Shields and Squyars: Mortality and hematology of Call/nectes sopidus infected with Hematodinium perezi 



141 



of hemolymph from an uninfected donor crab and 

 an experimental gi'oup («=20) injected individually 

 with 100 ]iL of infected hemolymph from a donor 

 crab containing an estimated 1.3 x 10'' trophonts/mL 

 (1.3 X 10^ trophonts per crab). 



The mortality-II experiment consisted of a con- 

 trol group (n=8) injected individually with 100 ]iL 

 of physiological saline buffer (modified from Apple- 

 ton and Vickerman, 1998; NaCl, 19.31 g/L; KCl 0.65 

 g/L; CaCl.,-2H.,0 1.38 g/L; MgSO^-TH^O 1.73 g/L; 

 Na2S04 0."'38 g/L; HEPES 0.82 g/L;) adjusted to pH 

 7.8, with added glucose (1.0 mg/mL) and two exper- 

 imental treatments (high dose=1.0 x 10^ parasites/ 

 crab; low dose=1.0 x lO"' parasites per/crab, /i = 10, 10 

 respectively). To prepare the inoculum for the exper- 

 imental treatments, 2.0 mL of infected hemolymph 

 were drawn from a donor crab infected with 6.15 

 x 10" parasites/mL (comprising 97% plasmodia; 3% 

 trophonts). The infected hemolymph was diluted 1:1 

 with buffer, centrifuged at 4000 rpm for 10 minutes, 

 the supernatant was decanted, and the cells were 

 resuspended in buffer. The cells were then adjusted 

 to 1.0 X 10'' parasites/mL, centrifuged through two 

 more washes, and serially diluted to attain densities 

 of 1.0 X 10*' parasites/mL and 1.0 x lO'* parasites/mL 

 (for inoculum of 100 pL, 1.0 x 10^ parasites/crab and 

 1.0 X 10^ parasites/crab, respectively). 



In both experiments, crabs were monitored daily for 

 mortalities. Deaths within the first nine days of each 

 experiment were excluded because of handling stress 

 arising from infrequent, bacterial infections (e.g. John- 

 son, 1976). None of the crabs in the experiments were 

 infected with amoebae, microsporans, or overt bacte- 

 rial infections (but see Welsh and Sizemore, 1985 for 

 background levels of Vibrio spp. in hemolymph of C. 

 sapidus). Ten crabs from each treatment in the mor- 

 tality-! experiment, and all of the crabs in the mortal- 

 ity-II experiment were bled approximately weekly to 

 assess infection status. In the mortality-I experiment, 

 the same ten crabs were bled approximately weekly 

 until they died; other crabs from within the experi- 

 ment were added as replacements. 



Crab hemolymph was taken by using a tubercu- 

 lin syringe (1 mL) with a 25.5-ga. needle from the 

 arthrodial membrane at the juncture of the basis 

 and the ischium of the 5th pereopod (swimming 

 leg). Ethanol (70*7^^) was used to sterilize the site of 

 inoculation and blood letting. Total and differential 

 counts of host hemocytes and estimates of parasite 

 density were obtained from individual crabs with a 

 hemocytometer (Neubauer improved, Bright Line, 

 two counts per crab) with phase contrast microscopy 

 at 400x. Host hemocytes were identified as granu- 

 locytes, semigranulocytes (intermediate cells with 

 relatively few granules, Bodammer, 1978; Johnson, 



1980) and hyalinocytes (cell types defined in Soder- 

 hall and Cerenius, 1992). Hemocyte and parasite 

 densities higher than 1.0 x IC cells/mL were diluted 

 1:5 with buffer and recounted to provide better esti- 

 mates of cell density. For comparative purposes, 

 total hemocyte densities and differential counts from 

 naturally infected male and female crabs were also 

 obtained. 



Parasites were easily distinguished from host cells 

 by using phase contrast microscopy (Fig. 1): uni- 

 nucleate trophonts (9-15 pm) possessed few small, 

 refractile vacuoles and were rounded or amoeboid, 

 without filopodia; multinucleate plasmodia (20-100 

 pm) were slender, vermiform, and motile. The den- 

 sity of infection refers to the number of parasites 

 per mL of hemolymph. Total hemocyte density refers 

 to the number of hemocytes per mL of hemolymph. 

 Mean intensity refers to the mean number of para- 

 sites per quantity of infected host tissue (Margolis et 

 al., 1982). 



Permanent preparations of hemolymph were pro- 

 cessed and stained as described in Messick ( 1994). 

 Briefly, acid-cleaned, poly-1-lysine-coated microslides 

 were smeared with fresh hemolymph, allowed to 

 stand for 2-3 minutes, and fixed in Bouin's fixative. 

 The smears were processed through a routine Harris 

 hematoxylin and eosin-Y procedure ( Humason, 1979, 

 p. 123 without acid destain). 



The early infection experiment consisted of a con- 

 trol group (n=5 crabs) injected individually with 

 100 pL of hemolymph from an uninfected donor crab 

 and an experimental group {n=20) injected with 100 

 pL of hemolymph from a donor crab containing an 

 estimated 4.1 x 10^ parasites/ml (4.1 x 10^ parasites 

 per crab; comprising 79% plasmodia, 21% trophonts). 

 Three days prior to infection, cell counts were con- 

 ducted on all crabs to serve as a benchmark (presa- 

 mple) for before-after comparisons. On days 3, 5, and 

 7 after inoculation, five infected crabs were bled and 

 dissected. Differential cell counts were conducted and 

 tissue samples taken for histological analysis. Tissue 

 samples were processed through a routine hematox- 

 ylin and eosin procedure and included muscle, hepa- 

 topancreas, heart, and, in some cases, foregut. The 

 control crabs were bled and tissue samples taken 10 

 days after injection. 



For statistical analyses, the proportional hazards 

 model with the Weibull distribution was used to 

 examine survival data and associated variables (Cox 

 and Oakes, 1984 ). The Tarone-Ware log-rank test was 

 used to examine differences between survival curves 

 (Wilkinson, 1997). ANOVA was used to analyze rela- 

 tionships in hemocyte densities and proportion of cell 

 type (cell type density divided by total hemocyte den- 

 sity) between inoculated and uninfected crabs. Simi- 



