CIM6P/IGF2R, a type one integral membrane glycoprotein with a short cytoplasmic tail and 15 extracytoplasmic repeats, binds to a variety of ligands, among which the better characterized belong to two groups that bind to distinct sites: glycoproteins conjugated with mannose-6-phosphate (M6P), including lysosomal enzymes, and IGF2 (Ghosh et al., 2003; Morgan et al., 1987; Wang et al., 2017). CIM6P/IGF2R is implicated in numerous cellular processes including growth, trafficking of lysosomal enzymes, lysosome-mediated clearance of the polypeptide IGF2, and proteolytic activation of enzymes and growth factor precursors (El-Shewy and Luttrell, 2009; Ghosh et al., 2003; Hawkes and Kar, 2004; Wang et al., 2017). CIM6P/IGF2R traffics newly synthesized lysosomal enzymes from the trans-Golgi network to lysosomes via early and late endosomes, where the enzymes are released in the acidic environment; late endosomes then fuse with lysosomes, while the CIM6P/IGF2R recycles back to the Golgi or to the cell surface via endosomal transport. CIM6P/IGF2R expressed on the cell membrane is best known for sequestering and internalizing IGF2, leading to its degradation in lysosomes (Oka et al., 1985); in this case as well, the receptor recycles back to its intracellular or plasma membrane compartments. The same trafficking pathway is also used by a subset of lysosomal enzymes that are secreted into the extracellular space: the enzymes bind to cell surface CIM6P/IGF2R, and are then targeted to lysosomes; see Ghosh et al. (2003), for a thorough review of the binding partners and sorting signals for CIM6P/IGF2R trafficking. CIM6P/IGF2R-null mice exhibit perinatal lethality, significant overgrowth abnormalities in organ development, particularly in cardiac tissue, and mis-sorting of M6P-tagged lysosomal enzymes. Collectively, these phenotypes indicate that CIM6P/IGF2R is crucial for normal growth and development, as well as for the correct trafficking of lysosomal enzymes (Lau et al., 1994; Ludwig et al., 1996); Z.-Q. Wang et al. (1994); Wylie et al., 2003).

Outside of the developmental context, CIM6P/IGF2R plays important roles in growth inhibition, including tumor suppressor functions. In many types of human cancers, including tumors of the lung, liver, and breast, CIM6P/IGF2R frequently undergoes loss of heterozygosity, often accompanied by loss-of-function mutation in the remaining allele (De Souza et al., 1995; Hankins et al., 1996; Kong et al., 2000; Ouyang et al., 1997). Loss of CIM6P/IGF2R function is thought to allow abnormally high levels of circulating IGF2 to exert mitogenic effects, presumably acting via its low-affinity receptor IGF1R (Ludwig et al., 1996), resulting in tumor growth.

Notably, CIM6P/IGF2R is widely expressed throughout the developed brain, including the choroid plexus, cortex, hippocampus, thalamus, hypothalamus, cerebellum, and spinal cord (Kar et al., 2006), with higher expression in cortex, hippocampus, cerebellum, and certain brainstem nuclei (Hawkes and Kar, 2004). In addition, specific variants of the gene encoding CIM6P/IGF2R have been associated with high cognitive ability in a quantitative trait locus study (Chorney et al., 1998). CIM6P/IGF2R has also been implicated in neurodegenerative diseases, although the mechanisms underlying this effect remain to be established (Wang et al., 2017). In spite of this knowledge, relatively little is known about the function of the CIM6P/IGF2R in the mature central nervous system and its specific role in cognition.

Previous studies from our lab showed that IGF2 expressed in the hippocampus is required for long-term memory formation, and that administration of recombinant IGF2 either directly into the hippocampus or via systemic injection significantly enhances memories and prolongs their persistence (Agis-Balboa et al., 2011; Chen et al., 2011; Stern et al., 2014). Furthermore, IGF2 prevents aging-related memory loss in rats (Steinmetz et al., 2016), and reverses multiple deficits, including cognitive ones, in mouse models of Alzheimer’s disease (Mellott et al., 2014; Pascual‐Lucas et al., 2014). In addition, subcutaneous injections of IGF2 reversed most core deficits in a mouse model of autism spectrum disorders (Steinmetz et al., 2018). These studies also demonstrated that the effects of IGF2 in memory enhancement and recovery of function in disease models are dependent on CIM6P/IGF2R (Chen et al., 2011; Steinmetz et al., 2018; Stern et al., 2014).

