Revolutionary improvements in resolving protein and cellular structures, and genetic identification of ciliopathies, have created renewed interest in primary cilia. These small protuberances (5–10 μm in length) are found in nearly every cell type. Primary cilia house key downstream elements of the sonic hedgehog pathway, which regulates embryonic development and some cancers. Recently, new tools have enabled measurements of other signaling elements within cilia. Primary cilia have elevated resting internal Ca²⁺ concentrations ([Ca²⁺]=300–700 nM [Delling et al., 2013]), and possess at least two TRP channels, polycystin-2 and polycystin 2-l1. These ion channels are encoded by the PKD2 and PKD2L1 genes of the TRPP subfamily. Electrophysiological measurements demonstrate that polycystin 2-l1 and polycystin-2 both underlie ionic currents in primary cilia (DeCaen et al., 2013, 2016; Kleene and Kleene, 2017). This suggests that voltage gradients and Ca²⁺, in addition to cAMP, are relevant signals within primary cilia.

PKD2 and PKD2L1 share high degrees of sequence identity (52%) and similarity (71%). Mutations in PKD2 account for ~15% of cases of individuals afflicted with Autosomal Dominant Polycystic Kidney Disease (ADPKD). In contrast, there are no diseases currently linked to PKD2L1. Deletion of PKD2 is lethal in mice (Wu et al., 1998), but deletion of PKD2L1 results in occasional gut malrotation, a relatively mild situs inversus phenotype (Delling et al., 2013). Finally, while the functional characterization of polycystin 2-l1 protein has been enabled by its plasma membrane expression (DeCaen et al., 2013, 2016), polycystin-2 protein function has been hampered by its restriction to surface expression in primary cilia. Recent approaches have been successful, however, in establishing the key biophysical properties of polycystin-2. These approaches include a gain-of-function mutation that enables expression to the plasma membrane in Xenopus oocytes (Arif Pavel et al., 2016) and direct patch clamp recording of polycystin-2 in primary cilia (Kleene and Kleene, 2017; Liu et al., 2018). These studies establish polycystin-2 as a monovalent-selective cation channel with little or no calcium permeation. In contrast, polycystin 2-l1 shows significant and relevant Ca²⁺ permeability (P_Ca/P_Na ~6) (DeCaen et al., 2013, 2016). Despite this knowledge, the physiological stimulus for activation and gating is unclear for both channels.

In a similar fashion, structural information about membrane proteins has undergone a profound advance. This is largely due to the explosion of structures from single particle electron cryo-microscopy (cryo-EM) in recent years. Accordingly, multiple structures have been solved for the TRP ion channel family, including the TRPP family to which polycystin-2 and polycystin 2-l1 belong.

Here we present a structure of the full-length human polycystin 2-l1 protein at 3.3 Å resolution in cryo-EM. The overall architecture conforms to that of other TRP channels, and in particular to the TRPP polycystin-2 structures. We establish the core structure and point out differences in regions and residues that may account for polycystin-2 and polycystin 2-l1’s distinct permeation and gating.

We first expressed and purified full-length recombinant polycystin 2-l1 protein tagged with maltose binding protein (MBP) (Figure 1—figure supplement 1) using HEK293 GnT I^- cells and the BacMam system (Goehring et al., 2014). To determine whether the MBP-tagged protein was functional, we measured single channel activity in reconstituted liposomes under voltage clamp. The channel’s conductance in the same buffer used for purification (HKN, see Materials and methods) was 105 pS, consistent with previous single-channel studies from cells expressing polycystin 2-l1 (DeCaen et al., 2013) (Figure 1—figure supplement 1d).

General architecture of polycystin 2-l1

Polycystin 2-l1’s structure exhibits many of the hallmarks of TRP channels: it forms a homotetramer with a domain-swapped Voltage Sensing-Like Domain (VSLD, the S1-S4 transmembrane domains) (Figure 1a) and shares remarkable architectural similarity to polycystin-2 (Figure 1—figure supplement 2) (RMSD of 1.5 Å measured with polycystin-2 model pdb 5T4D). As characteristic of group 2 TRP channels (TRPPs and TRPMLs), polycystin 2-l1 has a long S1-S2 extracellular loop, termed the polycystin mucolipin domain (PMD). This domain, a series of 3 α-helices, 4 β-sheets, and a glycosylated three loop region (see below, TLC or three-leaf clover), forms a cover, or lid, above the channel (Figure 1b). In contrast to the 3 or 4 observed glycans in polycystin-2's polycystin mucolipin domain (PMD) (Grieben et al., 2017; Shen et al., 2016; Wilkes et al., 2017), polycystin 2-l1’s PMD has one clear glycan density located at residue N207. An additional density, at N241, could possibly support an additional glycan. Notably, the PMD interacts with underlying elements of the pore, the voltage sensing-like domains, and adjacent PMDs. The role of glycosylation of the PMD in TRPPs is unknown. Two possibilities include serving as a folding/quality control mechanism or as a site for intermolecular interactions (e.g., ligand involved in gating).

Figure 1 with 5 supplements see all

Download asset Open asset

The intracellular face of polycystin 2-l1 is too poorly resolved to model the N terminus and the S4-S5 linker densities. Also, despite expression of full-length protein (Figure 1—figure supplement 1) the C terminus of polycystin 2-l1 was not resolved. We conclude that these elements are either unstructured or connected with flexible regions, which, along with the lack of cytoplasmic elements in 2D classification in polycystin-2 (Shen et al., 2016), prevents model building of these regions.

The voltage sensing-like domain

The VSLD is comprised of four transmembrane helical spanning elements that, while similar in structure to the voltage-sensing domains of voltage-gated ion channels, do not convert the energy of the transmembrane electric field into pore gating. Indeed, none of the group II TRP channels have significant voltage dependence. Comparison of the VSLD of polycystin 2-l1 and polycystin-2 reveals that the polycystin 2-l1 S2 helix is tilted an additional 4.5° away from the core, is laterally shifted away from the rest of the VSLD, and shows the greatest local RMSD difference (Figure 1—figure supplement 2a, b and c). Polycystin 2-l1’s S3 is a near-continuous helix extending from the membrane-spanning region into a pocket created by the PMD (Figures 1c and 3b). This S3 helix exhibits greater secondary structure than the presumed closed polycystin-2 structure of Shen et al. (Shen et al., 2016), but is similar to that of the multiple- and single-ion models of Wilkes et al., (Wilkes et al., 2017). The S3 extended helix abuts the PMD of the same monomer but does not display the same cation-π interactions as polycystin-2’s F545 and R320 residues. The S3-S4 linker and the top portion of the S4 helix also fit into the same cleft of the PMD as in polycystin-2. In polycystin-2, the S4 adopts a 3₁₀-helix configuration (I571 – F579). The density of the S4 helix of polycystin 2-l1 is incomplete at this region so a 3₁₀ configuration is not observed. Similarly, the equivalent residues of polycystin-2 that are thought to form a salt bridge between the S3 and S4 helix (K572, K575) (Shen et al., 2016), do not have sufficient side chain density to model accurately. The equivalent residue for polycystin-2’s acidic D511 residue is D390 in polycystin 2-l1, which does not form a salt-bridge with neighboring residues.

