Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

CATP-6, a C. elegans Ortholog of ATP13A2 PARK9, Positively Regulates GEM-1, an SLC16A Transporter

  • Eric J. Lambie ,


    Affiliation Department of Cell and Developmental Biology, Ludwig-Maximillians-University, Munich, Planegg-Martinsried, Germany

  • Pamela J. Tieu,

    Affiliation Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire, United States of America

  • Nadja Lebedeva,

    Affiliation Department of Cell and Developmental Biology, Ludwig-Maximillians-University, Munich, Planegg-Martinsried, Germany

  • Diane L. Church,

    Affiliation Parkinson's Center, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, United States of America

  • Barbara Conradt

    Affiliations Department of Cell and Developmental Biology, Ludwig-Maximillians-University, Munich, Planegg-Martinsried, Germany, CIPSM Center for Integrated Protein Science, Ludwig-Maximilians-University, Munich, Germany

CATP-6, a C. elegans Ortholog of ATP13A2 PARK9, Positively Regulates GEM-1, an SLC16A Transporter

  • Eric J. Lambie, 
  • Pamela J. Tieu, 
  • Nadja Lebedeva, 
  • Diane L. Church, 
  • Barbara Conradt


In previous work, we found that gain-of-function mutations that hyperactivate GEM-1 (an SLC16A transporter protein) can bypass the requirement for GON-2 (a TRPM channel protein) during the initiation of gonadogenesis in C. elegans. Consequently, we proposed that GEM-1 might function as part of a Mg2+ uptake pathway that functions in parallel to GON-2. In this study, we report that CATP-6, a C. elegans ortholog of the P5B ATPase, ATP13A2 (PARK9), is necessary for gem-1 gain-of-function mutations to suppress the effects of gon-2 inactivation. One possible explanation for this observation is that GEM-1 serves to activate CATP-6, which then functions as a Mg2+ transporter. However, we found that overexpression of GEM-1 can alleviate the requirement for CATP-6 activity, suggesting that CATP-6 probably acts as a non-essential upstream positive regulator of GEM-1. Our results are consistent with the notion that P5B ATPases govern intracellular levels of Mg2+ and/or Mn2+ by regulating the trafficking of transporters and other proteins associated with the plasma membrane.


P-type ATPase transporters are widespread within the genomes of both prokaryotes and eukaryotes, and include such well-studied proteins as the Na+, K+-ATPase, and the sarcoplasmic/endoplasmic reticulum Ca2+ pump (SERCA) [1], [2]. Eukaryotic P-type ATPases have been classified into five groups, P1-P5 [3], with the P5 subfamily the least well understood. The P5 ATPases are represented in plants, animals and fungi [4] and they have been divided into two subcategories, P5A and P5B based on amino acid similarity and structural organization. Furthermore, differences between the signature PPxxP motif within M4 suggest that P5A and P5B ATPases are likely to recognize different transport substrates [5]. Despite their deep evolutionary conservation, the biological functions and transport substrates of the P5 ATPases remain uncertain. In this paper, we will focus on the P5B ATPases.

The S. cerevisiae genome encodes a single P5B ATPase, Ypk9, which is localized to the vacuole membrane and has been suggested to transport Mn2+ into the vacuole [6], although no direct demonstration of this activity has been reported. In Dicytostelium discoideum the sole P5B ATPase, Kil2, is expressed on the phagosomal membrane and is required for Mg2+-dependent killing of ingested Klebsiella; this has prompted the suggestion that Kil2 pumps Mg2+ into the phagosome [7].

The C. elegans genome contains three P5B ATPase genes: catp-5, catp-6 and catp-7, which are likely derived from a single ancestral gene that was present at the time of origin of the nematode phylum (the genomes of many other nematodes, e.g., Ascaris and Brugia, contain only a single P5B ATPase gene; WormBase WS238). The Caldwell group identified CATP-6 in a high-throughput screen for factors that prevent aggregation of human alpha-synuclein heterologously expressed in C. elegans body muscles [8]. Subsequently, CATP-6 was shown to be able to partially substitute for Ypk9p in S. cerevisiae, implying that it might be able to act as a Mn2+ transporter [6]. CATP-5 localizes to the apical brush border of C. elegans intestinal epithelial cells [9]. Inactivation of catp-5 confers resistance to the toxic polyamine analog, norspermidine, and also impairs uptake of polyamines by the intestinal cells; therefore, Heinick et al. (2010) have suggested that CATP-5 might be a polyamine transporter. No characterization of CATP-7 has been reported.

The mouse and human genomes encode four P5B ATPases, ATP13A2-ATP13A5 [10]. Potential biological functions have been suggested for each of these except ATP13A5. ATP13A3/AFURS1 was found to be upregulated in senescent cultured human parenchymal kidney cells, suggesting an association with cellular aging [11]. A chromosomal rearrangement that disrupts ATP13A4 was found to be associated with autism, suggesting a potential role in neural development [12]. Vallipuram et al. (2010) subsequently reported that ATP13A4 localized to the endoplasmic reticulum (ER) when expressed in COS-7 cells, and that overexpression of ATP13A4 resulted in increased cytosolic levels of free Ca2+ [13].