Given these effects, in this study, we investigated the role of CIM6P/IGF2R in memory formation and elucidated the underlying mechanisms in adult rats and mice.

Our results demonstrate that CIM6P/IGF2R expressed by hippocampal neurons is required for the consolidation, but not learning, retrieval or reconsolidation of hippocampus-dependent memories in rodents. We also found that learning rapidly engages CIM6P/IGF2R to increase protein levels in response to learning, including the IEGs Arc, Egr1, and c-Fos. However, the receptor is dispensable for the learning-dependent increase in the mRNA levels of these IEGs. Finally, administration of the receptor ligand M6P, like IGF2, significantly enhances long-term memory, indicating that CIM6P/IGF2R represents a target mechanism for memory enhancement.

These data identified novel functional roles of the CIM6P/IGF2R, which are added to its previously known functions on the regulation of organ growth and development, tumor suppression, and its possible involvement in Alzheimer’s disease (El-Shewy and Luttrell, 2009; Wang et al., 2017). Because CIM6P/IGF2R is abundantly expressed in the mature brain, particularly in neurons, it is important to understand its contributions to brain functions under normal conditions and in disease states.

Our acute inhibition of CIM6P/IGF2R at various timepoints before and after training revealed that the receptor in the dorsal hippocampus is not required for encoding, retrieval, or reconsolidation but is rapidly engaged by learning to allow the regulation of immediate changes required for memory consolidation. In fact, blockade of the receptor pre-training completely abolished memory and led to memory impairment very rapidly: at 1 hr after IA learning in rats, and 5 min after contextual fear conditioning in mice. These results indicated that the CIM6P/IGF2R is promptly recruited by training and regulates rapid mechanisms critical for memory consolidation. However, once engaged by learning, the function of CIM6P/IGF2R in memory consolidation continued for several hours: only a prolonged post-training blockade of the receptor by two injections, one immediately after training and another 8 hr later, produced memory impairment. The reason for this prolonged functional requirement of CIM6P/IGF2R following training remains to be fully understood, and future studies are needed to address this issue.

Rapid changes induced by learning that then persists for several hours or days as critical mechanisms of memory consolidation include de novo transcription and translation, hallmark molecular signatures of long-term memory (Alberini and Kandel, 2015; Costa-Mattioli et al., 2009; Santini et al., 2014). Our data showed that CIM6P/IGF2R is necessary for the induction of de novo protein synthesis in response to learning, including the increase of the IEGs Arc, Egr1, and c-Fos proteins. In fact, receptor blockade abolished their induction specifically at the protein level, without affecting induction of the corresponding mRNAs. Thus, CIM6P/IGF2R is a critical upstream regulator of de novo protein synthesis increase evoked by learning.

These results provided key information on mechanisms downstream of the CIM6P/IGF2R; however, the specific mechanistic steps that link the receptor to de novo translation, and particularly translation of IEGs, remain to be understood. Because CIM6P/IGF2R has a small intracellular domain and does not have known direct signal transduction mechanisms (El-Shewy and Luttrell, 2009), there are no predictable hypotheses about which molecular mechanisms directly activate its downstream steps. We speculate that these mechanisms reside in domains that link of endosomal trafficking, lysosomal degradation and Golgi functions to learning-induced de novo translation, a complex area of investigation that shall be explored in future studies.