Figure 2

Download asset Open asset

The polycystin mucolipin domain

The polycystin mucolipin domain (PMD) rests on top of the VSLD of the same monomer and the adjacent pore domain (Figure 3b and c), with a series of three α-helices facing inward towards the funnel/turret of the pore (Figure 3c). A series of β-sheets is sandwiched between these α-helices with several loops that interface with the extracellular space and adjacent PMD (Figures 1b and 3c). Finally, a disulfide bond is present between residues C210-C223 in the PMD, similar to several polycystin-2 structures (Figure 3d). This element is proposed to stabilize a loop in polycystin-2 (Shen et al., 2016; Wilkes et al., 2017) and seems likely to play the same role in polycystin 2-l1.

Figure 3

Download asset Open asset

Three leaf clovers (TLCs) of the polycystin mucolipin domain (PMD)

Grieben et al., (Grieben et al., 2017) described the three-lobed area of polycystin-2’s PMD, naming it the ‘three-leafed clover’. The same area is found in polycystin 2-l1, but with several differences. Polycystin 2-l1 lacks the small α-helix in TLC1 of polycystin-2. Interestingly, this region (D208-D225) displays moderately lower sequence conservation among the TRPPs (Figure 3a). As noted, TLC1 in polycystin-2 appears to extend into the adjacent PMD and interact with the S3 helix and S3-S4 linker (Grieben et al., 2017; Wilkes et al., 2017). In polycystin 2-l1, the analogous TLC1 extends from one monomer into the PMD of the adjacent subunit (Figure 3b). However, the nature of the interaction differs in that F216 of TLC1 is near Y308 of the adjacent subunit’s PMD, representing a possible π - π stacking interaction or a hydrophobic pocket (Figure 4a). Additionally, N311’s carboxamide oxygen group, located on a hairpin between β3 and β4 of the PMD, is within hydrogen bonding distance (3.3 Å) of Y224’s amide in TLC1 of the adjacent subunit of polycystin 2-l1 (Figure 4b). Finally, residue W259 of TLC3 formats a cation-π stacking interaction with the upper pore domain at residue R534 (Figure 4c). We interpret these interactions as supporting tetrameric assembly and stability.

Figure 4

Download asset Open asset

Fenestrations

Despite moderately low identity and an increase in charged residues in polycystin 2-l1’s TLC3 (SPDKEE (residues 228–232) versus SVSSED in polycystin-2), the loop appears essentially the same as in polycystin-2. This element appears to reach under the PMD’s β-sheet. Proximity of the returning loop of TLC1 towards TLC2, as well as the S5-PH1 loop of the monomer and the PH2-S6 loop adjacent monomer creates four lateral openings at the base of the PMD in polycystin 2-l1, as in polycystin-2. These could present an alternative route for ion permeation (Grieben et al., 2017; Wilkes et al., 2017). This area is less conserved among the polycystin 2-l1 homologs (PKD2, PKD2L2)(Figure 3a).

Polycystin 2-l1’s PMD also interacts with the adjacent subunit’s upper pore domain. In the polycystin-2 multiple- (pdb 5MKF) and single-ion (pdb 5MKE) structures, Wilkes et al., (Wilkes et al., 2017) observed glycosylation-dependent interactions of the PMD of one subunit with the loop pore-helices of the adjacent (Wilkes et al., 2017), with mutually exclusive glycosylation states. One last interaction between TLC3 and the loop between pore-helix 2 and S6, noted in the polycystin-2 structure (Shen et al., 2016), is seen in polycystin 2-l1: W259, from TLC3, forms a cation-π stacking interaction with R534 and a hydrogen bond with G260’s carbonyl (2.8 Å) (Figure 4c). This difference may help rationalize the intermediate pore size observed in the polycystin-2 multiple-ion structure (1.4 Å) (Wilkes et al., 2017) and the single-ion structure (1.0 Å) (Wilkes et al., 2017) where such interactions do not exist, and our polycystin 2-l1 structure (2.6 Å).

The pore domain

As in polycystin-2, polycystin 2-l1’s selectivity filter is flanked by two pore-helices, which are themselves each flanked by a transmembrane spanning helix (S5 and S6). The S6 helix of polycystin 2-l1 remains α-helical throughout its length, whereas the S6 helix of polycystin-2 is broken by a middle π-helix element, before continuing to the C terminus (Shen et al., 2016) (Figure 1—figure supplement 2b and c). The S6 helix of polycystin 2-l1 is also tilted 7.9° relative to polycystin-2 (Figure 1—figure supplement 2c). The key filter residues of polycystin 2-l1 are L521, G522 and D523, in which the carbonyls of L521 and G522 point to the central pore axis (Figure 2a). D523’s side chain is well-resolved; the side chains face upward in parallel to the axis of ion conduction. The pore domain shows moderately strong conservation in S5, S6, and elements of the selectivity filter, as well as in pore helix 1 (Figure 2c). Notably, however, K511 in polycystin 2-l1 is highly variable among equivalent residues in other TRPPs (Figure 2c). The corresponding residue is a negatively-charged glutamate in polycystin-2 and an asparagine in polycystin 2-l2. The effect of this residue is to confer a net positive electrostatic potential (Figure 5a) relative to polycystin-2’s net negative charge.

Figure 5

Download asset Open asset

Relative to polycystin-2’s pore helix 1 (PDB 5T4D), the angle between the beginning of pore helix 1 and the top of the selectivity filter is similar (77° for polycystin 2-l1 and 75° for polycystin-2) compared to the 10° pitch difference between polycystin-2 and TRPV1 (Grieben et al., 2017). However, polycystin 2-l1’s pore helix, measured from beginning to end of the α-helix at the Cα, is ~1.5 Å shorter than that of polycystin-2.

The narrowest apertures in the selectivity filter, measured by HOLE (Smart et al., 1996), have radii of 2.6 and 2.8 Å at L521 and G522 respectively (Figure 2a). These distances represent a larger opening than that in any of the three polycystin-2 structures (1.0 to 1.4 Å) (Grieben et al., 2017; Shen et al., 2016; Wilkes et al., 2017). Such a diameter is sufficient to accommodate a partially hydrated Ca²⁺ ion with an average ~2.4 Å metal-oxygen distance (Katz et al., 1996; Marcus, 1988) and an ionic radius of 0.99 Å (Pauling, 1988; Hille, 2001). An observed density in the selectivity filter extends across the two carbonyls of L521 and G522, contrasting with polycystin-2 densities that localize above and below the selectivity filter (Wilkes et al., 2017) (Figure 2b). However, cryo-EM cannot determine the ion’s identity with the same degree of reliability of X-ray anomalous scattering. Although defined buffers are used, we cannot rule out the possibility of less abundant ions occupying the site.

