Functional Specialization of the Extrinsic Venom Gland Musculature within the Crotaline Snakes (Reptilia: Serpentes) and the Role of the M. Pterygoideus Glandulae

-The extrinsic venom gland musculature in snakes exhibits a reasonably conserved spatial pattern around the venom gland. A prominent exception to this pattern is the M. Pterygoideus Glandulae, which originates from the surface of lhe venom gland itself. This muscle is found only in crotaline snakes, and has a spotty distribution within that group. Previous workers proposed that the M. Pterygoideus Glandulae represents a muscular specialization to increase venom expulsion. Here, we test this hypothesis by directly quantifying the relative venom pressure produced upon stimulation of the individual extrinsic venom gland muscles of three crotaline snakes: Calloselasma rhodoslOma (where the M. Pterygoideus Glandulae is pronounced), Crotalus atrox (which has a M. Pterygoideus Glandulae), and Agkisll'udull collturtrix (which lacks this muscle). Our results suggest that the M. Pterygoideus Glandulae does increase venom expulsion, and thal crotaline snakes which lack this muscle, like A. contortrix, have not compensated in other ways. The key to understanding the functional significance of this muscle, and perhaps its phylogenetic development, may simply be the relative length of this muscle's course perpendicular to the venom gland, which is related ta the spatial relationship between the venom gland and palata-maxillary arch in these snakes. tloes not appear to be homologous among the major taxa of venomous snakes. It is referred to as the M. Compressor Glantlulae, or as a portion of the M. Adductor Mandibulae Externus complex (either Superficialis, Metlialis, or Profun­ dus), tlepending on the taxa (Haas, 1973; Jackson, 2003). The at least superficially analogous arrangement of the extrinsic venom glantl musculature, eombinetl with the ap­ parent intlependent evolution of the venom delivery systems in different snake taxa (Cundall and Greene, 2000; Jackson, 2003), suggests a tlegree of functional convergence in ven­ om expulsion. One of the most interesting specializations within this potentially convergent system occurs within the pitvipers, or crotaline snakes. Within this ratliation, the M. Pterygoideus may give rise to a separate head, the M, Ptery­ goideus Glandulae, which has a direct origin on the medial surface of the venom gland. Groombritlge (1986) reported that, within the crotaline snakes, there is significant varia­ tion in the development of the M. Pterygoideus Glantlulae; indeetl, GroombriJge recognized four different states of development. Herein, we recognize only three states, as the first two stales of Groombridge were not functionally tlillerent from one another. As Groombridge (1986) noted, and Kardong (L990) later emphasizetl, the tlistribution of these states within crotaline snakes does not correlate well with any recognized taxonomy or phylogeny for the group, with different states being reported from species within the same genus. Groombritlge (1986) and Kardong (1990) both speculatetl that increasetl development of the M. Pterygoi­ tleus GlanJulae woulJ result in increased venom expulsion. Contraction of the M. Pterygoideus Glandulae woulJ place tensile forces on the medial surface of the venom glanJ, _.u!rh, Department of Physical Therapy, University of which contrasts with the compressive forces produced by e-mail: Bruce [email protected] U, at Lowell, Lowell, rvlassachuselts 018544 USA Iinlil t.'niversily Press hone and courses craniulaterally, along the me­ 'nlromedial surface of thc venom gland, to insert cctopterygoid-maxilla juint. The latter muscle hI he rrc~enl among all venomous (and non-venLes. In contrast, the M. Compressor Glandulae INTRODUCTION In all venomous snakes, the venom is secreted and '1r~u in a specialiLcd gland (the venom gland) located lI111g the posterolateral margin of the upper jaw. Venom is \ 'lieu from the gland when one of the surrounding skel­ muscles comprising the extrinsic venom gland muscu­ ur~Clln\raCls, thereby increasing the fluid pressure within 'cnom gland (Rosenberg, L967; Young et aI., 2000). etal kinesis may indirectly produce limited venom ex­ Ill" IYoung and Zahn, 2001); however, the extrinsic gland musculature is the primary motive force for \low (though olh(~r factors will influence how much is lI\:tually discharged; see Young el aI., 2002). , exact composition of the extrinsic venom gland ,!;lUlrc varies considerably among venomous taxa (see 1m. and Zaher, 1994). Generally, though, there are llIn;ic I'enom gland muscles. One, the M. Compres­ IJnJulae, originates from (or is closely applied to) the allli dllr;olateral surface uf the gland before cours­ !l1\cntrally to insert on the lower jaw. The second, I'tc~gllideus, originates from the caudal end of the 48 B. A. Young and K. Jackson the other extrinsic venom gland muscles. The different force vector generated hy the M. Pterygoideus Glandulae could influence how venom is expelled from the venom gland parenchyma and central chamber into the venom duct. The purpose of this study was to test the hypothesis that contraction of the M. Pterygoideus Glandulae enhances venom expulsion. MATERIALS AND METHODS Animal caI-e.