Freeze-etch studies on the innervation of mesenteric arteries and vas deferens

Peripheral adrenergic neuromuscular relationships in smooth muscle cells have been documented in electron-microscope studies of small arteries and arterioles (e.g. Simpson & Devine, 1966; Devine & Simpson, 1967; Bumstock, Gannon & Iwayama, 1970) and vas deferens (e.g. Richardson, 1962; Merrillees, Bumstock & Holman, 1963; Merrillees, 1968). Merrillees (1968) showed by serial section sampling of the guineapig vas deferens that the axons lay between smooth muscle cells at distances of 20–30 nm at regions of close contact. Similar studies have not been made upon blood vessel neuromuscular contacts, but random sectioning indicates that the axons lie in the vascular adventitia, the minimum distance from smooth muscle cells being approximately 90 nm (Devine & Simpson, 1967). Usually the distances are much larger, e.g. 100–700 nm (Devine & Simpson, 1967).

Freeze-etch studies provide views of the surfaces of cells in the unfixed frozen state. In favourably oriented specimens the neuromuscular relationships, axonal varicosities and axonal contents can be studied. In the present study the neuromuscular relationships of mouse and guinea-pig vas deferens and of guinea-pig small mesenteric arteries are investigated by the freeze-etch method.

The mesenteric blood vessels of guinea-pigs anaesthetized with sodium pentobarbitone, 55 mg intraperitoneally, were exposed and flooded with 1 % procaine hydrochloride in either Krebs-Ringer phosphate buffer or Tyrode’s solution. Small arteries approximately 300 μm in diameter were removed, stripped of fat, placed in 25 % glycerol in Krebs-Ringer phosphate or Tyrode’s solution for 30–60 min and cut into small pieces. Portions of mouse vas deferens were placed in 25 % glycerol in phosphate buffer for 30–60 min and cut into small pieces. The pieces of tissue were then frozen, fractured, etched, shadowed and replicated in a Balzers BA500R freeze-etch apparatus as described by Moor & Mühlethaler (1963) and Moor (1964, 1965).

Small mesenteric arteries and vas deferens of guinea-pigs were prepared for conventional electron microscopy by fixing in 2% glutaraldehyde-formaldehyde in Tyrode’s solution, and post-fixing in osmium tetroxide in Tyrode’s solution. The tissue was then block-stained with 2 % aqueous uranyl acetate and embedded in Epon 812. Thin sections were stained with alkaline lead citrate. Replicas and sections were examined electron microscopically.

The freeze-etch appearances of smooth muscle cells in the vas deferens and blood vessels have been previously described (Devine, Simpson & Bertaud, 1969, 1971). Longitudinal rows of surface vesicles are a prominent feature of the smooth muscle cell surfaces (Fig. 4). The fact that the vascular axons lie exclusively in the adventitia, and usually in bundles, makes them comparatively easy to identify. In the vas deferens, however, the axons run between the smooth muscle cells, and many different fragments of surfaces of axons, Schwann cells and smooth muscle cells are usually present, so that only relatively few axons can be identified with certainty. Axons are similar in both types of smooth muscle tissue and will be described together.

Large bundles of axons are present in the serosal coat of the vas deferens and in the adventitia of blood vessels, surrounded by large amounts of collagen. Transverse and longitudinal fractures of such bundles reveal outer and inner aspects of the plasma membranes of axons and Schwann cells. Axons are mainly cylindrical in shape and are found in depressions or grooves of the Schwann cell, sometimes completely enveloped by the latter (Fig. 1). The outer and inner aspects of the axonal membrane are covered with numerous particles up to 15 run in diameter. Small bundles of axons are present between the smooth muscle cells of the vas deferens (Fig. 2), and in the adventitia of blood vessels (Figs. 3–9) lying close to the outer layer of smooth muscle cells. The axons in these small bundles are less fully enveloped by Schwann cells and appear less cylindrical than the axons in the large bundles (Fig. 1); varicosities can be seen in many instances (Figs. 2–9). The axonal surface does not appear to differ from that described for the larger axon bundles. The varicosities are large, up to 1·6 μm in diameter and 2 μm long, and are devoid of Schwann cell over much of their surface (Figs. 6, 9). Often they are rounded (Fig. 8) but irregular varicosities are also found (Fig. 6). Complete varicosities and adjacent narrow regions are seldom seen, but clear suggestions of alternating regions can be discerned (e.g. Figs. 3, 8).

Apart from the 15-nm particles on the inner and outer surfaces of axons, there are few other observable surface features; only occasional small shallow depressions, if viewed from the outer surface or protrusions if viewed from the inner surface, are seen (Figs. 4, 7, 9).

Axons lie in the adventitia of the blood vessels, surrounded by collagen fibrils, and come close to smooth muscle cells at intervals to form ‘close contacts’. Although elastic tissue is present in the adventitia between the axons and the smooth muscle cells, it is usually absent at regions of close contact (Figs. 3–5). The investment of Schwann cells is also usually deficient at these points. Neuromuscular distances less than 180 run have not been seen in this study (Figs. 3–6).