Ion permeation differences between polycystin-2 and polycystin 2-l1

Polycystin-2 and polycystin 2-l1 core selectivity filters residues (LGD) are conserved, leading us to examine other explanations as to why polycystin 2-l1 conducts Ca²⁺ while polycystin-2 does not (P_Ca/P_Na ~6) (DeCaen et al., 2013, 2016; Kleene and Kleene, 2017; Liu et al., 2018). If the pore helix 1-selectivity filter-pore helix 2 region of polycystin 2-l1 (C512-P538) is substituted by polycystin-2’s analogous region (C632-P658), the polycystin 2-l1 chimera has roughly (within 4-fold) similar permeability to Na⁺, K⁺, and Ca²⁺ (Shen et al., 2016). Point mutation experiments in which polycystin 2-l1’s D523 and D525 were mutated to alanine or serine yielded no measurable currents (DeCaen et al., 2013). However, polycystin 2-l1(LGD₅₂₃: D523N) was 18-fold, and polycystin 2-l1(LGDFD₅₂₅: D525N) 9-fold less calcium-permeant (P_Ca²⁺/P_Cs⁺) than wt polycystin 2-l1. Also, and significantly, the D525N mutation reduced outward current block by Ca²⁺, while D523N did not (DeCaen et al., 2016). These experiments indicate that order and placement of negative charge in the narrow region of the pore is important for both Ca²⁺ permeation and block. Higher resolution cryo-EM structures that identify Ca²⁺ and/or monovalents that are occupying the pore with high certainty (or equivalent crystallographic studies like those of CaVAb [Tang et al., 2014]), and molecular dynamic simulations, should shed light on these issues.

As a final attempt to understand the differences in the relative selectivity of TRP channels, we compared the charge landscape of the pore helix-selectivity filter-pore helix assembly of polycystin-2 and polycystin 2-l1. When the PMD is stripped away and only the pore-helices and selectivity filter elements of both channels are visualized, a striking difference appears (Figure 5a). Polycystin 2-l1 has a more restricted negative charge density immediately surrounding the pore than does polycystin-2. Interestingly, its pore-helices form a ‘Maltese cross’ of net positive charge, rotated 45° to the negative charge distribution ‘cross’ in polycystin-2. We compare these charge distributions to the much more Ca²⁺-selective TRPV5 and TRPV6 (P_Ca²⁺/P_Na⁺> 100) (Ramsey et al., 2006) structures (Hughes et al., 2018; McGoldrick et al., 2018) (Figure 5a) in which the TRPV5/6 channels’ negative charge is much more localized around their respective pores.

An additional mechanism to explain different selectivities of TRPPs is a local bonding environment difference at and near the selectivity filter between polycystin-2 and polycystin 2-l1. A conserved salt bridge between D523 and R518 in polycystin 2-l1 exists in polycystin-2 (D643 and R638). Yet a hydrogen bond between residues D525 and A528 in polycystin 2-l1 is not present in polycystin-2; the residues are not conserved (Figure 5b and c). Local hydrogen bonding networks at the selectivity filter of ion channels have been observed to influence function such as the entry into a C-inactivated state for KcsA (Cordero-Morales et al., 2011), but such details for non-selective channels are currently unknown.

In both observations, we rely on one structure, which may change with gating. The use of molecular dynamics (MD) to observe the simulated trajectory of ions and their relationship to the energy landscapes of the selectivity filter is an excellent starting point for future experiments. Coupling future structures with functional mutagenic studies (e.g., the weakly conserved residue K511) should answer these questions.

Although polycystin 2-l1 and polycystin-2 structures are very similar, there are notable differences that may explain reported functional differences in their selectivity and permeation. For discussion, we divide polycystin 2-l1 into three regions to highlight these differences; the voltage sensing-like domain (VSLD), the pore mucolipin domain (PMD), and the pore (PD). In the VSLD, polycystin 2-l1’s S3 helix extends into the PMD as a complete α-helix. In addition, the top of the S3-S4 helical region abuts the PMD. Comparing the PMDs, polycystin 2-l1 lacks oligosaccharides that act as bridges between adjacent subunits. Finally, although similar, polycystin 2-l1’s pore diameter is slightly larger.

Due to our inchoate understanding of the TRPP family, we must be cautious in linking the observed differences in structure to functional differences. The two main features we seek to understand are selectivity and gating. Comparing selectivities of the two channels, we know that polycystin 2-l1 conducts both monovalent ions and Ca²⁺, while polycystin-2 is monovalent-selective (Kleene and Kleene, 2017; Liu et al., 2018). We explain this difference with two features: pore diameter and electrostatic fields. The relatively larger size of polycystin 2-l1’s pore radius (2.6–2.8 Å) may enable partially hydrated calcium’s transit, thus avoiding the larger energy required for pore residues to dehydrate calcium. Most striking are the unique electrostatic maps (Figure 5a) surrounding the polycystin-2 and polycystin 2-l1 pores. Such differences in charge suggest that the putative energy landscapes may create small variations in selectivity, a characteristic suggested for the prokaryotic NaK ion channel (Alam and Jiang, 2011). These slight charge differences, and so landscapes, may impart the subtle relative permeability differences observed among most TRP channels. For organelles in which the TRPP (cilia) and TRPML (endolysosomes) appear, these subtle differences may have driven the evolution of small changes in relative permeation in order to titrate levels of ions within these sub-femtoliter compartments. Research into the relative permeabilities of the related TRPML family of ion channels will further our understanding of mechanisms beyond direct side chain coordination.

TRPPs are not appreciably voltage- or mechanically-gated within physiologically relevant ranges of these forces (DeCaen et al., 2013, 2016), which raises the question of what other mechanisms activate these channels. The large surface area of the PMD is a logical candidate for binding of ligands. An appealing hypothesis is that unknown ligands bind the PMD, initiates an interaction with the VSLD (possibly at the extended S3 helix) and/or pore domain which alters channel function. Similarly, interactions of the PMD with the pore domain may elicit conformational changes leading to gating. Given the association of polycystin-1 (a protein of unknown function) and polycystin-2 proteins, it will be important to determine if any of the domains of polycystin-1 proteins interact with polycystin-2 family PMDs.

PI(4,5)P₂ was recently observed to facilitate the gating of both polycystin-2 and polycystin 2-l1 by interacting with residues on both the N and C termini (Zheng et al., 2018). The current lack of resolution of N- and C-terminal ion channel flexible regions limits our interpretation of gating influences for the group II TRP channels. Among group I TRP channels, the large interaction surfaces for intracellular ligands, such as the ankyrin repeats found on several TRPs, the more common TRP domains, and predicted C-terminal EF-hand calcium-binding sites and calmodulin-binding domains, are difficult to interpret without better understanding of proteins or ligands that may bind these regions.

