We recently demonstrated that nematodes expressing a non-functional copy of intestinal CHCA-1, the nematode homolog of the mammalian copper transporter Ctr1, fail to accumulate wild-type levels of copper in their body and arrest during larval development when grown under copper deficient conditions (Yuan et al., 2018). Given the functional conservation between nematode CHCA-1 and mammalian Ctr1, its facile manipulation and genetic tractability, we set out to determine if C. elegans is a suitable animal model to explore ES as a potential therapeutic for copper deficiency disorders. To that end, we first asked if ES can rescue the copper deficiency phenotypes of the chca-1(tm6506) mutant strain that we previously characterized (Yuan et al., 2018). Synchronized L1 cultures of chca-1(tm6506)) mutants and wild-type (WT) worms were grown on standard growth media in the absence (Basal) or presence of 100 µM bathocuproinedisulfonic acid (BCS)—a membrane-impermeable copper chelator—or BCS-containing media supplemented with 10 µM copper chloride (CuCl₂) or various concentrations of ES (from 1 to 50 µM). As shown in Figure 1A (panel d), chca-1 mutants grown on BCS-containing plates have a dramatic larval arrest phenotype when compared to mutants grown on control plates (Figures 1A,C) and WT worms grown under either condition. As expected, this larval arrest phenotype is fully rescued by copper supplementation, and importantly, by supplementation with ES in a dose dependent matter (Figure 1A, e-h).

We used COPAS sorting to quantitatively examine the larval phenotype of the chca-1(tm6506) mutant. As shown in Figure 1B, rescue of the larval arrest phenotype—as determined by an increase in body size—is observed at doses as low as 5 µM ES with full rescue observed at 50 µM ES supplementation. Moreover, the ability of ES to rescue the larval phenotype is comparable to rescue observed by copper supplementation. Furthermore, ICP-MS analysis of worms grown under the same conditions demonstrated that whole animal copper levels in the chca-1(tm6506) are fully rescued by ES supplementation (Figure 1C). These results strongly suggest that ES can efficiently restore copper levels and reverse phenotypes caused by copper deficiency even in the presence of the high-affinity copper chelator, BCS. Importantly, these data suggest that C. elegans represents a viable experimental animal model to interrogate the biology of ES and its potential as a therapeutic for copper deficiency disorders.

Our previous studies identified cua-1, the C. elegans homolog of ATP7A/B, as a key regulator of systemic copper homeostasis in the worm (Chun et al., 2017a). While CUA-1 is expressed in the intestine, hypodermis, pharynx, and neurons in C. elegans, cua-1 primarily functions as a copper exporter at the basolateral membrane of the intestine, delivering dietary copper to extraintestinal tissues as needed. Notably, cua-1 mutant model Menkes disease in that they exhibit systemic copper deficiency (Chun et al., 2017a). cua-1 RNAi in the presence of 100 µM BCS results in a severe developmental arrest phenotype (Figures 2A,D). Importantly, this phenotype is rescued by treatment with 5 µM ES even in the absence of exogenous copper supplementation (Figures 2A,H). Worms simultaneously grown in cua-1(RNAi) and in the presence 100 µM BCS that manage to escape arrest and reach reproductive age were interrogated for their ability to produce viable embryos as measured by the hatching rate. As shown in Figure 2B, cua-1 RNAi in the presence of 100 µM BCS results in a severe defect in hatching rate. However, this phenotype can be partially rescued by treatment with 5 µM ES.

A number of groups have shown that cua-1 is expressed in multiple tissues in the worm (Wakabayashi et al., 1998; Chun et al., 2017a). To specifically demonstrate ES rescue of copper transport across the intestinal membrane, we made use of the BK015 worm line [P _vha-6 ::CUA-1.1::GFP::unc-54 3′UTR; cua-1(ok904)] which ectopically expresses a GFP-tagged version of CUA-1 driven by an intestine-specific promoter in a cua-1-null mutant background (Chun et al., 2017a). BK015 worms require added dietary copper for optimal growth but can be maintained with about 50% survival on standard OP50 bacteria without dietary copper amendments (Figure 2C, a-b and D). This morbidity is presumably caused by the lack of CUA-1 in extraintestinal tissues such as pharynx and neurons. Knockdown of cua-1 in BK015 worms results in complete larval lethality (Figures 2C,D). However, this lethality can be rescued almost fully in a dose-dependent manner by 500 nM ES supplementation (Figure 2D). Thus, ES is specifically capable of rescuing defective intestinal copper export, the major hallmark of Menkes disease, in a whole animal model. Notably, 5 µM ES also rescues suboptimal growth of BK015 under basal media conditions (Figures 2C,F), indicating that ES is capable of restoring copper homeostasis defects further downstream of intestinal copper export.