ATP13A2 (also known as PARK9) has been the most intensively studied human P5B ATPase, since mutations in this gene lead to juvenile onset Parkinson disease [14], [15]. ATP13A2 localizes to lysosomal membranes when expressed in human tissue culture cells [15], suggesting that it might perform a function comparable to that of Ypk9p and CATP-6, e.g., sequestering Mn2+ in a vesicular compartment, and thus protecting dopaminergic neurons from alpha- synuclein aggregation and cellular dysfunction [6], [8]. A series of recent reports have suggested that ATP13A2 function is important for general aspects of lysosome function [16] and neuronal physiology [17]. Based on high-throughput interactome studies, Usenovic et al. suggested that ATP13A2 directly interacts with components of the vesicular trafficking machinery [18], and Covy et al. [19] have proposed that ATP13A2 is required for lysosomal import of an unspecified physiological cofactor.

In this paper, we describe our identification of catp-6 as a locus that genetically interacts with the gon-2 and gem-1 loci of C. elegans. gon-2 encodes a TRPM cation channel protein that is required for Mg2+ uptake by the intestinal epithelial cells, and likely also by the somatic gonadal precursor cells, Z1 and Z4 [20], [21]. Inactivation of gon-2 leads to a severe gonadogenesis defect, and in some animals Z1 and Z4 fail to undergo any divisions [22]. gem-1 encodes an SLC16A major facilitator transporter protein [23]. Although the founding members of the SLC16A family are well-characterized as monocarboxylate transporters, the identities of the transport substrates for most proteins in this family are not known [24]. In previous work, we found that hypermorphic gain-of-function alleles of gem-1 are able to bypass the requirement for gon-2 in gonadogenesis [23]. We also showed that expression of gem-1 within Z1 and Z4 is sufficient to rescue the gonadogenesis defect of gon-2 mutants. Therefore, we proposed that hyperactivation of GEM-1 leads to increased activity of a parallel Mg2+ uptake system within the gonadal precursors. In this paper, we demonstrate that CATP-6 activity within Z1 and Z4 is necessary and sufficient in order for gem-1(gf) to suppress gon-2(lf). Furthermore, we find that overexpression of GEM-1 is able to bypass the requirement for CATP-6 during gonadogenesis. Therefore, we propose that CATP-6 normally promotes the activity of GEM-1, which then acts to increase the level of free Mg2+ within the gonadal precursors.


Nematode culture and genetic manipulation

Nematodes were maintained on NGM plates with the E. coli strain AMA1004 [25] as food source. With the exception of SNP mapping experiments, all strains used were in an N2 Bristol background. The wild strain CB4856 was used for SNP mapping experiments, essentially as described by Jakubowski et al. [26]. Standard methods were used for strain constructions [27]. PCR, sometimes in conjunction with DNA sequencing, was done to validate genotypes as necessary.

Molecular biology

Standard methods were used for DNA analysis and manipulation. Oligonucleotides were obtained from Eurofins. For large PCR products we used either Phusion or LongAmp DNA polymerase (New England Biolabs). We obtained a catp-6::gfp fosmid clone derivative of WRM067B_F08 from the C. elegans TransgeneOme project [28] and used this for transformation rescue and expression analyses. We obtained very similar results when we used in vivo recombination between fosmid WRM067B_F08 and a PCR fragment to generate a C-terminally tagged version of catp-6 that lacked the catp-6 3′UTR. However, since expression from the TransgeneOme clone was typically brighter, we used this construct for all of the analyses detailed below.

Isolation of catp-6(lf) mutations

Hermaphrodites of genotype gon-2(q388); gem-1(gf) were mutagenized with 50 mM EMS for 4 hrs [27]. F1 progeny were plated in groups of three in 12-well plates containing NGM solid medium seeded with E. coli and allowed to produce F2 progeny at 23.5°. Plates were then screened for wells that contained multiple gonadless individuals. In total, we screened approximately 3,000 mutagenized genomes. Fertile siblings were then cloned at 15°, and F3 progeny were subcloned and analyzed to identify derivatives that were homozygous for catp-6(lf).

Transgenic strains

Standard microinjection methods were used to generate and maintain transgenic animals. Plasmid pRF4 [29], which contains the dominant Roller marker, rol-6(su1006), was used in all injection mixes at a concentration of approximately 150 μg/ml. Expression/rescue constructs were injected at a concentration of approximately 50 μg/ml.

The oligonucleotide primer pairs and templates used to generate DNA segments used for injection mixes were as follows:

o1594 (GGCCCCAAATAATGATTTTATTTTGCGGGTGGCGCACGACGC) plus o1843 (aggtcgtcccgaatgttctg) were used to amplify the entire catp-6 transcription unit, plus sequences flanking the 3′ UTR from the recombineerome fosmid. The underlined section of o1594 corresponds to the sequence from −23 to −7, relative to the initiation codon for catp-6. The first 25 nucleotides of o1594 provide homology for in vivo recombination with PCR-amplified promoter segments (see below). o1843 is identical to nucleotides +406 to +387 relative to the stop codon for catp-6.