The explosion of structures available for TRP channels lays the groundwork for interpreting future detailed characterization. The structure of polycystin 2-l1 presented here, will help in understanding what drove evolution to create the diversity that exists in the TRPP subfamily, but only when we more completely understand their biophysical and physiological functions. The recent progress made in cryo-EM underscores the importance of discovering these features.

The cryo map and model have been deposited to the Worldwide Protein Data Bank (6DU8) and Electron Microscopy Data Bank (8912).

1. 1. Clapham DE
   2. Hulse RE
   3. Li Z
   4. Huang RK
   5. Zhang J

   (2018) Human Polycystin 2-l1

   Publicly available at the RCSB Protein Data Bank (Accession no: 6DU8).

   https://www.rcsb.org/structure/6DU8
2. 1. Clapham DE
   2. Hulse RE
   3. Li Z
   4. Huang RK
   5. Zhang J

   (2018) Human Polycsytin 2-l1

   Publicly available at the Electron Microscopy Data Bank (Accession no: 8912).

   http://www.ebi.ac.uk/pdbe/entry/emdb/EMD-8912

1. 1. Shen PS
   2. Yang X
   3. DeCaen PG
   4. Liu X
   5. Bulkley D
   6. Clapham DE
   7. Cao E

   (2016) Cryo-EM structure of Polycystic Kidney Disease protein 2 (PKD2), residues 198-703

   Publicly available at the RCSB Protein Data Bank (accession no: 5T4D).

   https://www.rcsb.org/structure/5T4D
2. 1. Wilkes M
   2. Madej MG
   3. Ziegler C

   (2016) cryoEM Structure of Polycystin-2 in complex with calcium and lipids

   Publicly available at the RCSB Protein Data Bank (accession no: 5MKF).

   https://www.rcsb.org/structure/5MKF
3. 1. Wilkes M
   2. Madej MG
   3. Ziegler C

   (2016) cryoEM Structure of Polycystin-2 in complex with cations and lipids

   Publicly available at the RCSB Protein Data Bank (accession no: 5MKE).

   https://www.rcsb.org/structure/5MKE

1. 1. Adams PD
   2. Afonine PV
   3. Bunkóczi G
   4. Chen VB
   5. Davis IW
   6. Echols N
   7. Headd JJ
   8. Hung LW
   9. Kapral GJ
   10. Grosse-Kunstleve RW
   11. McCoy AJ
   12. Moriarty NW
   13. Oeffner R
   14. Read RJ
   15. Richardson DC
   16. Richardson JS
   17. Terwilliger TC
   18. Zwart PH

   (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution

   Acta Crystallographica Section D Biological Crystallography 66:213–221.

   https://doi.org/10.1107/S0907444909052925

   • PubMed
   • Google Scholar
2. 1. Alam A
   2. Jiang Y

   (2011) Structural studies of ion selectivity in tetrameric cation channels

   The Journal of General Physiology 137:397–403.

   https://doi.org/10.1085/jgp.201010546

   • PubMed
   • Google Scholar
3. 1. Arif Pavel M
   2. Lv C
   3. Ng C
   4. Yang L
   5. Kashyap P
   6. Lam C
   7. Valentino V
   8. Fung HY
   9. Campbell T
   10. Møller SG
   11. Zenisek D
   12. Holtzman NG
   13. Yu Y

   (2016) Function and regulation of TRPP2 ion channel revealed by a gain-of-function mutant

   PNAS 113:E2363–E2372.

   https://doi.org/10.1073/pnas.1517066113

   • PubMed
   • Google Scholar
4. 1. Barad BA
   2. Echols N
   3. Wang RY
   4. Cheng Y
   5. DiMaio F
   6. Adams PD
   7. Fraser JS

   (2015) EMRinger: side chain-directed model and map validation for 3D cryo-electron microscopy

   Nature Methods 12:943–946.

   https://doi.org/10.1038/nmeth.3541

   • PubMed
   • Google Scholar
5. 1. Cordero-Morales JF
   2. Jogini V
   3. Chakrapani S
   4. Perozo E

   (2011) A multipoint hydrogen-bond network underlying KcsA C-type inactivation

   Biophysical Journal 100:2387–2393.

   https://doi.org/10.1016/j.bpj.2011.01.073

   • PubMed
   • Google Scholar
6. 1. Cortes DM
   2. Perozo E

   (1997) Structural dynamics of the Streptomyces lividans K+ channel (SKC1): oligomeric stoichiometry and stability

   Biochemistry 36:10343–10352.

   https://doi.org/10.1021/bi971018y

   • PubMed
   • Google Scholar
7. 1. DeCaen PG
   2. Delling M
   3. Vien TN
   4. Clapham DE

   (2013) Direct recording and molecular identification of the calcium channel of primary cilia

   Nature 504:315–318.

   https://doi.org/10.1038/nature12832

   • PubMed
   • Google Scholar
8. 1. DeCaen PG
   2. Liu X
   3. Abiria S
   4. Clapham DE

   (2016) Atypical calcium regulation of the PKD2-L1 polycystin ion channel

   eLife 5:e13413.

   https://doi.org/10.7554/eLife.13413

   • PubMed
   • Google Scholar
9. 1. Delling M
   2. DeCaen PG
   3. Doerner JF
   4. Febvay S
   5. Clapham DE

   (2013) Primary cilia are specialized calcium signalling organelles

   Nature 504:311–314.

   https://doi.org/10.1038/nature12833

   • PubMed
   • Google Scholar
10. 1. Emsley P
    2. Lohkamp B
    3. Scott WG
    4. Cowtan K

    (2010) Features and development of coot

    Acta Crystallographica. Section D, Biological Crystallography 66:486–501.

    https://doi.org/10.1107/S0907444910007493

    • PubMed
    • Google Scholar
11. 1. Glaser F
    2. Pupko T
    3. Paz I
    4. Bell RE
    5. Bechor-Shental D
    6. Martz E
    7. Ben-Tal N

    (2003) ConSurf: identification of functional regions in proteins by surface-mapping of phylogenetic information

    Bioinformatics 19:163–164.

    https://doi.org/10.1093/bioinformatics/19.1.163

    • PubMed
    • Google Scholar
12. 1. Goehring A
    2. Lee CH
    3. Wang KH
    4. Michel JC
    5. Claxton DP
    6. Baconguis I
    7. Althoff T
    8. Fischer S
    9. Garcia KC
    10. Gouaux E

    (2014) Screening and large-scale expression of membrane proteins in mammalian cells for structural studies

    Nature Protocols 9:2574–2585.

    https://doi.org/10.1038/nprot.2014.173

    • PubMed
    • Google Scholar
13. 1. Grant T
    2. Grigorieff N

    (2015) Measuring the optimal exposure for single particle cryo-EM using a 2.6 Å reconstruction of rotavirus VP6

    eLife 4:e06980.

    https://doi.org/10.7554/eLife.06980

    • PubMed
    • Google Scholar
14. 1. Grant T
    2. Rohou A
    3. Grigorieff N

    (2018) cisTEM, user-friendly software for single-particle image processing

    eLife 7:e35383.

    https://doi.org/10.7554/eLife.35383

    • PubMed
    • Google Scholar
15. 1. Grieben M
    2. Pike AC
    3. Shintre CA
    4. Venturi E
    5. El-Ajouz S
    6. Tessitore A
    7. Shrestha L
    8. Mukhopadhyay S
    9. Mahajan P
    10. Chalk R
    11. Burgess-Brown NA
    12. Sitsapesan R
    13. Huiskonen JT
    14. Carpenter EP

    (2017) Structure of the polycystic kidney disease TRP channel Polycystin-2 (PC2)

    Nature Structural & Molecular Biology 24:114–122.

    https://doi.org/10.1038/nsmb.3343

    • PubMed
    • Google Scholar
16. Book

    1. Hille B

    (2001)

    Ion Channels of Excitable Membranes

    Sunderland, MA: Sinauer.