Previous studies have shown that multiple waves of regulated de novo protein synthesis are evoked by learning and required for memory consolidation (Alberini, 2009; Barzilai et al., 1989; Bourtchouladze et al., 1998; Stork and Welzl, 1999; Taubenfeld et al., 2001). More specifically, it has been found that IA training in rats very rapidly activates de novo translation in the hippocampus, and this translation is required for memory consolidation. This critical de novo protein synthesis continues, presumably through multiple waves, for more than 24 hr after training, returning to control conditions by 48 hr post-training (Arguello et al., 2013; Bambah-Mukku et al., 2014; Chen et al., 2011; Garcia-Osta et al., 2006; Taubenfeld et al., 2001). Our data showed that while a pre-training injection of receptor blocking antibody completely blocks long-term memory formation, a single injection given after training has no effect. This suggests that, once engaged, the receptor remains functionally critical for some time. In agreement with this explanation, two injections 8 hr apart given after training, but not single injections or even a single injection of 10x higher concentration of blocking antibody, were able to impair long-term memory. Together with our in vivo SUnSET data showing that the receptor controls de novo translation induced by learning, these results lead us to speculate that CIM6P/IGF2R is intimately coupled to the induction and persistence of experience-regulated de novo translation. If this were the case, we would expect that the inhibition of the receptor and of protein synthesis have similar temporal profiles of functional requirement. This is indeed the case, as, like with blockers of the receptor, a pre-training injection of the protein synthesis inhibitor anisomycin completely blocks long-term memory formation, whereas post-training injections have graded effects for the first 24 hr after training (Bambah-Mukku et al., 2014).

Nevertheless, the exact mechanisms by which CIM6P/IGF2R controls the increase in protein production remains to be determined, and ours are only speculative explanations. Importantly, our experiments could not dissect whether the decrease in puromycin incorporation upon CIM6P/IGF2R blockade was due to inhibition of mRNA translation or acceleration of protein degradation. In addition, a recent study reported that mRNA translation in axons of retinal ganglion cells occurs on ribosomes associated with late endosomes docked at mitochondria (Cioni et al., 2019). Given that CIM6P/IGF2R traffics cargo via endosomes, we speculate that the receptor may regulate endosomal availability and trafficking required for protein synthesis. Moreover, because CIM6P/IGF2R is crucial for maintaining lysosomal function (Ludwig et al., 1996; Wang et al., 1994), it may be involved in the regulation of protein degradation, which is activated in long-term plasticity and memory (Bingol and Sheng, 2011; Fernández-Monreal et al., 2012; Goo et al., 2017).

The rapid engagement of CIM6P/IGF2R implies that of one of its major ligands, IGF2, is rapidly recruited; however, previous studies showed that IGF2 is significantly upregulated by training not immediately but several hours (20 hr) later (Chen et al., 2011). It is possible that the rapid activation of the receptor occurs through other types of ligands, as the CIM6P/IGF2R can indeed bind a variety of molecules including mannosylated proteins such as lysosomal enzymes, TGF-β1 precursor, proliferin, plasminogen, and retinoic acid (El-Shewy and Luttrell, 2009).

It is also possible that IGF2 present at the basal level is released upon training, thus immediately recruiting the receptor, without changing its expression level for several hours. IGF2 upregulation at the level of transcription and translation may be needed for the persistence of the consolidation process. A similar type of regulation has been documented for the neurotrophin, brain derived neurothrophic factor (BDNF), which is immediately engaged but its requirement persists via an autoregulatory positive feedback loop throughout the whole molecular consolidation temporal window (Bambah-Mukku et al., 2014).

Finally, the signaling capabilities of CIM6P/IGF2R remain controversial: some authors suggest that CIM6P/IGF2R directly or indirectly activates G-proteins (El-Shewy et al., 2006; Hawkes et al., 2006; Ikezu et al., 1995; McKinnon et al., 2001; Nishimoto et al., 1989; Okamoto et al., 1990), whereas others have presented opposing evidence (Körner et al., 1995; Sakano et al., 1991). Therefore, further studies are needed to elucidate whether signaling activated by CIM6P/IGF2R is involved in the regulation of de novo protein synthesis or degradation.

