Scalable cryopreservation of infectious Cryptosporidium hominis oocysts by vitrification

Abstract

Cryptosporidium hominis is a serious cause of childhood diarrhea in developing countries. The development of therapeutics is impeded by major technical roadblocks including lack of cryopreservation and simple culturing methods. This impacts the availability of optimized/standardized singular sources of infectious parasite oocysts for research and human challenge studies. The human C. hominis TU502 isolate is currently propagated in gnotobiotic piglets in only one laboratory, which limits access to oocysts. Streamlined cryopreservation could enable creation of a biobank to serve as an oocyst source for research and distribution to other investigators requiring C. hominis. Here, we report cryopreservation of C. hominis TU502 oocysts by vitrification using specially designed specimen containers scaled to 100 μL volume. Thawed oocysts exhibit ~70% viability with robust excystation and 100% infection rate in gnotobiotic piglets. The availability of optimized/standardized sources of oocysts may streamline drug and vaccine evaluation by enabling wider access to biological specimens.

Fig 1. Permeabilization of C. hominis to water and cryoprotective agent.

a) Exclusion of water from oocysts was achieved under pressure of a hyperosmotic gradient of trehalose and is reported as change in oocyst volume (V, estimated oocyst volume; V₀, starting volume of control oocysts in PBS). The concentration of trehalose (0 M, 0.33 M, 0.5 M and 1 M) is expressed as inverse osmolarity on the X-axis. Dehydration observed in response to the trehalose gradient is substantial (One-way ANOVA; p < 0.0001, F = 200, df = 11), and linear (Pearson`s correlation coefficient; R² = 0.91). Data met requirements of normality (Shapiro-Wilk test; p > 0.21 for all trehalose concentrations) and homoscedasticity (Brown–Forsythe test; p = 0.47) for inclusion in ANOVA analysis. Values indicate mean and error bars indicate standard error (n = 3). b) The kinetics of DMSO toxicity were measured with 2-week-old oocysts in response to time and temperature of DMSO exposure. Oocysts were first dehydrated with 1 M trehalose for 10 min and then treated with a solution of DMSO to achieve a final concentration of 0.5 M trehalose/ 50% DMSO. DMSO-induced mortality was measured by PI inclusion using flow cytometry and was normalized to control oocysts treated with trehalose and PBS in lieu of DMSO. Increased temperature significantly accelerates mortality (Two-way ANOVA; p <0.0001, F = 214, df = 2 with requirements of normality homoscedasticity met: Shapiro-Wilk test; p >0.054 and Brown–Forsythe test; p >0.53), which is consistent with increased permeabilization and intracellular uptake of DMSO. Values indicate mean and error bars indicate standard error (n = 3). c) DIC micrographs in upper panel demonstrate shrink-swell response of oocysts during the multi-stage CPA loading procedure. Oocysts first shrink in response to 1 M trehalose and swell after exposure to DMSO for 2 min at 37°C (solid arrows). Control oocysts in the lower panel exposed to PBS in lieu of DMSO swell only minimally (dashed arrows) due to dilution of trehalose to 0.5 M and thus reduction of osmotic pressure. Scale indicates 5 μm.

Fig 2. Optimization of thermal permeabilization and cryoprotective agent exposure.

Matched 6- and 12-week-old C. hominis oocysts were dehydrated in 1 M trehalose (10 min) and then exposed to 50% DMSO at a) 30°C for 5–30 min or b) 37°C for 1–5 min. Oocyst mortality was measured by flow cytometry based on PI inclusion and was normalized to control oocysts treated with PBS in lieu of DMSO. A different response between oocyst ages is observed (Two-way ANOVA; **p = 0.006, F = 9.8, df = 1, Shapiro-Wilk normality test; p > 0.52, and Brown–Forsythe homoscedasticity test; p > 0.29). In contrast, exposure to DMSO at 37°C leads to rapid increase in toxicity with no differential age response (Two-way ANOVA; p = 0.13, F = 2.4, df = 1, Shapiro-Wilk normality test; p > 0.47, and Brown–Forsythe homoscedasticity test; p > 0.47 for all age and time conditions tested). Values indicate mean and error bars indicate standard error (n = 3). ns = non-significant. c) After exposure to 1M trehalose (10 min) and 50% DMSO (2 min, 37°C) or trehalose alone, functional viability was evaluated microscopically using an excystation assay. Excystation rate was calculated as a percent of oocysts which released sporozoites after 30 min incubation with 0.75% taurocholic acid at 37°C (induced) or in PBS at 37°C (spontaneous). ‘PBS’ control incubated under identical conditions in PBS in lieu of CPA determines baseline excystation. Exposure to DMSO substantially increases induced excystation with minimal effect on spontaneous breakdown. Lines indicate mean (n = 6). d) Micrographs demonstrate the extent of induced excystation in DMSO-treated oocysts in comparison to untreated oocysts. Solid arrows pointing to refractory oocysts indicate full unexcysted oocysts and dashed arrows indicate empty shells of excysted oocysts. Sporozoites released after DMSO-treatment are of comparable quality to those yielded by untreated oocysts. Scale indicates 20 μm.