-Seven live specimens were obtained commercially: three Western Diamondback Rattlesnakes (Crotalus atrox, snout-vent length, SVL =60-65 em; mass = 100-140 g); two Copperheads (Agkistrudon contortrix, SVL = 54--{i7 cm, mass = 110-180 g); and two Malayan Pit Vipers (Calloselasma rhodostema. SVL =43.5-69 cm, mass =50-300 g). These three species were chosen because they span the range of variation in the development of the M. Pterygoideus Glandulae. Agkistrodon contortrix has no evidence of a separate M. Pterygoideus Glandulae (state A), whereas C. rhodostoma has one of the most pronounced developments of a M. Pterygoideus Glandulae among cro­ taline snakes (state C). Crotalus atrox has a distinct M. Pterygoideus Glandulae (state B), but it is not as distinct as in C. rhodostema. At Lafayette College, these snakes were maintained in a specially designed venomous snake room with a 12: 12 photoperiod, a temperature of 26-3\ °C, water ad lihitum, and a diet of pre-killed mice. To ensure a replen­ ished venom supply, none of the snakes were fed within two weeks of the experiments. All animal maintenance and experimental procedures complied with existing guidelines for both live reptiles and venomous snakes, and were ap­ proved by the Lafayette College Institutional Animal Care and Use Committee. Anatomy.-Two preserved specimens each of C. atrox and A. contortrix, from the private collection of Bruce Figure 1. Generalized depiction of the venom delivery system in crotaline snakes. The M. Compressor Glandulae (C) originated from the dorsal surface (stippled) of the venom gland (V) and courses ventrad to insert on the compound. The M. Pterygoideus (P) courses along the ventromedial surface of the venom gland. The dorsal deflection in the venom duct was the insertion site for the pressure cannula. Young (BAY), were examined, as was one specimen of C. rhodostoma on loan from the Smithsonian Institutillll (USNM 53440). In each specimen. the lower jaw was dt­ flected ventrolaterally to expose the roof of the mOllth.1lr lining epithelium of the mouth was removed on one silk to expose the venom gland and M. Pterygoideus. FollOW­ ing the removal of some superficial connective tissue, It' dissection was photographed using a Nikon SMZ-IO til section microscope (Nikon Inc., Tokyo, Japan). At the 5 of the greatest interaction between the venom gland and M. Pterygoideus Glandulae (or M. Pterygoideus), a nil blade was used to excise a block of tissue consisting of venom gland and portions of both extrinsic venom g muscles. This excised tissue was dehydrated through ethanol series, cleared in Hemo-De® (Thermo FisherS tific, Waltham, Massachusetts, USA), and then em in Paraplast® (Thenno Fisher Scientific). Transverse tions (10 11m) were cut from each block using a Lipsl' rotary microtome (Shandon Lipshaw Inc., PittsburghJ sylvania. USA). Mounted sections were alternately sta with Hematoxylin and Eosin, as well as Van Gieson'l (Luna, 1968; Presnell and Schreibman, 1997). Slides photographed using a Nikon Eclipse compound mie (Nikon Inc.). Functional morphology.-Each specimen was I. anesthetized through exposure to Isoflurane®. Once thetized, the specimen was weighed using an Ohau digital balance (Ohaus Corp., Pine Brook, Nell' J USA), and its body length (SVL) determined. Further thesia was provided by an intramuscular injection of kg Ketamine Hydrochloride:Acepromazine in a 9:1 The anesthetized specimen was placed on a heated table (VSSI Inc., Carthage. Missouri, USA) and pr additional exposure to Isoflurane. With the specimen positioned on its side, thl' overlying the venom gland and duct were remOld the epithelium from the roof of the mouth. Using a gical scope (Leica Inc., Wetzlar. Germany). the veil. was surgically isolated from the adjacent connecli\~ vasculature, and nerves. In crotaline snakes the portion of the venom duct includes a prominent directed convolution (Fig. I). The proximal v was catheterized using flexible plastic tuhing: the of the tubing was adjusted according to the dia venom duct, but the inner diameter of the tubiJ!! from 0.02-0.045 mm. Immediately proximal to tion site of the catheter, the catheter was held in silk sutures around the venom duct; these sutures enough to preclude both movement of the callI( to the venom duct and discharge of venom dislaJ~ eter. Due to the variations in the contours of venom duct and in the adjacent soft tissues.l!JI'r. nificant variation among the specimens in the the catheter tip within the venom delivery s)'s!e~ The free end of the catheter was connecteJ· hypodermic needle, to a PT300 pressure trait\; S = supralabial scaIation; V = venom gland. Functional specialization o/venom gland musculature 49