o1587 (TCGCGTTAACGCTAGCATGGATCTCGAAGCTTGGGCTGCAGGTCGG) plus o1588 (CAAAATAAAATCATTATTTGGGGCC TTGGGTCCTTTGGCCAATCC) were used to amplify the myo-3 promoter from pPD96.52 (Fire Lab 1995 Vector Kit). The underlined section of o1587 is a forward primer at the 5′ end of the promoter segment. The underlined section of o1588 is a reverse primer at the 3′ end of the promoter segment. The first 25 nucleotides of o1588 provide homology for recombination with the catp-6::gfp PCR product.

o1725 (AAGAGGTCCCGCTCCAACAAC) plus o1600 (CAAAATAAAATCATTATTTGGGGCC TTTGTAATTTGGAAGCTGGGAGGAATA) were used to amplify the ehn-3 promoter from wild type genomic DNA. o1725 is a forward primer at the 5′ end of the promoter. The underlined section of o1600 is a reverse primer at 3′ end of promoter. The first 25 nucleotides of o1600 provide homology for recombination with catp-6::gfp.

o1880 (ttgagccaatttatccaagtcc) and o1881 (CAAAATAAAATCATTATTTGGGGCC atcggtttggttggaagcgg) were used to amplify the unc-119 promoter from pCFJ150 (Addgene plasmid 19329) [30]. o1880 is a forward primer at the 5′ end of the promoter, whereas o1881 is reverse primer that includes homology for recombination with catp-6::gfp as described above.

Microscopy and imaging

Worms were mounted on 4% agarose pads for DIC and epifluorescence microscopy. In most cases, 10 mM levamisole was used to immobilize animals. Images were acquired using either a ) Zeiss Imager.M2 equipped with appropriate optics and a CCD camera connected to a PC running Time to Live (Caenotec) software, or b) a Zeiss Axioskop 2, equipped with appropriate optics and a PC running MetaMorph (Molecular Devices) software. Brightness and contrast of images was adjusted in Image J and/or PowerPoint.

Assessment of gonad development in different genetic backgrounds

The presence of a vulva was used to determine whether animals had completed the initial set of gonadal cell divisions required to generate an anchor cell [31]. In most cases, vulva development was well-correlated with the ability of an animal to produce fertilized eggs (fertility); however, animals scored as fertile did not necessarily produce a normal number of viable progeny.

For Tables 1 and 2, L4 stage hermaphrodites were incubated overnight at the indicated temperature, then cut open in M9 buffer the next day in order to release eggs from the uterus. Eggs were then transferred by micropipette to seeded plates and raised at the indicated temperature until adult, at which point gonadogenesis and fertility were assessed.

For transgenic experiments, L4 or young-adult stage Rol animals were transferred to the restrictive temperature for gon-2(q388) (23.5°), and Rol progeny were then scored 3–5 days later, as they reached adult stage. Animals were scored as fertile if they had multiple fertilized eggs in the uterus. This represents a more stringent measure of successful gonad development than the presence/absence of a vulva. The vast majority of such animals had normal vulva development.


Identification of catp-6 mutations as suppressors of gem-1(gf)

In an effort to identify null alleles of gem-1, we screened for EMS-induced mutations that reverted the gem-1(gf) phenotype, i.e., restored the Gon phenotype of gon-2(q388); gem-1(dx66) animals (see methods). Unexpectedly, four of the five mutations that we recovered did not map to the gem-1 locus, but instead were located on chromosome IV. These four mutations (dx110, dx112, dx113 and dx114), are all recessive and fail to complement each other. Since dx114 appeared to be the most penetrant, we chose this allele for further characterization. First, through standard two and three-factor mapping, we determined that dx114 is located between unc-24 and dpy-20 on chromosome IV. Next, through a series of SNP mapping experiments [26], we narrowed the location of dx114 to a 120 kb interval (Figure 1). Since we obtained mutations in the dx114 complementation group at a relatively high frequency, we sequenced the coding sequences of the largest predicted gene in this region, W08D2.5 ( =  catp-6). Consequently, we identified a single G -> A missense mutation associated with each of our mutant alleles.

Figure 1. SNP mapping of catp-6.

Recombinant progeny were analyzed from animals of three different genotypes: a) gon-2(q388); unc-24(e138) catp-6(dx114)/CB4856; gem-1(dx66gf), b) gon-2(q388); catp-6(dx114) dpy-20(e1282)/CB4856; gem-1(dx66gf) (top panels), and c) gon-2(q388); catp-6(dx114) unc-43(e408)/CB4856; gem-1(dx66gf) (bottom panel). Not all SNPs were analyzed for each recombinant. Distances between SNPs, but not flanking markers, are to scale.

According to WormBase (WS238), three different isoforms of CATP-6 are expected to be derived from the catp-6 locus, each from a distinct mRNA. CATP-6a is 1256 aa in length, and has the typical structure of a P5B ATPase: eleven transmembrane segments (M0, plus M1-M10), with a relatively large cytoplasmic loop between M4 and M5 (Figure 2). The transcript for CATP-6c begins slightly 3′ relative to CATP-6a, resulting in a protein of 1207 aa that has the same overall structure. CATP-6b is significantly shorter, with a predicted length of 893 aa beginning just before M3. We have not attempted to determine whether each of these different isoforms is functional.

Figure 2. Locations of amino acids affected by mutant alleles of catp-6 relative to conserved domains.