We also showed that expression of CIM6P/IGF2R in the hippocampus is expressed predominantly in neurons, as demonstrated by co-localization of CIM6P/IGF2R with MAP2 (a neuronal marker) but not GFAP (an astrocytic marker) or Iba1 (a microglial marker). These results are in agreement with several previous reports revealing relatively high [¹²⁵I]-IGF2 radio-labelled binding in the hippocampus (Hawkes and Kar, 2003; Kar et al., 1993; Lesniak et al., 1988; Mendelsohn et al., 1988; Smith et al., 1988), as well as other immunostaining studies with different anti-CIM6P/IGF2R antibodies (Hawkes and Kar, 2003; Kar et al., 2006; Lesniak et al., 1988; Nolan et al., 1987; Zhou et al., 1995).

Our data also indicated that training does not alter total levels of CIM6P/IGF2R mRNA or protein, leading us to conclude that although CIM6P/IGF2R is critical for memory consolidation, its total level does not change following learning. However, our experiments do not exclude the possibility that learning changes the intracellular distribution and trafficking of the CIM6P/IGF2R.

Cre-recombinase-mediated knockdown of neuronal CIM6P/IGF2R resulted in impairments of long-term, but not short-term, hippocampus-dependent memories without affecting non–hippocampus-dependent behaviors. This indicates that similar to rats, mice also require hippocampal CIM6P/IGF2R for memory consolidation, but not the ability to learn. The receptor specifically contributes to memory consolidation, but it is not required for other behavioral responses such as locomotion or anxiety responses.

Finally, our data also showed that the CIM6P/IGF2R can be targeted to enhance memory. Consistent with previous reports that CIM6P/IGF2R is required for the memory-enhancing effect of hippocampal or systemically administered IGF2 (Chen et al., 2011; Stern et al., 2014), we showed that a distinct ligand of the receptor, M6P, injected either in the hippocampus in rat or systemically in mouse, significantly increased memory strength and persistence. These effects were completely abolished in mice that had reduced receptor levels in their dorsal hippocampi, strengthening the conclusion that memory enhancement promoted by either ligand is indeed mediated by CIM6P/IGF2R in hippocampal neurons.

CIM6P/IGF2R has been implicated in several aspects of Alzheimer’s disease (AD): levels of CIM6P/IGF2R decrease with the presence of APOE4 alleles (Kar et al., 2006), and overexpression of CIM6P/IGF2R increases amyloid precursor protein processing and amyloid-beta (Aβ) production (Wang et al., 2015), a hallmark alteration in AD. Furthermore, CIM6P/IGF2R co-localizes with subsets of Aβ-positive neuritic plaques (Nixon, 2005; Wang et al., 2017), and may also exert neuroprotective effects (Martin-Montañez et al., 2014; Mellott et al., 2014; Mellott et al., 2017). In light of our observations showing that neuronal CIM6P/IGF2R in the hippocampus plays a crucial role in memory consolidation and is a target mechanism for memory enhancement, this receptor may represent a key target for novel therapies for AD.

All data generated or analysed during this study are included in the manuscript and supporting files. Information about sample-size estimation is provided in the Statiscal analysis sub-section of the Materials and Methods section.

1. 1. Agis-Balboa RC
   2. Arcos-Diaz D
   3. Wittnam J
   4. Govindarajan N
   5. Blom K
   6. Burkhardt S
   7. Haladyniak U
   8. Agbemenyah HY
   9. Zovoilis A
   10. Salinas-Riester G
   11. Opitz L
   12. Sananbenesi F
   13. Fischer A

   (2011) A hippocampal insulin-growth factor 2 pathway regulates the extinction of fear memories

   The EMBO Journal 30:4071–4083.

   https://doi.org/10.1038/emboj.2011.293

   • PubMed
   • Google Scholar
2. 1. Alberini CM

   (2009) Transcription factors in long-term memory and synaptic plasticity

   Physiological Reviews 89:121–145.

   https://doi.org/10.1152/physrev.00017.2008

   • PubMed
   • Google Scholar
3. 1. Alberini CM

   (2011) The role of reconsolidation and the dynamic process of long-term memory formation and storage

   Frontiers in Behavioral Neuroscience 5:12.

   https://doi.org/10.3389/fnbeh.2011.00012

   • PubMed
   • Google Scholar
4. 1. Alberini CM
   2. Kandel ER

   (2015) The regulation of transcription in memory consolidation

   Cold Spring Harbor Perspectives in Biology 7:a021741.