To further assess the effect of ES on organismal copper homeostasis in C. elegans, we analyzed the subcellular localization of CUA-1.1, which is encoded by one of two splice isoforms of cua-1. We have previously shown that CUA-1.1 localizes to the basolateral membrane of the worm intestine under basal conditions. In response to elevated amounts of dietary copper, CUA-1.1 redistributes to lysosome-like gut granules in intestinal cells, likely as a protective measure to sequester potentially toxic excess copper. The other splice variant, CUA-1.2, localizes constitutively to the basolateral membrane irrespective of dietary copper concentration (Chun et al., 2017a). Localization of CUA-1.1 can be perturbed genetically, as loss of chca-1 decreased distribution of CUA-1.1 to granules, indicative of copper deficiency even in the presence of sufficient dietary copper (Yuan et al., 2018). Surprisingly, ES treatment did not appear to influence the subcellular distribution of CUA-1.1 in the gut (Figure 3). This result is intriguing and may suggest an alternative copper uptake pathway exploited by ES that potentially preserves copper in a less toxic form that does not trigger CUA-1.1 re-localization to gut stress granules.

We next aimed to determine if ES was capable of escorting copper across the polarized intestinal epithelium in mammals. Loss of intestinal Ctr1 in mice leads to profound neonatal defects in copper delivery to peripheral tissues, and results in drastically stunted development with mutants exhibiting 100% lethality within 25 days after birth (Nose et al., 2006). Intestine-specific Ctr1 knockout mice (Ctr1 ^int/int ) were administered a 10 mg/kg dose of ES by oral gavage once every 2 days from postnatal day (PND) 7. We found a 35-days survival rate of 100% in ES-treated Ctr1 ^int/int mice, whereas all mutant mice treated with vehicle died between PND 12 and 23 (Figure 4A). ES-treated knockout pups showed normal development and weight gain with no apparent difference as compared to vehicle- or ES-treated WT mice, while all vehicle-treated Ctr1 ^int/int pups failed to thrive, lost weight, and began dying around PND 12 (Figure 4B). These results suggest that ES can deliver available copper from normal dietary sources—namely, milk from mothers fed a regular chow diet.

To further characterize the intestinal phenotype of ES-rescued Ctr1 ^int/int , we isolated IECs from PND 17 mice treated with ES or vehicle, and evaluated the abundance of copper chaperone for superoxide dismutase (CCS), a marker for intracellular copper-availability, as it is known that CCS levels are inversely proportional to cytosolic copper bioavailability (Bertinato and L'Abbé, 2003; Kim et al., 2010). As expected, we observed increased CCS steady-state levels in vehicle-treated Ctr1 ^int/int intestinal cells as compared to levels in cells from control littermates (Ctr1 ^flox/flox ) treated with vehicle or ES. The increased CCS levels in cells from ES-treated Ctr1 ^int/int were suppressed back to levels observed in Ctr1 ^flox/flox mice (Figures 4C,D), indicating that the limited copper availability in the Ctr1-deleted enterocytes was normalized by ES gavage. Hyperaccumulation of copper in the IECs observed in Ctr1 ^int/int (Nose et al., 2006) was also rescued in the ES-treated mutants, while ES did not influence overall zinc levels in the intestine (Figure 4D) suggesting a copper-specific function for ES. Taken together, our results indicate that oral treatment with ES is capable of escorting copper across a CTR1-deficient polarized enterocyte layer.

Bristol N2 was used as the wild-type (WT) C. elegans strain. N2 WT, BK014 transgenic worms [P _vha-6 ::CUA-1.1::GFP::unc-54 3′ UTR], BK015 [Pvha-6::CUA-1.1::GFP::unc-54 3′UTR; cua-1(ok904)] worms, chca-1 (tm6506 IV) and their respective WT broodmates were maintained at 20 °C on NGM plates seeded with E. coli OP50 or HT115 (DE3) as a food source. Some worm strains were obtained from the Caenorhabditis Genetics Center, which is funded by National Institutes of Health Office of Research Infrastructure Programs (P40 OD010440). RNAi bacteria strains against chca-1 or cua-1 were used, and bacteria transformed with the empty L4440 vector were used as a negative RNAi control as described previously (Yuan et al., 2018).

