Preliminary evaluations of 3-dimensional human skin models for their ability to facilitate in vitro the long-term development of the debilitating obligatory human parasite Onchocerca volvulus

Abstract

Onchocerciasis also known as river blindness is a neglected tropical disease and the world's second-leading infectious cause of blindness in humans; it is caused by Onchocerca volvulus. Current treatment with ivermectin targets microfilariae and transmission and does not kill the adult parasites, which reside within subcutaneous nodules. To support the development of macrofilaricidal drugs that target the adult worm to further support the elimination of onchocerciasis, an in-depth understanding of O. volvulus biology especially the factors that support the longevity of these worms in the human host (>10 years) is required. However, research is hampered by a lack of access to adult worms. O. volvulus is an obligatory human parasite and no small animal models that can propagate this parasite were successfully developed. The current optimized 2-dimensional (2-D) in vitro culturing method starting with O. volvulus infective larvae does not yet support the development of mature adult worms. To overcome these limitations, we have developed and applied 3-dimensional (3-D) culture systems with O. volvulus larvae that simulate the human in vivo niche using in vitro engineered skin and adipose tissue. Our proof of concept studies have shown that an optimized indirect co-culture of in vitro skin tissue supported a significant increase in growth of the fourth-stage larvae to the pre-adult stage with a median length of 816-831 μm as compared to 767 μm of 2-D cultured larvae. Notably, when larvae were co-cultured directly with adipose tissue models, a significant improvement for larval motility and thus fitness was observed; 95% compared to 26% in the 2-D system. These promising co-culture concepts are a first step to further optimize the culturing conditions and improve the long-term development of adult worms in vitro. Ultimately, it could provide the filarial research community with a valuable source of O. volvulus worms at various developmental stages, which may accelerate innovative unsolved biomedical inquiries into the parasite's biology.

Fig 1. Three critical factors that might support long-term in vitro culturing of O. volvulus L4s that better resemble the human in vivo niche, the nodule.

Factor 1: Presence of a 3-D ECM scaffold; Factor 2. Incorporation of skin-specific and subcutaneous cells with the ECM scaffold; and Factor 3. Facilitating direct contact between the larvae and skin tissue models.

Fig 2. Optimized 2-D culture system for O. volvulus.

(A) Schematic 2-D culture system. 1) HUVEC feeder layer; 2) culture insert with porous membrane; 3) L4 larvae loaded into the insert. (B) Loss of 80% of larvae during culture from L3 to early pre-adult stages over 8 weeks of culture using the 2-D culture system. The results presented are an average of 18 batchs of larvae (300–600 L3s per batch) that were cultured and monitored weekly for motility.

Fig 3. Compatibility of the experimental co-culture media on L4 fitness using the 2-D culturing setup with a HUVEC monolayer.

Growth (length, median per group) of larvae and % of L4 of the initial inoculum (N = 18–22 per group divided in two reproducible wells) that were still motile in the presence of experimental co-culture media (3D-1 to 3D-4) in comparison to the 2-D culturing system (2D-ctrl) after 7 days in culture. The number of L4s per group and the % motile L4s in each group of day 7 in culture are indicated below the graph.

Fig 4. Structural analysis of FTSM cultured in the experimental co-culture media.

(A) Schematic of the FTSM comprising epidermis and dermis cultured in (B) cell culture insert on a porous membrane. (C) Histological Hematoxylin and Eosin staining of FTSM in E10 skin medium (3D-1) compared to (D) 3D-3 and (E) 3D-4, the best experimental co-culture media that supported L4 fitness. Scale bar = 100 μm. (1) porous membrane, (2) outer-well medium, (3) dermis, (4) epidermis.

Fig 5. Methods for introducing larvae into a collagen-based tissue model.

(A) Injection of larvae. (B) Creation of a wound with a tissue punch, followed by loading of larvae. (C) Microscopy of L4 inside the undefined wound created within the 3-D model. Scale bar = 200 μm. (D) Homogenous distribution of larvae by their incorporation into the collagen during model creation. (E) Generation of a confined compartment by imprinting a well with a stamp. (F) L4 larvae inside the imprinted compartment within the dermal model on day 2 of culture. The image was taken using light microscopy through the insert membrane and a thin collagen layer, responsible for the blur. (G) Growth of L4 (N = 16 divided between two reproducible tissue models) cultured in direct contact within imprinted compartment in the 3-D models on day 7, 14, 21, and 28 in culture. The significance of differences in growth between D0 and D21 (P = 0.0061: **) and D0 and D28 (P = 0.0065: **) was analyzed using the Mann-Whitney U test. The number of worms, that could be observed is noted below the X-axis.

Fig 6. Indirect 3-D co-culture setup that supports the growth of O. volvulus L4.

(A) FTSM and larvae are co-cultured in separate culture inserts with a schematic representation of hDF migration into the larvae compartment. (B) Image of the two inserts in a 6-well-culture plate containing the 3-D indirect co-culture setup. (C) SEM picture depicting cells that migrated inside the larvae containing insert and grew over the viable O. volvulus L4. Inset shows the interface between L4, cells, and ECM. Scale bar = 20 μm. (D) Growth of 15-days old L4 co-cultured in the indirect 3-D setup using two experimental co-culture media, 3D-3 and 3D-4, versus 2D-ctrl (over a HUVEC monolayer) over 77 days of culture (observations reported are those for day 15, 36, and 92 of worm age). Each experimental condition was set up with 3 technical replicates containing ~10 L4s each. The % motile L4s in each group and time point are indicated below the graph. The significance of differences in growth between days in culture was analyzed using the Mann-Whitney U test (P < 0.0001: ****), and between the groups on day 77 of culture using ANNOVA with Dunnet’s analysis (P ≤ 0.001: ***; P < 0.0001: ****)

Fig 7. Development of an adipose tissue model with an enhanced differentiation efficiency.

(A) Schematic Adipose tissue model cultured on cell culture insert membrane. (B) Adipocyte aggregates stained positive for Oil Red O. (C) Histological Hematoxylin and Eosin staining depicts formation of adipose tissue-like organization containing mature uniocular adipocytes. Scale bar = 50 μm. (D) Tolerance testing of hMSCs cultured in 2D and differentiated in L4 medium supplemented with adipogenic differentiation factors, confirmed by Oil Red O. Scale bar = 100 μm. (E) Oil Red O Quantification of enhanced lipid accumulation by supplemented glucose and lipids (ADM+) compared and normalized to standard differentiation medium (ADM). Unpaired t-test P < 0.0001, N = 2.

Fig 8. Adipose tissue co-culture setups.

(A) Schematic of direct adipose aggregate co-culture with O. volvulus larvae. 1. culture insert with porous membrane; 2. L4 larvae; 3. adipose aggregates; 4. outer well; 5. culture medium. (B) Indirect co-culture setup of larvae and adipose tissues. (C) Microscopic picture of L4 with adjacent adipose tissue aggregate, indicated by dashed line. 1. Porous membrane; 2. L4 larva; 3. adipose aggregate. Scale bar = 200 μm. (D) % Motility of L4 co-cultured in the 2-D culture system (2D-ctrl), over adipose aggregates in indirect contact (soluble), and in direct contact with the adipose aggregates (direct). Two-way ANOVA–Significance tested against the 2-D culture system as the control. 2D-ctrl: N = 27 L4; soluble: N = 33 L4; direct: N = 30 L4.