   https://doi.org/10.1101/cshperspect.a021741

   • Google Scholar
5. 1. Arguello AA
   2. Ye X
   3. Bozdagi O
   4. Pollonini G
   5. Tronel S
   6. Bambah-Mukku D
   7. Huntley GW
   8. Platano D
   9. Alberini CM

   (2013) CCAAT enhancer binding protein δ plays an essential role in memory consolidation and reconsolidation

   Journal of Neuroscience 33:3646–3658.

   https://doi.org/10.1523/JNEUROSCI.1635-12.2013

   • PubMed
   • Google Scholar
6. 1. Bambah-Mukku D
   2. Travaglia A
   3. Chen DY
   4. Pollonini G
   5. Alberini CM

   (2014) A positive autoregulatory BDNF feedback loop via C/EBPβ mediates hippocampal memory consolidation

   Journal of Neuroscience 34:12547–12559.

   https://doi.org/10.1523/JNEUROSCI.0324-14.2014

   • PubMed
   • Google Scholar
7. 1. Barzilai A
   2. Kennedy TE
   3. Sweatt JD
   4. Kandel ER

   (1989) 5-HT modulates protein synthesis and the expression of specific proteins during long-term facilitation in Aplysia sensory neurons

   Neuron 2:1577–1586.

   https://doi.org/10.1016/0896-6273(89)90046-9

   • PubMed
   • Google Scholar
8. 1. Bingol B
   2. Sheng M

   (2011) Deconstruction for reconstruction: the role of proteolysis in neural plasticity and disease

   Neuron 69:22–32.

   https://doi.org/10.1016/j.neuron.2010.11.006

   • PubMed
   • Google Scholar
9. 1. Bourtchouladze R
   2. Abel T
   3. Berman N
   4. Gordon R
   5. Lapidus K
   6. Kandel ER

   (1998)

   Different training procedures recruit either one or two critical periods for contextual memory consolidation, each of which requires protein synthesis and PKA

   Learning & Memory 5:365–374.

   • PubMed
   • Google Scholar
10. 1. Chen DY
    2. Stern SA
    3. Garcia-Osta A
    4. Saunier-Rebori B
    5. Pollonini G
    6. Bambah-Mukku D
    7. Blitzer RD
    8. Alberini CM

    (2011) A critical role for IGF-II in memory consolidation and enhancement

    Nature 469:491–497.

    https://doi.org/10.1038/nature09667

    • PubMed
    • Google Scholar
11. 1. Chorney MJ
    2. Chorney K
    3. Seese N
    4. Owen MJ
    5. Daniels J
    6. McGuffin P
    7. Thompson LA
    8. Detterman DK
    9. Benbow C
    10. Lubinski D
    11. Eley T
    12. Plomin R

    (1998) A Quantitative Trait Locus Associated With Cognitive Ability in Children

    Psychological Science 9:159–166.

    https://doi.org/10.1111/1467-9280.00032

    • Google Scholar
12. 1. Cioni JM
    2. Lin JQ
    3. Holtermann AV
    4. Koppers M
    5. Jakobs MAH
    6. Azizi A
    7. Turner-Bridger B
    8. Shigeoka T
    9. Franze K
    10. Harris WA
    11. Holt CE

    (2019) Late endosomes act as mRNA translation platforms and sustain mitochondria in axons

    Cell 176:56–72.

    https://doi.org/10.1016/j.cell.2018.11.030

    • PubMed
    • Google Scholar
13. 1. Costa-Mattioli M
    2. Sossin WS
    3. Klann E
    4. Sonenberg N

    (2009) Translational control of long-lasting synaptic plasticity and memory

    Neuron 61:10–26.

    https://doi.org/10.1016/j.neuron.2008.10.055

    • PubMed
    • Google Scholar
14. 1. De Souza AT
    2. Hankins GR
    3. Washington MK
    4. Orton TC
    5. Jirtle RL

    (1995) M6P/IGF2R gene is mutated in human hepatocellular carcinomas with loss of heterozygosity