The P (phosphorylation) domain, N (nucleotide binding) domain and A (actuator) domains are indicated. The transmembrane domains (M0-M10) are numbered 0-10. Relative locations of mutant alleles are indicated, including left breakpoint of ok3473 deletion allele. Sizes of different domains are not strictly to scale.

Three of the four mutant alleles that we identified affect residues situated within the large cytoplasmic domain between M4 and M5 (Figures 2 and 3). dx113 converts a highly conserved glycine in the sequence LHGDP to a valine; this glycine is predicted to be situated between the first two helices of the N domain, and is immediately adjacent to residues that interact with Mg2+/ATP [32]. Therefore, dx113 is likely to interfere with nucleotide binding. dx114, converts an invariant glycine in the middle of the sequence GDGAN to a glutamate. This glycine is within one of the essential Mg2+/ATP binding sites, so this mutation is also expected to disrupt nucleotide binding [32]. dx110, the third cytoplasmic mutation converts a threonine in the sequence GPTFA to an isoleucine; this residue is neither highly conserved nor expected to be in close proximity to the nucleotide binding site, but its alteration might disrupt the structure/function of the P domain. sThe sole mutation that affects one of the transmembrane domains, dx112, converts the highly conserved VPPALP sequence within M5 to VLPALP. Since this sequence is predicted to create the substrate binding pocket [33], dx112 is expected to alter or severely impair transport activity.

Figure 3. Locations of catp-6 mutant alleles in comparison to other P-type ATPases.

Alignments were done in TCoffee [35], followed by formatting in BoxShade. Accession numbers for protein sequences used in alignments are as follows: ATP13A1 NP_065143, ATP13A2 NP_001135445, ATP13A3 NP_078800, ATP13A4 NP_115655, ATP13A5 NP_940907, Na/K ATPase NP_001172014, CATP-5 NP_001024768, CATP-6 NP_001024768, CATP-7 NP_001023542, MgtA WP_020898295, SERCA NP_777613, Ypk9 NP_014934.

As further verification of gene identity, we obtained the C. elegans Knockout Consortium allele, catp-6(ok3473), and determined that it defines a 934 bp deletion within catp-6. Both of the endpoints of ok3473 are located within exons, but since the junction results in a reading frame shift the last correctly coded amino acid is Ser765 (CATP-6a numbering); a stop codon occurs after 60 incorrectly coded amino acids (Figure 2). ok3473 is thus expected to be a null allele; not only would the mRNA encode a protein lacking more than half of the transmembrane domains, but the mRNA is also expected to be destabilized due to nonsense-mediated decay [34]. Consistent with these expectations, we found that ok3473 prevents gem-1(dx66gf) from suppressing gon-2(q388), and its phenotype closely resembles that of dx114. WormBase (WS238) describes the catp-6(0) phenotype as embryonic lethal or sterile, based on the deletion allele, tm3190. We obtained the tm3190 mutation from the Mitani laboratory and found that tm3190 homozygotes closely resemble dx114 and ok3473 homozygotes. Therefore, the assignment of tm3190 as lethal/sterile evidently is due to the poor growth and low brood size of catp-6(0) animals.

In addition to blocking suppression of gon-2(q388) by gem-1(gf), elimination of catp-6 activity also causes an obvious Gro (slow growth) phenotype; postembryonic development takes approximately 20% longer than in wild type. Furthermore, catp-6(0) animals are almost always Egl (egg-laying defective). We have not characterized either of these phenotypes in detail, although they are exhibited by both dx114 and ok3473.

Effects of different genotypes on gonadogenesis

We generated a series of single, double and triple mutant strains to investigate the effects of different genetic combinations on gonadogenesis (Tables 1 and 2). Initiation of gonadal cell divisions was not blocked in either the catp-6(0) or gem-1(0) single mutants, or in the catp-6(0); gem-1(0) double mutant. However, the catp-6(0) and gem-1(0) mutations clearly enhance the gon-2(q388) mutant phenotpe when animals are raised at 20°, the upper range of permissive temperature for gon-2(q388) (Table 1). gem-1(0) enhances gon-2(ts) more strongly than does catp-6(0), and the degree of enhancement is not further increased in the gon-2(ts); catp-6(0); gem-1(0) mutant (Table 1). Unexpectedly, catp-6(0) exhibits slight suppression of gem-1(0) at 23.5° (Table 2, lines 5 vs 6, and 7 vs. 8; see Discussion). The gain-of-function mutation, gem-1(dx66gf), appears to exhibit weak suppression of gon-2(q388), even in a catp-6(0) background (Table 2, line 6 vs. 9); however, the difference observed is not very large, and the opposite trend is observed at 20°. Possibly, gem-1(dx66gf) is more active at 23.5°, or it could be that it simply doesn't have much of an effect compared to gem-1(+) in a catp-6(0) background.

The gem-1(dx66gf) mutation alters a residue located within the last of the twelve transmembrane domains of GEM-1, whereas some of the other gem-1(gf) mutations that we isolated are situated within the large cytoplasmic loop situated between transmembrane segments 6 and 7 [23]. Since different alleles of gem-1 might affect different aspects of gem-1 regulation, we tested whether catp-6(0) can block suppression of gon-2(ts) by two of the cytoplasmic loop alleles, dx69gf and dx75gf. We found that catp-6(0) also prevents these alleles from efficiently suppressing gon-2(q388) (Tables 1 and 2); however, both of these alleles appear to be able to weakly suppress catp-6(0) at 20°, and dx75gf also appears to weakly suppress catp-6(0) at 23.5°.