    • Google Scholar
17. 1. Hughes TET
    2. Lodowski DT
    3. Huynh KW
    4. Yazici A
    5. Del Rosario J
    6. Kapoor A
    7. Basak S
    8. Samanta A
    9. Han X
    10. Chakrapani S
    11. Zhou ZH
    12. Filizola M
    13. Rohacs T
    14. Han S
    15. Moiseenkova-Bell VY

    (2018) Structural basis of TRPV5 channel inhibition by econazole revealed by cryo-EM

    Nature Structural & Molecular Biology 25:53–60.

    https://doi.org/10.1038/s41594-017-0009-1

    • PubMed
    • Google Scholar
18. 1. Katz AK
    2. Glusker JP
    3. Beebe SA
    4. Bock CW

    (1996) Calcium ion coordination: a comparison with that of Beryllium, Magnesium, and Zinc

    Journal of the American Chemical Society 118:5752–5763.

    https://doi.org/10.1021/ja953943i

    • Google Scholar
19. 1. Kleene SJ
    2. Kleene NK

    (2017) The native TRPP2-dependent channel of murine renal primary cilia

    American Journal of Physiology-Renal Physiology 312:F96–F108.

    https://doi.org/10.1152/ajprenal.00272.2016

    • PubMed
    • Google Scholar
20. 1. Kucukelbir A
    2. Sigworth FJ
    3. Tagare HD

    (2014) Quantifying the local resolution of cryo-EM density maps

    Nature Methods 11:63–65.

    https://doi.org/10.1038/nmeth.2727

    • PubMed
    • Google Scholar
21. 1. Landau M
    2. Mayrose I
    3. Rosenberg Y
    4. Glaser F
    5. Martz E
    6. Pupko T
    7. Ben-Tal N

    (2005) ConSurf 2005: the projection of evolutionary conservation scores of residues on protein structures

    Nucleic Acids Research 33:W299–W302.

    https://doi.org/10.1093/nar/gki370

    • PubMed
    • Google Scholar
22. 1. Liu X
    2. Vien T
    3. Duan J
    4. Sheu SH
    5. DeCaen PG
    6. Clapham DE

    (2018) Polycystin-2 is an essential ion channel subunit in the primary cilium of the renal collecting duct epithelium

    eLife 7:e33183.

    https://doi.org/10.7554/eLife.33183

    • PubMed
    • Google Scholar
23. 1. Marcus Y

    (1988) Ionic radii in aqueous solutions

    Chemical Reviews 88:1475–1498.

    https://doi.org/10.1021/cr00090a003

    • Google Scholar
24. 1. Mastronarde DN

    (2005) Automated electron microscope tomography using robust prediction of specimen movements

    Journal of Structural Biology 152:36–51.

    https://doi.org/10.1016/j.jsb.2005.07.007

    • PubMed
    • Google Scholar
25. 1. McGoldrick LL
    2. Singh AK
    3. Saotome K
    4. Yelshanskaya MV
    5. Twomey EC
    6. Grassucci RA
    7. Sobolevsky AI

    (2018) Opening of the human epithelial calcium channel TRPV6

    Nature 553:233–237.

    https://doi.org/10.1038/nature25182

    • PubMed
    • Google Scholar
26. Book

    1. Pauling L

    (1988)

    General Chemistry

    New York: Dover Publications Inc.

    • Google Scholar
27. 1. Pettersen EF
    2. Goddard TD
    3. Huang CC
    4. Couch GS
    5. Greenblatt DM
    6. Meng EC
    7. Ferrin TE

    (2004) UCSF chimera--a visualization system for exploratory research and analysis

    Journal of Computational Chemistry 25:1605–1612.

    https://doi.org/10.1002/jcc.20084

    • PubMed
    • Google Scholar
28. 1. Ramsey IS
    2. Delling M
    3. Clapham DE

    (2006) An introduction to TRP channels

    Annual Review of Physiology 68:619–647.

    https://doi.org/10.1146/annurev.physiol.68.040204.100431

    • PubMed
    • Google Scholar
29. 1. Rosenthal PB
    2. Henderson R

    (2003) Optimal determination of particle orientation, absolute hand, and contrast loss in single-particle electron cryomicroscopy

    Journal of Molecular Biology 333:721–745.

    https://doi.org/10.1016/j.jmb.2003.07.013

    • PubMed
    • Google Scholar
30. 1. Shen PS
    2. Yang X
    3. DeCaen PG
    4. Liu X
    5. Bulkley D
    6. Clapham DE
    7. Cao E

    (2016) The structure of the polycystic kidney disease channel PKD2 in lipid nanodiscs

    Cell 167:763–773.

    https://doi.org/10.1016/j.cell.2016.09.048

    • PubMed
    • Google Scholar
31. 1. Smart OS
    2. Neduvelil JG
    3. Wang X
    4. Wallace BA
    5. Sansom MS

    (1996) HOLE: a program for the analysis of the pore dimensions of ion channel structural models

    Journal of Molecular Graphics 14:354–360.

    https://doi.org/10.1016/S0263-7855(97)00009-X

    • PubMed
    • Google Scholar
32. 1. Tan YZ
    2. Baldwin PR
    3. Davis JH
    4. Williamson JR
    5. Potter CS
    6. Carragher B
    7. Lyumkis D

    (2017) Addressing preferred specimen orientation in single-particle cryo-EM through tilting

    Nature Methods 14:793–796.

    https://doi.org/10.1038/nmeth.4347

    • PubMed
    • Google Scholar
33. 1. Tang L
    2. Gamal El-Din TM
    3. Payandeh J
    4. Martinez GQ
    5. Heard TM
    6. Scheuer T
    7. Zheng N
    8. Catterall WA

    (2014) Structural basis for Ca2+ selectivity of a voltage-gated calcium channel

    Nature 505:56–61.