    Nature Genetics 11:447–449.

    https://doi.org/10.1038/ng1295-447

    • PubMed
    • Google Scholar
15. 1. Descalzi G
    2. Gao V
    3. Steinman MQ
    4. Suzuki A
    5. Alberini CM

    (2019) Lactate from astrocytes fuels learning-induced mRNA translation in excitatory and inhibitory neurons

    Communications Biology 2:1–11.

    https://doi.org/10.1038/s42003-019-0495-2

    • PubMed
    • Google Scholar
16. 1. El-Shewy HM
    2. Johnson KR
    3. Lee MH
    4. Jaffa AA
    5. Obeid LM
    6. Luttrell LM

    (2006) Insulin-like growth factors mediate heterotrimeric G protein-dependent ERK1/2 activation by transactivating sphingosine 1-phosphate receptors

    Journal of Biological Chemistry 281:31399–31407.

    https://doi.org/10.1074/jbc.M605339200

    • PubMed
    • Google Scholar
17. 1. El-Shewy HM
    2. Luttrell LM

    (2009) Chapter 24 insulin‐like growth factor‐2/Mannose‐6 phosphate receptors

    Vitamins & Hormones 80:667–697.

    https://doi.org/10.1016/S0083-6729(08)00624-9

    • Google Scholar
18. 1. Fernández-Monreal M
    2. Brown TC
    3. Royo M
    4. Esteban JA

    (2012) The balance between receptor recycling and trafficking toward lysosomes determines synaptic strength during long-term depression

    Journal of Neuroscience 32:13200–13205.

    https://doi.org/10.1523/JNEUROSCI.0061-12.2012

    • PubMed
    • Google Scholar
19. 1. Gallo FT
    2. Katche C
    3. Morici JF
    4. Medina JH
    5. Weisstaub NV

    (2018) Immediate early genes, memory and psychiatric disorders: focus on c-Fos, Egr1 and arc

    Frontiers in Behavioral Neuroscience 12:180.

    https://doi.org/10.3389/fnbeh.2018.00079

    • PubMed
    • Google Scholar
20. 1. Garcia-Osta A
    2. Tsokas P
    3. Pollonini G
    4. Landau EM
    5. Blitzer R
    6. Alberini CM

    (2006) MuSK expressed in the brain mediates cholinergic responses, synaptic plasticity, and memory formation

    Journal of Neuroscience 26:7919–7932.

    https://doi.org/10.1523/JNEUROSCI.1674-06.2006

    • PubMed
    • Google Scholar
21. 1. Ghosh P
    2. Dahms NM
    3. Kornfeld S

    (2003) Mannose 6-phosphate receptors: new twists in the tale

    Nature Reviews Molecular Cell Biology 4:202–213.

    https://doi.org/10.1038/nrm1050

    • Google Scholar
22. 1. Goo MS
    2. Sancho L
    3. Slepak N
    4. Boassa D
    5. Deerinck TJ
    6. Ellisman MH
    7. Bloodgood BL
    8. Patrick GN

    (2017) Activity-dependent trafficking of lysosomes in dendrites and dendritic spines

    Journal of Cell Biology 216:2499–2513.

    https://doi.org/10.1083/jcb.201704068

    • PubMed
    • Google Scholar
23. 1. Goodman CA
    2. Mabrey DM
    3. Frey JW
    4. Miu MH
    5. Schmidt EK
    6. Pierre P
    7. Hornberger TA

    (2011) Novel insights into the regulation of skeletal muscle protein synthesis as revealed by a new nonradioactive in vivo technique

    The FASEB Journal 25:1028–1039.

    https://doi.org/10.1096/fj.10-168799

    • PubMed
    • Google Scholar
24. 1. Hankins GR
    2. De Souza AT
    3. Bentley RC
    4. Patel MR
    5. Marks JR
    6. Iglehart JD
    7. Jirtle RL

    (1996)

    M6P/IGF2 receptor: a candidate breast tumor suppressor gene

    Oncogene 12:2003–2009.