Expression pattern of catp-6::gfp

In order to investigate the expression pattern of catp-6, we used a modified version of fosmid WRM067B_F08 obtained from TransgeneOme project in which the gfp coding sequence is fused to the 3′ end of catp-6, separated by a short linker region [28].

We found that CATP-6::GFP is expressed in multiple tissues throughout development. These include a) many neurons in the head and tail (Figure 4), b) all body muscles (Figure 5), c) most pharyngeal cells, particularly in the posterior bulb (Figure 4), d) vulval muscles, e) coelomocytes, f) spermatheca, g) gonadal sheath cells (Figure 6), and h) lateral hypodermis. In many cases, particularly neurons, the fusion protein localizes to cytoplasmic puncta that probably correspond to membranous vesicles (Figures 4 and 5). In other tissues, e.g., pharyngeal cells and gonadal sheath cells, CATP-6::GFP is closely associated with the plasma membrane (Figures 4 and 6). Most significantly with regard to the effect of catp-6(0) on gonadogenesis, CATP-6::GFP is associated with the plasma membrane of the somatic gonad precursor cells, Z1 and Z4 (Figure 7).

Figure 4. Expression of Pcatp-6::catp-6::gfp in the adult head.

A, DIC, B, GFP. Genotype gon-2(q388); catp-6(ok3473); Ex [Pcatp-6::catp-6::gfp;rol-6(d)] Arrows indicate two representative neuronal cell bodies. Arrowhead indicates posterior pharyngeal bulb.

Figure 5. Expression of Pcatp-6::catp-6::gfp in adult body muscle.

A, DIC, B, GFP. Genotype gon-2(q388); catp-6(ok3473); Ex [Pcatp-6::catp-6::gfp;rol-6(d)]Arrowheads indicate regions where body muscles abut each other. Arrows indicate two neuronal cell bodies. Bright globular patches of fluorescence are autofluorescent gut granules.

Figure 6. Expression of Pcatp-6::catp-6::gfp in adult gonad.

A, DIC, B, GFP. Genotype gon-2(q388); catp-6(ok3473); Ex [Pcatp-6::catp-6::gfp;rol-6(d)]. Arrowheads indicate gonadal sheath cell membrane expression. Arrows indicate sheath cell outlining of germ cells undergoing apoptosis. Bright globular patches of fluorescence are autofluorescent gut granules.

Figure 7. Expression of catp-6::gfp in the L1-stage gonad.

Genotype gon-2(q388); catp-6(ok3473); Ex [Pcatp-6::catp-6::gfp;rol-6(d)]. A, DIC B, GFP. Bright globular patches of fluorescence are autofluorescent gut granules.

CATP-6 expression within Z1 and Z4 rescues gonadogenesis

Since gem-1 and catp-6 interact genetically, the simplest scenario would be that both genes act within the same cell type, i.e, Z1 and Z4. Indeed, we found that when we used the ehn-3 promoter to drive catp-6::gfp expression within Z1 and Z4 we were able to rescue the catp-6(0) phenotype (Table 3). The ehn-3 promoter also drives expression in a small number of neurons in the head and tail region (Figure 8), so it remained formally possible that catp-6 functions within these cells, rather than the somatic gonad precursors. Therefore, we also tested whether driving catp-6 using the pan-neuronal unc-119 promoter could rescue catp-6(0). Although we did observe widespread expression of catp-6::gfp within the nervous system (Figure 9), this did not result in rescue of the catp-6(0) phenotype (Table 3). Similarly, when we used the myo-3 promoter to drive catp-6::gfp in body muscles we did not observe any rescuing activity (Table 3), despite successful expression (Figure 10).

Figure 8. L1-stage expression of Pehn-3::catp-6::gfp.

A, DIC, B, GFP. Genotype gon-2(q388); catp-6(0); gem-1(dx66gf); Ex [Pehn-3::catp-6::GFP; rol-6(d)]. Arroweads indicate neurons in the head and tail. Bright globular patches of fluorescence are autofluorescent gut granules.

Figure 9. L1-stage expression of Punc-119::catp-6::gfp.

A, DIC, B, GFP. gon-2(q388); catp-6(0); gem-1(dx66gf); Ex [Punc-119::catp-6::GFP; rol-6(d). Arrowheads indicate representative neurons. Bright globular fluorescence represents autofluorescent gut granules.

Figure 10. L1-stage expression of Pmyo-3::catp-6::gfp.

A, DIC, B, GFP. gon-2(q388); catp-6(0); gem-1(dx66gf); Ex [Pmyo-3::catp-6::GFP; rol-6(d). Arrowheads indicate body muscles. Bright globular fluorescence represents autofluorescent gut granules.