Synchronized L1 larval cultures of WT and chca-1(tm6506) mutant worms were cultured on basal media or media containing 100 µM BCS plus 10 µM CuCl₂ or various concentrations of ES for ∼2.5 days. Worms from each condition were analyzed for time of flight (length), extinction (width), and GFP fluorescence using a COPAS Biosort FP-250 (Union Biometrica) to determine body size and copper content by ICP-MS. To measure brood sizes, N2 worms were exposed to cua-1 (RNAi) or control (L4440 vector) from synchronized L1 stage with or without 100 µM BCS treatment, supplemented with vehicle (DMSO) or 5 µM ES. After 4 days, worms under each condition were picked to individual plates in triplicate, allowed to lay eggs, and transferred to fresh plates every 24 h for 3 days. Eggs were incubated overnight to allow hatching and progeny number was determined as the total number of hatched larvae. For the survival assay, BK015 transgenic worms were cultured from the L1 stage on RNAi plates supplemented with either vehicle (DMSO) or 5 µM ES for 4 days, and then scored for survival. Animals were considered dead when no signs of viability (movement, pharyngeal pumping, or response to prodding) were detected. The ES concentration required to observe rescue in different tests reflects technical requirements for each assay—e.g., type of bacterial food source, bacterial culture requirements, and the nature of the genetic background or transgene present in the worm strains.

Metal content of worms was measured using ICP-MS as described previously (Chun et al., 2017b; Yuan et al., 2018). Values were normalized to the wet weight of worms. Synchronized L1 worms were grown on NGM plates seeded with OP50 supplemented with the indicated amounts of copper, BCS or ES until worms reached the L4 stage. At least three independent biological replicates were analyzed. Copper, and zinc concentrations of mice were measured from nitric acid-digested tissues by ICP-MS as described (Nose et al., 2006). The values were normalized by tissue protein concentration.

BK014 transgenic worms (P _vha-6 ::CUA-1.1::GFP::unc-54 3’ UTR) were maintained on standard NGM agar plates seeded with OP50 bacteria. Synchronized L1 larvae were cultured on NGM plates with indicated amount of CuCl₂, vehicle (DMSO) or ES for 3 days. To visualize live worms, animals were paralyzed in M9 buffer containing 10 mM sodium azide (NaN₃) and mounted on agarose pads. GFP in worms was imaged using an LSM710 confocal microscope (Zeiss).

The intestine-specific Ctr1 deletion mouse (Ctr1 ^int/int ) was generated by crossing Ctr1 ^flox/flox mice with mice expressing Cre recombinase driven by the Villin-promoter as described previously (Nose et al., 2006; Chun et al., 2017b). Age-matched Ctr1 ^flox/+ or Ctr1 ^flox/flox siblings not expressing Cre were served as a control group in this study. All animal procedures were performed in accordance with National Institutes of Health Guide and approved by the Institutional Animal Care and Use Committee at the University of Maryland, College Park (protocol #: R-APR-18-14). Starting from postnatal day (PND) 7, Ctr1 ^int/int and control littermates were treated with 10 mg/kg body weight of ES (Selleckchem) or vehicle by gavage every 2 days until PND 31. Mouse body weights were recorded every day from PND 7 to PND 35. Depending on individual mouse weight, an appropriate amount of ES was dissolved in DMSO, and then mixed with 5% methyl cellulose solution (Sigma) to reach a 2% final concentration of ES-DMSO. The vehicle control solution contained 0.5% methyl cellulose solution with 2% DMSO. A subset of mice were dissected at PND 17 for tissue analysis.

The isolation and extraction of intestinal epithelial cells from mice for immunoblot assays and metal measurements have been described previously (Chun et al., 2017b). The anti-CCS antibody (Santa Cruz Biotechnology) and anti-GAPDH antibody (Sigma) were used at 1:1,000 dilution and 1:5,000 dilution, respectively.

Statistical analysis was performed with one-way ANOVA followed by Sidak post hoc test, or two-way ANOVA followed by Tukey’s post hoc test, using GraphPad Prism (GraphPad). Differences were considered statistically significant at p < 0.05.