    • PubMed
    • Google Scholar
25. 1. Hartig SM

    (2013) Basic Image Analysis and Manipulation in ImageJ

    Current Protocols in Molecular Biology 102:14.15.1–14.1514.

    https://doi.org/10.1002/0471142727.mb1415s102

    • Google Scholar
26. 1. Hawkes C
    2. Jhamandas JH
    3. Harris KH
    4. Fu W
    5. MacDonald RG
    6. Kar S

    (2006) Single transmembrane domain insulin-like growth factor-II/mannose-6-phosphate receptor regulates central cholinergic function by activating a G-protein-sensitive, protein kinase C-dependent pathway

    Journal of Neuroscience 26:585–596.

    https://doi.org/10.1523/JNEUROSCI.2730-05.2006

    • PubMed
    • Google Scholar
27. 1. Hawkes C
    2. Kar S

    (2003) Insulin-like growth factor-II/mannose-6-phosphate receptor: widespread distribution in neurons of the central nervous system including those expressing cholinergic phenotype

    The Journal of Comparative Neurology 458:113–127.

    https://doi.org/10.1002/cne.10578

    • PubMed
    • Google Scholar
28. 1. Hawkes C
    2. Kar S

    (2004) The insulin-like growth factor-II/mannose-6-phosphate receptor: structure, distribution and function in the central nervous system

    Brain Research Reviews 44:117–140.

    https://doi.org/10.1016/j.brainresrev.2003.11.002

    • PubMed
    • Google Scholar
29. 1. Ikezu T
    2. Okamoto T
    3. Giambarella U
    4. Yokota T
    5. Nishimoto I

    (1995) In vivo Coupling of Insulin-like Growth Factor II/Mannose 6-Phosphate Receptor to Heteromeric G Proteins

    Journal of Biological Chemistry 270:29224–29228.

    https://doi.org/10.1074/jbc.270.49.29224

    • Google Scholar
30. 1. Jedlicka P
    2. Vnencak M
    3. Krueger DD
    4. Jungenitz T
    5. Brose N
    6. Schwarzacher SW

    (2015) Neuroligin-1 regulates excitatory synaptic transmission, LTP and EPSP-spike coupling in the dentate gyrus in vivo

    Brain Structure and Function 220:47–58.

    https://doi.org/10.1007/s00429-013-0636-1

    • PubMed
    • Google Scholar
31. 1. Kar S
    2. Chabot JG
    3. Quirion R

    (1993) Quantitative autoradiographic localization of [125I]insulin-like growth factor I, [125I]insulin-like growth factor II, and [125I]insulin receptor binding sites in developing and adult rat brain

    The Journal of Comparative Neurology 333:375–397.

    https://doi.org/10.1002/cne.903330306

    • PubMed
    • Google Scholar
32. 1. Kar S
    2. Poirier J
    3. Guevara J
    4. Dea D
    5. Hawkes C
    6. Robitaille Y
    7. Quirion R

    (2006) Cellular distribution of insulin-like growth factor-II/mannose-6-phosphate receptor in normal human brain and its alteration in Alzheimer's disease pathology

    Neurobiology of Aging 27:199–210.

    https://doi.org/10.1016/j.neurobiolaging.2005.03.005

    • PubMed
    • Google Scholar
33. 1. Kim JJ
    2. Fanselow MS

    (1992) Modality-specific retrograde amnesia of fear

    Science 256:675–677.

    https://doi.org/10.1126/science.1585183

    • PubMed
    • Google Scholar
34. 1. Kong FM
    2. Anscher MS
    3. Washington MK
    4. Killian JK
    5. Jirtle RL

    (2000) M6P/IGF2R is mutated in squamous cell carcinoma of the lung

    Oncogene 19:1572–1578.

    https://doi.org/10.1038/sj.onc.1203437

    • PubMed
    • Google Scholar
35. 1. Körner C
    2. Nürnberg B
    3. Uhde M
    4. Braulke T

    (1995) Mannose 6-Phosphate/Insulin-like Growth Factor II Receptor Fails to Interact with G-proteins

    Journal of Biological Chemistry 270:287–295.