    https://doi.org/10.1038/nature12775

    • PubMed
    • Google Scholar
34. 1. Wilkes M
    2. Madej MG
    3. Kreuter L
    4. Rhinow D
    5. Heinz V
    6. De Sanctis S
    7. Ruppel S
    8. Richter RM
    9. Joos F
    10. Grieben M
    11. Pike AC
    12. Huiskonen JT
    13. Carpenter EP
    14. Kühlbrandt W
    15. Witzgall R
    16. Ziegler C

    (2017) Molecular insights into lipid-assisted Ca²⁺ regulation of the TRP channel Polycystin-2

    Nature Structural & Molecular Biology 24:123–130.

    https://doi.org/10.1038/nsmb.3357

    • PubMed
    • Google Scholar
35. 1. Williams CJ
    2. Headd JJ
    3. Moriarty NW
    4. Prisant MG
    5. Videau LL
    6. Deis LN
    7. Verma V
    8. Keedy DA
    9. Hintze BJ
    10. Chen VB
    11. Jain S
    12. Lewis SM
    13. Arendall WB
    14. Snoeyink J
    15. Adams PD
    16. Lovell SC
    17. Richardson JS
    18. Richardson DC

    (2018) MolProbity: more and better reference data for improved all-atom structure validation

    Protein Science 27:293–315.

    https://doi.org/10.1002/pro.3330

    • PubMed
    • Google Scholar
36. 1. Wu G
    2. D'Agati V
    3. Cai Y
    4. Markowitz G
    5. Park JH
    6. Reynolds DM
    7. Maeda Y
    8. Le TC
    9. Hou H
    10. Kucherlapati R
    11. Edelmann W
    12. Somlo S

    (1998) Somatic inactivation of Pkd2 results in polycystic kidney disease

    Cell 93:177–188.

    https://doi.org/10.1016/S0092-8674(00)81570-6

    • PubMed
    • Google Scholar
37. 1. Zheng W
    2. Cai R
    3. Hofmann L
    4. Nesin V
    5. Hu Q
    6. Long W
    7. Fatehi M
    8. Liu X
    9. Hussein S
    10. Kong T
    11. Li J
    12. Light PE
    13. Tang J
    14. Flockerzi V
    15. Tsiokas L
    16. Chen XZ

    (2018) Direct binding between Pre-S1 and TRP-like domains in TRPP channels mediates gating and functional regulation by PIP2

    Cell Reports 22:1560–1573.

    https://doi.org/10.1016/j.celrep.2018.01.042

    • PubMed
    • Google Scholar

Thank you for submitting your article "Cryo-EM structure of the polycystin 2-l1 ion channel" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by Richard Aldrich as the Senior Editor, a Reviewing Editor, and three reviewers. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

In this study, Hulse et al., presented a structure of the full-length human polycystin 2-l1 protein at 3.1Å resolution using single particle electron cryo-microscopy (cryo-EM). Although it shares high sequence similarity to polycystin 2, whose structure has been determined by multiple labs recently, polycystin 2-l1 exhibits distinct ion permeability. The study aims to provide a detailed view of the structural differences between polycystin 2-l1 and polycystin 2, and these structural differences may underlie the functional difference between the two channels.

Essential revisions:

1) The manuscript is structured from the viewpoint of understanding the function of the channel but, at present, it is not clear at this point whether the structural differences contribute to function. The main impact of the present study would be to compare the structures of polycystin 2-l1 and PKD2 in as much detail as possible. Oddly, the introduction to the manuscript barely mentions the fact that a structure for PKD2 has already been reported by this and other groups, but the trajectory of the manuscript would be better served by addressing this up front and moving on from there. In this context, figures showing direct comparisons or overlays of these two TRPP structures would be very useful and help convey important points that are difficult to ascertain from the current format. We are not expecting new functional data but the new structures at the least should help develop a cogent hypothesis for further testing.

2) There seems to be a disconnect between the stated resolution of the structure and the quality of the maps. This is suggested by the inability to resolve many regions and, in particular, side chains. From the figures provided (for example, Figure 2B and Figure 3D), it is not clear how well the model fits the density map in various regions. The maps look more to be in the 4 – 4.5Å range, and it isn't clear where the possible discrepancy lies. Overfitting is a known issue with cisTEM program for EM structure determination. The authors should re-evaluate their data processing and analysis.

3) Related to the above point, please provide the map for this structure which would help reviewers assess the quality of the data and thus the validity of the conclusions.

4) Another question concerns the unusual protocol for channel protein purification, which is very different from that described for PKD2 and which involves initial extraction into C12E9 detergent, batch overnight incubation with amylose resin, and multiple gel filtration steps. Could this be enhancing protein instability that degrades resolution?

5) Overall, the manuscript is difficult to follow. Most figures are not properly labeled and many of the descriptions in the paper are not adequately supported by figures, making it difficult to comprehend, particularly for readers who are not familiar with the TRPP family of channels. Specifically:

Results section: "We conclude that these elements are either unstructured or connected with flexible regions, which, along with the lack of cytoplasmic elements in 2D classification (Shen et al., 2016), prevents model building of these regions." This sentence is unclear and I am not sure what the phrase "cytoplasmic elements" refer to.

Subsection “The Voltage Sensing-Like Domain”: "This S3 helix exhibits greater secondary structure than the presumed closed polycystin-2 structure of Shen et al." What does greater secondary structure mean here – longer S3?

Subsection "The Polycystin Mucolipin Domain". The authors spend so much effort pointing out very detailed structural differences between polycystin 2-l1 and polycystin 2. However, these differences do not seem to have any functional relevance.

Subsection “Three leaf clovers of the polycystin mucolipin domain”: "However, the nature of the interaction differs in that F216 of TLC1 is near W308 of the adjacent subunit's PMD, representing a possible pi-pi stacking interaction or a hydrophobic pocket". A hydrophobic pocket is not a proper way to describe this interaction.

Subsection “Fenestrations and TLC3”: If I recall correctly, the fenestration appears to be sealed in polycystin-2 by an extended loop based on Shen's structure. If there is a fenestration in the present structure, a proper figure should be provided here.

Subsection “Fenestrations and TLC3”: What does "upper pore domain" refer to: the extracellular side of the pore domain? What do "the loop pore helices" refer to here: the loop and/or the pore helices? What does "mutually exclusive glycosylation states" refer to? In the last sentence of this section: "This difference may help rationalize the intermediate diameter observed in the polycystin-2 multiple-ion structure (1.4Å) (Wilkes et al., 2017) and the single-ion structure (1.0Å) (Wilkes et al., 2017) where such interactions do not exist, and our polycystin 2-l1 structure (3.1Å)." What diameters are the authors describing here and what does "this difference" refer to? Again, there is no figure to support all the descriptions in this paragraph.

Subsection “The Pore Domain”:

I am not sure how the figures are ordered. Figure 2 is described after Figure 4. Also, there are no proper labels on Figure 2.

First paragraph of subsection “The Pore Domain”: "Notably, one element, K511 in polycystin 2-l1, is highly variable among TRPPs." Where is this K511? Again, there is no figure to support it.