Independent expression and localization of GEM-1 and CATP-6

Since gem-1(0) and catp-6(0) each enhance gon-2(ts) (Tables 1 and 2), their actions could potentially be explained by a simple regulatory relationship in which one gene acts upstream of the other. Given that each gene encodes a membrane protein expressed within Z1 and Z4, one simple possibility would be that one of the proteins acts to recruit the other to the plasma membrane. We tested this possibility by examining the expression/localization of GEM-1::GFP in a catp-6(0) background, and CATP-6::GFP in a gem-1(0) background. We found that GEM-1::GFP associated normally with the plasma membrane of Z1 and Z4 in a catp-6(0) background (Figure 11), as did CATP-6::GFP in a gem-1(0) background (Figure 12). Therefore, neither protein is strictly dependent on the activity of the other in terms of expression or subcellular localization. However, since each fusion construct is present on an extrachromosomal array, we cannot fully exclude the possibility that normal regulatory constraints might be overwhelmed by overexpression from the transgene. Furthermore, higher resolution imaging would be necessary to detect subtle changes in subcellular protein localization.

Figure 11. L1 stage expression of Pcatp-6::catp-6::gfp in a gem-1(0) background.

A, DIC, B, GFP. Genotype gon-2(q388); gem-1(bc364); Ex [Pcatp-6::catp-6::gfp; rol-6(d)].

Figure 12. Expression of gem-1::gfp in the first larval stage gonad.

A and C, DIC, B and D, GFP. A, C, Genotype gon-2(q388); gem-1(bc364); Ex [gem-1::gfp; rol-6(d)]. B, D, Genotype gon-2(q388); catp-6(ok3473); gem-1(dx66gf); Ex [gem-1::gfp; rol-6(d)]. Arrowead in D indicates overexposed body muscle membrane fluorescence. The animal in panels C and D has more gut granules and a larger nucleolus in Z4, because it is later in the L1 stage than the animal in A and B. Bright globular patches of fluorescence are autofluorescent gut granules.

Effects of overexpression of CATP-6 and GEM-1

Using the transgenic strains described above, we tested whether expression of CATP-6::GFP could bypass the requirement for gem-1(+). As discussed above, since the fusion protein is encoded on a multicopy extrachromosomal array, it is likely that its expression level exceeds that of the wild type CATP-6 protein. However, we observed no rescue of the gem-1(0) phenotype in the transgenic animals (Table 3). This is in contrast to our observations regarding GEM-1::GFP. In this case, we did observe rescue of the catp-6(0) phenotype, suggesting that high level expression of GEM-1 can bypass the requirement for CATP-6 (Table 3).


Our observation that catp-6 activity is required in order for gem-1(gf) to suppress gon-2(lf) could potentially be explained by various different scenarios. Based on the known ability of the prokaryotic P-type ATPases, MgtA and MgtB to mediate Mg2+ uptake, a simple model for the relationship between GEM-1 and CATP-6 would be that GEM-1 exerts some form of positive regulation on CATP-6, which then pumps Mg2+ into the cytoplasm of Z1 and Z4. Although this model is attractive because of its simplicity, several lines of evidence suggest that it is probably not correct. First, the most similar eukaryotic P-type ATPases, ypk9 and ATP13A2, are thought to pump divalent cations- into intracellular vesicular compartments (i.e., out of the cytoplasm). Second, with regard to gonadogenesis, the phenotype of gem-1(0) is more severe than that of catp-6(0). Third, neither expression nor subcellular localization of CATP-6::GFP is strictly dependent on GEM-1 activity. Fourth, overexpression of GEM-1::GFP is able to bypass catp-6(0). However, none of these individual arguments is completely definitive: In the first case, neither the ion specificity nor the directionality of transport of P5B ATPases has been directly measured. In the second, third and fourth cases, it is possible that one of the paralogous genes, catp-5 or catp-7, is able to provide compensatory activity in the absence of catp-6 function.

Since GEM-1 appears to be capable of functioning even in the absence of CATP-6 activity, we consider it most likely that CATP-6 acts as a positive regulator of GEM-1 that is dispensable under conditions of excess GEM-1 protein (Figure 13). Possible explanations for such a positive regulatory interaction include, but are not limited to, the following: a) CATP-6 could promote the proper targeting of GEM-1 to the plasma membrane, either via a direct protein-protein interaction, or by regulating vesicular trafficking, b) CATP-6 activity could be required to maintain normal lysosome function; dysfunction of lysosomes could lead to inappropriate sequestration and/or degradation of GEM-1, c) CATP-6 could pump Mg2+ into an intracellular compartment from which it is later released by GEM-1. In this case, overexpression of GEM-1 on the plasma membrane might permit sufficient Mg2+ import to render access from intracellular compartments unnecessary, d) CATP-6 could act as a polyamine importer, or positively regulate a polyamine transporter, (as proposed for CATP-5), and polyamines could promote GEM-1 activity, e) CATP-6 and GEM-1 could directly interact to form a Mg2+ importer, with CATP-6 acting as a non-essential, regulatory subunit.

Figure 13. Model of possible regulatory relationships between CATP-6, GEM-1, GEM-4 and GON-2.

