neurotransmitter involved in penile erection. Acta Physiol. Scand., 113: 545, 1981. 5. Polak, J.M., Mina, S., Gu, J. and Bloom, S. R.: VIPergic nerves-in the penis. Lancet, 2: 217, 1981. 6. Andersson,-P.-0.,-Bfoom,-S. R. andMellandex,S.:Haemodynamics of pelvic nerve induced penile erection in the dog: possible mediation by vasoactive intestinal polypeptide. J. Physiol. (Lond.), 350: 209, 1984. 7. Gu, J., Polak, J. M., Probert, L., Islam, K. N., Marangos, P. J., Mina, S., Adrian, T. E., McGregor, G. P., O'Shaughnessy, D. J. and Bloom, S. R.: Peptidergic innervation of the human male genital tract. J. Urol., 130: 386, 1983. 8. Kolodny, R. C., Kahn, C. B., Goldstein, H. H. and Barrett, D. M.: Sexual dysfunction in diabetic men. Diabetes, 23: 306, 1974. 9. Herman, A., Adar, R. and Rubinstein, Z.: Vascular lesions associated with impotence in diabetic and non-diabetic arterial occlusive disease. Diabetes, 27: 975, 1978. 10. Jevtich, M. J., Edson, M., Jarman, W. D. and Herrera, H. H.: Vascular factor in erectile failure among diabetics. Urology, 19: 163, 1982. 11. Lehman, T. P. and Jacobs, J. A.: Etiology of diabetic impotence. J. Urol., 129: 291, 1982. 12. Faerman, I., Glocer, L., Fox, D., Jadzinsky, M. N. and Rapaport, M.: Impotence and diabetes: histological studies of autonomic nervous fibers of the corpora cavernosa in impotent diabetic males. Diabetes, 23: 971, 1974. 13. Ellenberg, M.: Sexual function in diabetic patients. Ann. Intern. Med., 92: 331, 1980. 14. Karacan, I.: Diagnosis of erectile impotence in diabetes mellitus. Ann. Intern. Med., 92: 334, 1980.
AlJT01'10Iv'.iIC IfTNEl'-lVATIOJ\f Ol? P:El\IIS IN Dil-\BE'TIC Ih-'.e:POTEI\J'T lVIALES
15. 16.
17. 18.
19.
20. 21.
22. 23. 24.
M.: Impotence in diabetes: the neuroiogic factor. Ann. Intern. 75~ 213, 1971. Melman, A., Henry, D. P. and Felton, D. L.: Catecholamine content of the peniie corpora in patients with diabetes-associated impotence. Surg. Forum, 29: 634, 1978. Melman, A., Henry, D. P., Felten, D. L. and O'Connol', B.: Effect of diabetes upon penile sympathetic nerves in impotent patients. South. Med. J., 73: 307, 1980. Crowe, R., Lincoln, J., Blacklay, P. F., Pryor, J. P., Lumley, J. 8. P. and Burnstock, G.: Vasoactive intestinal polypeptidelike immunoreactive nerves in diabetic penis. Diabetes, 32: 1075, 1983. Gu, J., Polak, J. M., Lazarides, M., Morgan, R., Pryor, J. P., Marangos, P. J., Blank, M. A. and Bloom, S. R.: Decrease of vasoactive intestinal polypeptide (VIP) in the penises from impotent men. Lancet, 2: 315, 1984. Coons, A. H.: Fluorescent antibody methods. In: General Cytochemical Methods. Edited by J. F. Danielle. New York: Academic Press, pp. 399-422, 1958. Lindvall, 0. and Bjorklund, A.: The glyoxylic acid fluorescence histochemical method: a detailed account of the methodology for the visualisation of central catecholamine neurons. Histochemistry, 39: 97, 1974. Karnovsky, M. J. and Roots, L.: A 'direct coloring' thiocholine method for cholinesterases. J. Histochem. Cytochem., 12: 219, 1964. Keller, R., Oke, A., Mefford, I. and Adams, R. N.: Liquid chromatographic analysis of catecholamines: routine assay for :regional brain mapping. Life Sci., 19: 995, 1976. Moyer, To P. and Jiang, N.-S.: Optimized isocratic conditions for analysis of catecholamines by high-performance reversed-phase paired-ion chromatography with amperometric detection. J.
25. 1\!Ielman, LI!."""' Felten) D. L. and B. L.: Alteration of the penile corpora in patients with impo1980. tence. Invest. Urol., 17: 26. Meiman, A. and Hemy, D.: possible role of the catecholamines of the corpora in penile erection. J. Urol., 121: 1979. 27. Metz, P., Ebbehoj, J., Uhrenholdt, A. and Wagner, · Peyronie's disease and erectile failure. J. Urol., 130: 1103, 1983. 28. Vande Berg, J. S., Devine, C. J., Horton, C. E., Somers, K Do, Wright, G. L Jr., Leffel!, M. S., Dawson, D. M., G!eischman, S. H. and Rowe, M. J.: Peyronie's disease: an electron microscopic study. J. Urol., 126: 333, 1981. 29. Lue, T. F., Zeineh, S. J., Schmidt, R. A. and Tanagho, E. A.: Neuroanatomy of penile erection: its :relevance to iatrogenic impotence. J. Urol., 131: 273, 1984. 30. Kawatani, M., Lowe, I. P., Nadelhaft, I., Morgan, C. and de Groat, W. C.: Vasoactive intestinal polypeptide in visceral afferent pathways to the sacral spinal cord of the cat. Neurosci. Lett., 42: 311, 1983. 31. Chernow, B., Zaloga, G. P., Lake, C R., Coleman, M. D. and Ziegler, M. G.: Effect of antihypertensive therapy on sympathetic nervous system activity in patients with essential hype1tension. Fed. Proc., 43: 72, 1984. 32. Chubb, L W., Hodgson, A. J. and White, G. H.: Acetylcholinesterase hydrolyzes substance P. Neuroscience, 5: 2065, 1980. 33. Clarke, B. F., Ewing, D. J. and Campbell, I. W.: Diabetic autonomic neuropathy. Diabetologia, 17: 195, 1979. 34. Lin, J. T. and Bradley, W. E.: Penile neuropathy in insulindependent diabetes mellitus. J. 133: 213, 1985. 35. McConnell, J., Benson, G. S. and J.: Autonomic innervation of the mammalian penis: a histochemical and physiological study. J. Neural Transm., 45: 227, 1979.