Fig 3. Cryopreservation of infectious C. hominis oocysts by ultra-rapid vitrification.

a) Silica microcapillary used for vitrification. Scale indicates 1 cm. b) Viability was quantified using microscopy by means of propidium iodide (PI) exclusion in oocysts exposed to 0.5 M trehalose/ 30% DMSO for 5 min, both before (CPA control) and after cryogenic storage (vitrified). Values indicate mean and error bars indicate standard error (n = 6). c) Sporozoites excysted from oocysts vitrified in capillaries remain morphologically similar to fresh controls. Functional viability was assessed based on sporozoite structure, shape and observed motility in reference to fresh and killed control oocysts. Scale indicates 5 μm. d) C. hominis oocysts are infectious to gnotobiotic piglets after 16–40 months of cryogenic storage. To determine minimum infectious dose, piglets were inoculated orally with either 10⁵ (n = 3), 10⁴ (n = 1) or 10³ thawed PI^-oocysts (n = 1) in the presence of controls infected with 10⁵ fresh oocysts (n = 4) and an uninfected control (n = 2). Aside from two inocula dosed at 10⁵ oocysts, which were recovered after 40 months of cryogenic storage, the remainder of inocula were stored for 16 months. Fecal shedding of oocysts was determined daily by microscopic enumeration in 30 fields of acid-fast stained fecal smears examined at 1000x magnification. Values indicate mean of log transformed oocyst counts and bars indicate standard error. Untransformed infectivity data for each individual piglet can be found in S2 Fig. e) Micrographs of hematoxylin and eosin-stained ileal sections from piglets inoculated with oocysts, either fresh or cryopreserved in microcapillaries for 40 months, and from an uninfected control. Arrows indicate intracellular parasite stages located at the apex of enterocytes. Scale indicates 20μm.

Fig 4. High aspect ratio cassette enables vitrification of scalable volumes.

a) The outer cassette dimensions are 25 mm by 75 mm with internal volume of approximately 100 μL. More details on dimensions are provided in S3 Fig. Silicone tabs are utilized as self-healing loading ports and pressure release ports, while unloading is performed by cutting exit ports with sterile scissors. Details of the sample loading and unloading technique are provided in S4 Fig. b) Cross sectional view of the cassette shows two polycarbonate sheets (thickness 180 μm) bonded by a sheet of pressure sensitive adhesive (thickness 86 μm) cut out to form a sample chamber. c) Plunging a sample of 20% DMSO into liquid nitrogen results in ice formation (top panel) and 30% DMSO forms a partial glass (middle panel). A solution of 40% DMSO forms a transparent amorphous glass, revealing the pattern of background surface, as assessed by visual inspection immediately after removing cassette from liquid nitrogen (bottom panel).

Fig 5. C. hominis oocysts vitrified in cassettes are viable and excyst.

a) Prior to vitrification, oocysts were dehydrated in 1 M trehalose for 10 min and then incubated in 0.5 M trehalose/50% DMSO at either 30°C or 37°C for 2-, 5- or 10 min. Oocysts were then immediately loaded into cassettes and rapidly plunged into liquid nitrogen for 30 min. Oocysts were thawed by quickly transferring directly from liquid nitrogen to 40°C water for 10 sec. After removal of CPA by incubation in excess PBS for 30 min, viability was assessed based on PI exclusion and fitness of excysted sporozoites. b) The heat map indicates the percent of positive cryopreservation outcomes in relation to oocyst age and time of incubation in DMSO at a corresponding permeabilization temperature. Cryopreservation outcome was determined by an excystation assay based on sporozoite morphology and motility. Positive outcome is defined by the presence of full-bodied and motile sporozoites of correct curvature, in contrast to thin and non-motile (presumed dead) sporozoites, as demonstrated in DIC micrographs (scale indicates 5 μm). Each box indicates the number of successful trials over the number of total trials attempted. The CPA protocol including 2 min incubation in 0.5 M trehalose/50% DMSO at 37°C results in positive cryopreservation outcome irrespective of oocyst age. c) Oocyst viability was determined microscopically by means of PI exclusion, both before (CPA control) and after cryopreservation (vitrified) using the 2-min protocol of 0.5 M trehalose/50% DMSO exposure at 37°C. No difference in after-cryopreservation viability was observed across age groups (One-way ANOVA; p = 0.59, F = 0.63, df = 3, Shapiro-Wilk normality test; p > 0.16 for all age groups, Brown–Forsythe homoscedasticity test; p = 0.67). Lines indicate mean (n = 8–13).

Fig 6. C. hominis oocysts aged <10 weeks vitrified in cassettes are infectious to gnotobiotic piglets.

a) C. hominis oocysts aged 7–10 weeks originating from two different batches were cryopreserved in cassettes using the 2-min protocol of 0.5 M trehalose/50% DMSO exposure at 37°C. Gnotobiotic piglets were inoculated orally with 500,000 thawed PI^- oocysts (n = 5) in the presence of controls infected with 500,000 fresh, matched oocysts (n = 4) and negative controls (n = 3), of which 2 were uninoculated and 1 was inoculated with parasite frozen in absence of CPA. Fecal shedding of oocysts was determined daily by microscopic enumeration in 30 fields of acid-fast stained fecal smears examined under 1000x magnification. Eighty percent of piglets (4/5) produced a patent infection 1–3 days later than controls infected with fresh parasite. Values indicate mean of log transformed oocyst counts and bars indicate standard error. Untransformed infectivity data for each individual piglet can be found in S7 Fig. b) Hematoxylin and eosin-stained sections collected at 8 dpi from an experimental piglet show evidence of intestinal infection in both piglets infected with cryopreserved oocysts that tested negative and positive by acid fast (AF) staining. Arrows indicate intracellular parasite stages located at the apex of enterocytes. Scale indicates 20μm.