Second paragraph subsection “The Pore Domain”: I simply don't understand what is discussed here and again there is no figure to support the description. Should the filter diameter be radius? Does 5K47 refer to a PDB code?

Third paragraph of subsection “The Pore Domain”: The narrowest distance of 2.1 Å appears to be the narrowest filter radius based on Figure 2A (right panel), which excludes the van deer Waals radius of the G522 carbonyl oxygen. The "2.4 Å metal-oxygen distance" refers to atom-to-atom distance between Ca²⁺ and oxygen. These two distances are not comparable.

Third paragraph of subsection “The Pore Domain”: Here the author repeated the same discussion as the last paragraph, except using atom-to-atom diagonal distances. If the diagonal distance between the two G522 carbonyl oxygen atoms is 8.7Å, then the minimum distance between the G522 carbonyl oxygen and Ca²⁺ ion would be 4.4 Å. If so, the argument of 2.4 Å metal-oxygen distance is no longer correct. Basically, the whole description about the pore domain is messy and incomprehensible.

Subsection “Ion permeation differences between polycystin-2 and polycystin 2-l1”. The values of relative permeability ratio should be provided.

"Higher resolution Cryo-EM structures that identify Ca²⁺ and/or monovalents that are occupying the pore with high certainty (or equivalent crystallographic studies like those of CaVAb (Tang et al., 2014)), and realistic molecular dynamic simulations, should shed light on these issues." What is realistic MD simulation? What do "these issues" refer to here?

I simply don't understand the logic behind the electrostatic argument. Are the authors suggesting that the net positive charge of the polycystin 2-l1 pore helices contributes to the Ca²⁺ permeability of the channel? Also, how do they compare the relative Ca²⁺ permeability among four different channels?

Discussion section "The relatively larger diameter of polycystin 2-l1's pore (0.6Å – 1.1Å) may enable partially […]" Where is this distance value from? It does not match with the pore radius plot! Is it 0.6-1.1 Å larger than the pore of polycystin 2?

Figure 1A the pore radius and the solvent accessible pathway does not match, particularly the lower gate (green) part.

Essential revisions:

1) The manuscript is structured from the viewpoint of understanding the function of the channel but, at present, it is not clear at this point whether the structural differences contribute to function. The main impact of the present study would be to compare the structures of polycystin 2-l1 and PKD2 in as much detail as possible. Oddly, the introduction to the manuscript barely mentions the fact that a structure for PKD2 has already been reported by this and other groups, but the trajectory of the manuscript would be better served by addressing this up front and moving on from there. In this context, figures showing direct comparisons or overlays of these two TRPP structures would be very useful and help convey important points that are difficult to ascertain from the current format. We are not expecting new functional data but the new structures at the least should help develop a cogent hypothesis for further testing.

We have taken your point and extensively discuss the PKD2/PKD-2l1 structures. We believe that the most relevant comparisons are made in Figure 5 with PKD2 and TRPV5 and TRPV6. We emphasize that despite PDK2’s similarity with PKD2L1 (structural and sequence), there are differences in Ca²⁺ permeation physiologically: polycystin-2l1 is permeant to calcium, but polycystin-2 is not. We have included a comparison of the profile of the selectivity filter and mapped the RMSD difference onto the structure as well as provided an overlay (Figure 1—figure supplement 5).

2) There seems to be a disconnect between the stated resolution of the structure and the quality of the maps. This is suggested by the inability to resolve many regions and, in particular, side chains. From the figures provided (for example, Figure 2B and Figure 3D), it is not clear how well the model fits the density map in various regions. The maps look more to be in the 4 – 4.5Å range, and it isn't clear where the possible discrepancy lies. Overfitting is a known issue with cisTEM program for EM structure determination. The authors should re-evaluate their data processing and analysis.

We agree. Further evaluation of the data after consultation with Niko Grigorieff’s group revealed that the data was slightly overfit. To address this concern, we reviewed our data processing strategy and now have limited the resolution during refinement to 8Å. This decreased the nominal resolution but improved map density, with associated improved 3D FSCs (see below). We’ve re-evaluated the data with new tools from the Lyumkis group using 3D FSC plots to characterize the preferred orientation observed in our data Tan et al., 2017. The 3D FSC results are consistent with the angular distribution plot (Figure 1—figure supplement 4) which show that while the Z-resolution is high (labeled side view), the in-plane (x/y) resolution is weaker. This is also reflected in the fits of side chains in the map such as elements seen in Figure 1—figure supplement 5B. We have included the results of this analysis as well as new maps and models. We hope that our analysis and the new tools will help other ion channel groups facing similar problems with their structural efforts using cryoEM. While we cannot eliminate the preferred orientation of our channel, we believe that Lyumkis’ methods enable a better understanding of the effect on the data. Additional metrics, such as the fit of the map (Figure 1—figure supplement 5), the high degree of similarity to Erhu Cao’s polycystin-2 model (Figure 1—figure supplement 2), and the similarity to the mouse polycystin 2-l1 model that was released very shortly after we submitted this work (Su et al., 2018) all verify that our model is not significantly affected. The current estimation from cisTEM and the 3D FSC is 3.3Å and 3.2Å respectively. We report our resolution at the more conservative 3.3Å. The model was reevaluated with the new maps, and several minor adjustments were made including the ability to build in a side chain for residue D523, a critical component of the selectivity filter. We also reveal a more complete loop in the PMD, and some improved density fitting at linkers between transmembrane helices. All figures and the manuscript were updated with this information.

3) Related to the above point, please provide the map for this structure which would help reviewers assess the quality of the data and thus the validity of the conclusions.

Done.

4) Another question concerns the unusual protocol for channel protein purification, which is very different from that described for PKD2 and which involves initial extraction into C12E9 detergent, batch overnight incubation with amylose resin, and multiple gel filtration steps. Could this be enhancing protein instability that degrades resolution?

Our protocol is based on empirical evaluation of extraction, stabilization and exchange using standard techniques. C12E9 was selected as an efficient extraction detergent on an empirical screen using western blot against MBP. Stability is evident from the exchange into DDM:CHS during amylose chromatography and in the SEC (Figure 1—figure supplement 1) and the ability of the protein to remain in solution for up to 72 h with stable SEC profiles (not shown).

5) Overall, the manuscript is difficult to follow. Most figures are not properly labeled and many of the descriptions in the paper are not adequately supported by figures, making it difficult to comprehend, particularly for readers who are not familiar with the TRPP family of channels. Specifically:

Results section: "We conclude that these elements are either unstructured or connected with flexible regions, which, along with the lack of cytoplasmic elements in 2D classification (Shen et al., 2016), prevents model building of these regions." This sentence is unclear, and I am not sure what the phrase "cytoplasmic elements" refer to.