Electron micrographs of thin sections show the typical vascular neuromuscular relationship (Fig. 10): a vesicle-filled axon varicosity, free of Schwann cell cover at the point of nearest approach, making close contact with smooth muscle cells at regions devoid of elastic tissue (although collagen is present). Surface vesicles and dense bodies are seen in the smooth muscle cells.

In the vas deferens, axons lie between the smooth muscle cells and are occasionally in grooves in the smooth muscle cells (Fig. 2); the axons are thus never so far away from smooth muscle cells as is the case in vessels. The neuromuscular distance at regions of closest contact is about 20–50 nm, i.e. considerably less than the smallest distance found in the case of vascular smooth muscle close contacts. Varicosities of the axons, devoid of Schwann cell cover, can be seen (Fig. 2).

Axons lying in large bundles have stumps, probably of neurotubules, 25 nm in diameter, protruding from them (Fig. 1). Occasional synaptic vesicles are present in some of these axons. Larger numbers of synaptic vesicles are usually found in varicosities of axons making close contact with smooth muscle cells (Figs. 5–7, 9). The vesicles in freeze-etch micrographs range in diameter from 40–100 nm, the majority being approximately 50 nm in diameter. These vesicle sizes correspond closely to the diameters found in preparations fixed for conventional electron microscopy (Fig. 10). Particles up to 15 nm in diameter are seen on both inner and outer aspects of the vesicle membranes (Fig. 9). Spherical structures larger than those of vesicles probably correspond to portions of long mitochondria (Fig. 9). No special structure was observed in freeze-etch preparations which would correspond to the dense cores in the small synaptic vesicles, as seen in sectioned material.

The present study complements the conventional electron-microscopic investigations of sympathetic neuromuscular relationships. It has demonstrated the apposition of the axon and the smooth muscle cell over considerable distances and has also demonstrated the axonal varicosities. The morphology of the axonal varicosities is similar to that described by Nickel & Potter (1970) for cholinergic axons in the Torpedo electric organs.

The freeze-etch studies have so far revealed no clear modification of the smooth muscle membrane to form specialized receptor areas at the site of neuromuscular contact. Since the rows of surface vesicles on the smooth muscle membrane are relatively common (Fig. 4), chance apposition of the axon to such areas may provide the necessary transfer mechanism.

The sizes of the synaptic vesicles observed in this study confirm those found in conventionally fixed blood vessels (Devine & Simpson, 1967), and are also in agreement with observations of Moor, Pfenninger & Akert (1969), who found little difference in sizes of synaptic vesicles in conventionally fixed tissues, tissues pre-fixed with glutaraldehyde and glycerol before freezing and tissues treated with glycerol alone before freezing. As the great majority of vesicles seen in freeze-etched preparations had fractured along the membrane rather than through the centre, no evidence was available as to the nature of the densely staining cores frequently found in these bodies in sectioned material. In Torpedo (Nickel & Potter, 1970), the majority of the synaptic vesicles were the same size as the large vesicles in the present study.

Although the particles found on fracture faces of frozen biological membranes are a convenient identification feature, since particle populations frequently show characteristic differences between one membrane face and another, their nature and biological function are as yet unknown. Various investigators (e.g. Pinto da Silva, 1970; Tillack & Marchesi, 1970; Wrigglesworth, Packer & Branton, 1970) have confirmed that they occur within the membrane and not on its surface, and that they are exposed in freezeetch preparations by splitting of membranes along a central plane.

In freeze-etch studies of the subfornical organ by Moor et al. (1969) and of Torpedo electric organ (Nickel & Potter, 1970) depressions were seen in the surface of axons, and in cross-fractured axons openings of synaptic vesicles fused with the axonal membrane. Depressions occasionally seen in the surface of axons in the present study were not deep and did not appear to connect with synaptic vesicles in cross-fractured axons, but this possibility cannot be excluded on the present evidence.

The authors wish to thank Miss J. M. Ledingham and Mrs S. O’Kane for skilful technical assistance and the Medical Research Council of New Zealand for financial assistance. The use of the Hitachi HU 11E electron microscope provided by General Research Support Grant N.I.H. FR05610 of the Presbyterian-University of Pennsylvania Medical Center is gratefully acknowledged.

 
• a

  axon

 
• ai

  inner aspect of axonal membrane

 
• ao

  outer aspect of axonal membrane

 
• col

  collagen

 
• db

  dense body

 
• el

  elastic tissue

 
• m

  mitochondria

 
• mi

  inner aspect of mitochondrial membrane

 
• mo

  outer aspect of mitochondrial membrane

 
• myo

  myofilaments

 
• nt

  neuro tubules

 
• sch

  Schwann cell

 
• schi

  inner aspect of Schwann cell membrane

 
• scho

  outer aspect of Schwann cell membrane

 
• sm

  smooth muscle

 
• smi

  inner aspect of plasmalemma of smooth muscle cell

 
• smo

  outer aspect of plasmalemma of smooth muscle cell

 
• sv

  synaptic vesicle

 
• v

  vesicle

All the freeze-etch replicas are from small mesenteric arteries of guinea-pig and vas deferens of mouse. The direction of shadowing is indicated by (♂).

The words ‘inner ‘and ‘outer ‘aspects of the plasma membrane and other membranes are loosely used and are not meant to imply that the membranes are fractured along the inner or outer surfaces, but are used for convenience.