Arrows indicate positive regulation and “roadblocks” indicate negative regulation.

gon-2(lf); gem-1(0) hermaphrodites exhibit a highly penetrant gonadogenesis defect that is weakly suppressed by inactivation of catp-6. One possible explanation for this suppression is that CATP-6 might also be a positive regulator of GEM-4 (Figure 13). We previously found that inactivation of GEM-4 partially suppresses the gonadogenesis defect of gon-2(lf); gem-1(0) animals, possibly by relief of negative regulation of GON-2. Since GEM-4 also associates with the plasma membrane of Z1 and Z4, CATP-6 could potentially affect GEM-4 function either by a direct interaction, or indirectly via alteration of vesicular trafficking. A comparable situation might also exist in the case of CATP-5, where it could be that this protein exerts its effects on polyamine uptake by regulating the association of a separate tansporter protein with the plasma membrane.

Although we detect CATP-6::GFP in close association with the plasma membrane in some tissues, we cannot be certain that the protein is actually located within the plasma membrane. For example, in the case of Z1 and Z4 the fluorescence pattern of CATP-6::GFP (unlike that of GEM-1::GFP) is often discontinuous along the periphery of the cell. Thus, it remains possible that the protein localizes to vesicles that are just beneath the plasma membrane. Furthermore, it is possible that CATP-6 localizes to both plasma membrane and vesicular compartments and that the relative distribution differs between cell types. Higher resolution microscopy will be necessary to resolve this issue.

Clearly, further work must be done in order to distinguish between the many possible models for the interactions between CATP-6 and GEM-1, as well as to resolve the issues of transport specificity of each of these proteins. In the case of GEM-1, we have no direct evidence that the protein functions as a transporter; however, its similarity to known transporters makes it highly likely that this is the case. For CATP-6, our findings that one of the alleles that we isolated (dx112) alters a residue that is predicted to directly interact with substrate and another (dx114) affects a residue expected to be critical for enzymatic activity suggest that this protein is very likely to function as a typical P-type ATPase transporter protein. Although not unexpected, this was not a foregone conclusion, e.g., Ruaud and Bessereau (2007) have shown that C. elegans CATP-1, a type IIC P-type ATPase, has important functions that are independent of its catalytic/transport activity.


The authors wish to acknowledge Jessica Whitaker for excellent technical assistance, Sonia Yuen for the gon-2(q388); catp-6(dx114); gem-1(dx69gf) strain, the Caenorhabditis Genetics Center for various strains, Bob Barstead and the C. elegans Knockout Consortium for catp-6(ok3473), Shohei Mitani and the National Bioresource Project of Japan for catp-6(tm3190), Mihail Sarov and the C. elegans TransGeneome Project for the catp-6::gfp fosmid, and Andrew Fire for plasmids.

Author Contributions

Conceived and designed the experiments: EJL PJT NL DLC BC. Performed the experiments: EJL PJT NL DLC. Analyzed the data: EJL PJT NL DLC BC. Contributed reagents/materials/analysis tools: EJL PJT NL DLC BC. Wrote the paper: EJL PJT NL DLC BC.