Polycystin 2-l1’s C terminus is cytoplasmic, likely forming a coiled-coiled motif like other homotetrameric channels. In Shen, 2016, a truncation of the C terminus was required for a stable biochemical preparation. We were able to express full-length protein, but the C terminus was still unresolvable

Subsection “The Voltage Sensing-Like Domain”: "This S3 helix exhibits greater secondary structure than the presumed closed polycystin-2 structure of Shen et al." What does greater secondary structure mean here – longer S3?

Yes, S3 in polycystin 2-l1 has more α helical content and is thus continuously longer than the interrupted C terminus of Shen et al., 2016. We have clarified the text.

Subsection "The Polycystin Mucolipin Domain". The authors spend so much effort pointing out very detailed structural differences between polycystin 2-l1 and polycystin 2. However, these differences do not seem to have any functional relevance.

We believe that the PMD will turn out to be a ligand binding domain, but to date, none has been found (for the PMD of any type II TRP). Since disruption of the PMD results in nonfunctional channels, it will take some time to understand its function.

Subsection “Three leaf clovers of the polycystin mucolipin domain”: "However, the nature of the interaction differs in that F216 of TLC1 is near W308 of the adjacent subunit's PMD, representing a possible pi-pi stacking interaction or a hydrophobic pocket". A hydrophobic pocket is not a proper way to describe this interaction.

We have corrected the manuscript.

Subsection “Fenestrations and TLC3”: If I recall correctly, the fenestration appears to be sealed in polycystin-2 by an extended loop based on Shen's structure. If there is a fenestration in the present structure, a proper figure should be provided here.

The fenestration seen in polycystin 2-l1 is a small hole at the interface of the top of the VSLD and bottom of the PMD. Polycystin-2 has a similar configuration, with a small hole (5T4D). Whether the PMD is truly sealing this entry is unclear in our model since we can only approximate the position relative to the membrane.

Subsection “Fenestrations and TLC3”: What does "upper pore domain" refer to: the extracellular side of the pore domain? What do "the loop pore helices" refer to here: the loop and/or the pore helices? What does "mutually exclusive glycosylation states" refer to? In the last sentence of this section: "This difference may help rationalize the intermediate diameter observed in the polycystin-2 multiple-ion structure (1.4Å) (Wilkes et al., 2017) and the single-ion structure (1.0Å) (Wilkes et al., 2017) where such interactions do not exist, and our polycystin 2-l1 structure (3.1Å)." What diameters are the authors describing here and what does "this difference" refer to? Again, there is no figure to support all the descriptions in this paragraph.

The comparison of polycystin-2 and polycystin 2-l1 in the supplementary section helps to clarify this discussion.

Subsection “The Pore Domain”:

I am not sure how the figures are ordered. Figure 2 is described after Figure 4. Also, there are no proper labels on Figure 2.

First paragraph of subsection “The Pore Domain”: "Notably, one element, K511 in polycystin 2-l1, is highly variable among TRPPs." Where is this K511? Again, there is no figure to support it.

Second paragraph of subsection “The Pore Domain”: I simply don't understand what is discussed here and again there is no figure to support the description. Should the filter diameter be radius? Does 5K47 refer to a PDB code?

Third paragraph of subsection “The Pore Domain”: The narrowest distance of 2.1 Å appears to be the narrowest filter radius based on Figure 2A (right panel), which excludes the van deer Waals radius of the G522 carbonyl oxygen. The "2.4 Å metal-oxygen distance" refers to atom-to-atom distance between Ca²⁺ and oxygen. These two distances are not comparable.

We apologize for the confusion. Our attempt to discuss both the radii of the measurements from the HOLE program (which calculates the VDWr for each atom, with a spherical probe for our case) and measurements from Pymol (as either Cα or CO distances) present different information. HOLE is better suited to this discussion. We’ve removed the Pymol distances. We’ve also corrected the specific questions, improved the figure legends and ensured the order is correct.

Third paragraph of subsection “The Pore Domain”: Here the author repeated the same discussion as the last paragraph, except using atom-to-atom diagonal distances. If the diagonal distance between the two G522 carbonyl oxygen atoms is 8.7Å, then the minimum distance between the G522 carbonyl oxygen and Ca²⁺ ion would be 4.4Å. If so, the argument of 2.4Å metal-oxygen distance is no longer correct. Basically, the whole description about the pore domain is messy and incomprehensible.

Subsection “Ion permeation differences between polycystin-2 and polycystin 2-l1”. The values of relative permeability ratio should be provided.

"Higher resolution Cryo-EM structures that identify Ca²⁺ and/or monovalents that are occupying the pore with high certainty (or equivalent crystallographic studies like those of CaVAb (Tang et al., 2014)), and realistic molecular dynamic simulations, should shed light on these issues." What is realistic MD simulation? What do "these issues" refer to here?

I simply don't understand the logic behind the electrostatic argument. Are the authors suggesting that the net positive charge of the polycystin 2-l1 pore helices contributes to the Ca²⁺ permeability of the channel? Also, how do they compare the relative Ca²⁺ permeability among four different channels?

We have added the relative permeabilities from previous lab work (DeCaen, 2013) and cited literature (Ramsey, 2006).

Yes, we believe that the overall difference in charge at this region contributes to the moderate increase in Ca²⁺ relative to polycystin-2.

Polycystin-2 is nonselective among monovalents and does not conduct calcium, whereas polycystin 2-l1 is nonselective among monovalens but also conducts Ca²⁺ (with weak selectivity over Na⁺). TRPV5 and TRPV6 have a much higher selectivity (>10x) for Ca²⁺ over Na⁺ than polycystin 2-l1. We believe that both the local energy landscape and geometry of the selectivity filter are at least two variables that distinguish polycystin 2-l1 from other TRPP and TRPV channels.

Discussion section "The relatively larger diameter of polycystin 2-l1's pore (0.6Å – 1.1Å) may enable partially […]" Where is this distance value from? It does not match with the pore radius plot! Is it 0.6-1.1 Å larger than the pore of polycystin 2?

This was an error and has been fixed.

Figure 1A the pore radius and the solvent accessible pathway does not match, particularly the lower gate (green) part.

This figure was updated with the new model.

Author details

1. Raymond E Hulse

   1. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United States
   2. Department of Cardiology, Howard Hughes Medical Institute, Boston Children’s Hospital, Boston, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0110-3752

2. Zongli Li

   Department of Biological Chemistry and Molecular Pharmacology, Howard Hughes Medical Institute, Harvard Medical School, Boston, United States

   Contribution

   Resources, Investigation, Writing—review and editing

   Competing interests

   No competing interests declared

3. Rick K Huang

   Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United States

   Contribution

   Resources, Investigation, Writing—review and editing

   Competing interests

   No competing interests declared

4. Jin Zhang

   1. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United States
   2. School of Basic Medical Sciences, Nanchang University, Nanchang, China

   Contribution

   Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

5. David E Clapham

   1. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United States
   2. Department of Cardiology, Howard Hughes Medical Institute, Boston Children’s Hospital, Boston, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   claphamd@hhmi.org

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4459-9428

• 2,823

  Page views

• 594

  Downloads

• 34

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.