  1. 1. Chan H, Babayan V, Blyumin E, Gandhi C, Hak K, et al. (2010) The P-type ATPase superfamily. J Mol Microbiol Biotechnol 19: 5–104.
  2. 2. Palmgren MG, Nissen P (2011) P-type ATPases. Annu Rev Biophys 40: 243–266.
  3. 3. Axelsen KB, Palmgren MG (1998) Evolution of substrate specificities in the P-type ATPase superfamily. J Mol Evol 46: 84–101.
  4. 4. Moller AB, Asp T, Holm PB, Palmgren MG (2008) Phylogenetic analysis of P5 P-type ATPases, a eukaryotic lineage of secretory pathway pumps. Mol Phylogenet Evol 46: 619–634.
  5. 5. Sorensen DM, Buch-Pedersen MJ, Palmgren MG (2010) Structural divergence between the two subgroups of P5 ATPases. Biochim Biophys Acta 1797: 846–855.
  6. 6. Gitler AD, Chesi A, Geddie ML, Strathearn KE, Hamamichi S, et al. (2009) Alpha-synuclein is part of a diverse and highly conserved interaction network that includes PARK9 and manganese toxicity. Nat Genet 41: 308–315.
  7. 7. Lelong E, Marchetti A, Gueho A, Lima WC, Sattler N, et al. (2011) Role of magnesium and a phagosomal P-type ATPase in intracellular bacterial killing. Cell Microbiol 13: 246–258.
  8. 8. Hamamichi S, Rivas RN, Knight AL, Cao S, Caldwell KA, et al. (2008) Hypothesis-based RNAi screening identifies neuroprotective genes in a Parkinson's disease model. Proc Natl Acad Sci U S A 105: 728–733.
  9. 9. Heinick A, Urban K, Roth S, Spies D, Nunes F, et al. (2010) Caenorhabditis elegans P5B-type ATPase CATP-5 operates in polyamine transport and is crucial for norspermidine-mediated suppression of RNA interference. FASEB J 24: 206–217.
  10. 10. Schultheis PJ, Hagen TT, O'Toole KK, Tachibana A, Burke CR, et al. (2004) Characterization of the P5 subfamily of P-type transport ATPases in mice. Biochem Biophys Res Commun 323: 731–738.
  11. 11. Habtemichael N, Kovacs G (2002) Cloning the AFURS1 gene which is up-regulated in senescent human parenchymal kidney cells. Gene 283: 271–275.
  12. 12. Kwasnicka-Crawford DA, Carson AR, Roberts W, Summers AM, Rehnstrom K, et al. (2005) Characterization of a novel cation transporter ATPase gene (ATP13A4) interrupted by 3q25-q29 inversion in an individual with language delay. Genomics 86: 182–194.
  13. 13. Vallipuram J, Grenville J, Crawford DA (2010) The E646D-ATP13A4 mutation associated with autism reveals a defect in calcium regulation. Cell Mol Neurobiol 30: 233–246.
  14. 14. Di Fonzo A, Chien HF, Socal M, Giraudo S, Tassorelli C, et al. (2007) ATP13A2 missense mutations in juvenile parkinsonism and young onset Parkinson disease. Neurology 68: 1557–1562.
  15. 15. Ramirez A, Heimbach A, Grundemann J, Stiller B, Hampshire D, et al. (2006) Hereditary parkinsonism with dementia is caused by mutations in ATP13A2, encoding a lysosomal type 5 P-type ATPase. Nat Genet 38: 1184–1191.
  16. 16. Usenovic M, Krainc D (2012) Lysosomal dysfunction in neurodegeneration: the role of ATP13A2/PARK9. Autophagy 8: 987–988.
  17. 17. Ramonet D, Podhajska A, Stafa K, Sonnay S, Trancikova A, et al. (2012) PARK9-associated ATP13A2 localizes to intracellular acidic vesicles and regulates cation homeostasis and neuronal integrity. Hum Mol Genet 21: 1725–1743.
  18. 18. Usenovic M, Knight AL, Ray A, Wong V, Brown KR, et al. (2012) Identification of novel ATP13A2 interactors and their role in alpha-synuclein misfolding and toxicity. Hum Mol Genet 21: 3785–3794.
  19. 19. Covy JP, Waxman EA, Giasson BI (2012) Characterization of cellular protective effects of ATP13A2/PARK9 expression and alterations resulting from pathogenic mutants. J Neurosci Res 90: 2306–2316.
  20. 20. Teramoto T, Lambie EJ, Iwasaki K (2005) Differential regulation of TRPM channels governs electrolyte homeostasis in the C. elegans intestine. Cell Metab 1: 343–354.
  21. 21. West RJ, Sun AY, Church DL, Lambie EJ (2001) The C. elegans gon-2 gene encodes a putative TRP cation channel protein required for mitotic cell cycle progression. Gene 266: 103–110.
  22. 22. Sun AY, Lambie EJ (1997) gon-2, a gene required for gonadogenesis in Caenorhabditis elegans. Genetics 147: 1077–1089.
  23. 23. Kemp BJ, Church DL, Hatzold J, Conradt B, Lambie EJ (2009) gem-1 encodes an SLC16 monocarboxylate transporter-related protein that functions in parallel to the gon-2 TRPM channel during gonad development in Caenorhabditis elegans. Genetics 181: 581–591.
  24. 24. Halestrap AP, Wilson MC (2012) The monocarboxylate transporter family--role and regulation. IUBMB Life 64: 109–119.
  25. 25. Casadaban MJ, Martinez-Arias A, Shapira SK, Chou J (1983) Beta-galactosidase gene fusions for analyzing gene expression in Escherichia coli and yeast. Methods Enzymol 100: 293–308.
  26. 26. Jakubowski J, Kornfeld K (1999) A local, high-density, single-nucleotide polymorphism map used to clone Caenorhabditis elegans cdf-1. Genetics 153: 743–752.
  27. 27. Brenner S (1974) The genetics of Caenorhabditis elegans. Genetics 77: 71–94.
  28. 28. Sarov M, Murray JI, Schanze K, Pozniakovski A, Niu W, et al. (2012) A genome-scale resource for in vivo tag-based protein function exploration in C. elegans. Cell 150: 855–866.
  29. 29. Mello CC, Kramer JM, Stinchcomb D, Ambros V (1991) Efficient gene transfer in C.elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J 10: 3959–3970.
  30. 30. Frokjaer-Jensen C, Davis MW, Hopkins CE, Newman BJ, Thummel JM, et al. (2008) Single-copy insertion of transgenes in Caenorhabditis elegans. Nat Genet 40: 1375–1383.
  31. 31. Kimble J (1981) Alterations in cell lineage following laser ablation of cells in the somatic gonad of Caenorhabditis elegans. Dev Biol 87: 286–300.
  32. 32. Patchornik G, Munson K, Goldshleger R, Shainskaya A, Sachs G, et al. (2002) The ATP-Mg2+ binding site and cytoplasmic domain interactions of Na+, K+-ATPase investigated with Fe2+-catalyzed oxidative cleavage and molecular modeling. Biochemistry 41: 11740–11749.
  33. 33. Toyoshima C, Nakasako M, Nomura H, Ogawa H (2000) Crystal structure of the calcium pump of sarcoplasmic reticulum at 2.6 A resolution. Nature 405: 647–655.
  34. 34. Hodgkin J, Papp A, Pulak R, Ambros V, Anderson P (1989) A new kind of informational suppression in the nematode Caenorhabditis elegans. Genetics 123: 301–313.
  35. 35. Notredame C, Higgins DG, Heringa J (2000) T-Coffee: A novel method for fast and accurate multiple sequence alignment. J Mol Biol 302: 205–217.