    https://doi.org/10.1074/jbc.270.1.287

    • Google Scholar
36. 1. Lau MM
    2. Stewart CE
    3. Liu Z
    4. Bhatt H
    5. Rotwein P
    6. Stewart CL

    (1994) Loss of the imprinted IGF2/cation-independent mannose 6-phosphate receptor results in fetal overgrowth and perinatal lethality

    Genes & Development 8:2953–2963.

    https://doi.org/10.1101/gad.8.24.2953

    • PubMed
    • Google Scholar
37. 1. Lesniak MA
    2. Hill JM
    3. Kiess W
    4. Rojeski M
    5. Pert CB
    6. Roth J

    (1988) Receptors for insulin-like growth factors I and II: autoradiographic localization in rat brain and comparison to receptors for insulin

    Endocrinology 123:2089–2099.

    https://doi.org/10.1210/endo-123-4-2089

    • PubMed
    • Google Scholar
38. 1. Liu X-B
    2. Murray KD

    (2012) Neuronal excitability and calcium/calmodulin-dependent protein kinase type II: Location, location, location

    Epilepsia 53:45–52.

    https://doi.org/10.1111/j.1528-1167.2012.03474.x

    • Google Scholar
39. 1. Lonergan ME
    2. Gafford GM
    3. Jarome TJ
    4. Helmstetter FJ

    (2010) Time-Dependent expression of arc and Zif268 after acquisition of fear conditioning

    Neural Plasticity 2010:1–12.

    https://doi.org/10.1155/2010/139891

    • Google Scholar
40. 1. Lovett-Barron M
    2. Kaifosh P
    3. Kheirbek MA
    4. Danielson N
    5. Zaremba JD
    6. Reardon TR
    7. Turi GF
    8. Hen R
    9. Zemelman BV
    10. Losonczy A

    (2014) Dendritic inhibition in the Hippocampus supports fear learning

    Science 343:857–863.

    https://doi.org/10.1126/science.1247485

    • PubMed
    • Google Scholar
41. 1. Ludwig T
    2. Eggenschwiler J
    3. Fisher P
    4. D'Ercole AJ
    5. Davenport ML
    6. Efstratiadis A

    (1996) Mouse mutants lacking the type 2 IGF receptor (IGF2R) are rescued from perinatal lethality in Igf2 and Igf1r null backgrounds

    Developmental Biology 177:517–535.

    https://doi.org/10.1006/dbio.1996.0182

    • PubMed
    • Google Scholar
42. 1. Martin-Montañez E
    2. Pavia J
    3. Santin LJ
    4. Boraldi F
    5. Estivill-Torrus G
    6. Aguirre JA
    7. Garcia-Fernandez M

    (2014) Involvement of IGF-II receptors in the antioxidant and neuroprotective effects of IGF-II on adult cortical neuronal cultures

    Biochimica Et Biophysica Acta (BBA) - Molecular Basis of Disease 1842:1041–1051.

    https://doi.org/10.1016/j.bbadis.2014.03.010

    • Google Scholar
43. 1. McKinnon T
    2. Chakraborty C
    3. Gleeson LM
    4. Chidiac P
    5. Lala PK

    (2001) Stimulation of human extravillous trophoblast migration by IGF-II is mediated by IGF type 2 receptor involving inhibitory G protein(s) and phosphorylation of MAPK

    The Journal of Clinical Endocrinology & Metabolism 86:3665–3674.

    https://doi.org/10.1210/jcem.86.8.7711

    • PubMed
    • Google Scholar
44. 1. Mellott TJ
    2. Pender SM
    3. Burke RM
    4. Langley EA
    5. Blusztajn JK

    (2014) IGF2 ameliorates amyloidosis, increases cholinergic marker expression and raises BMP9 and neurotrophin levels in the Hippocampus of the APPswePS1dE9 alzheimer's disease model mice

    PLOS ONE 9:e94287.

    https://doi.org/10.1371/journal.pone.0094287

    • PubMed
    • Google Scholar
45. 1. Mellott TJ
    2. Huleatt OM
    3. Shade BN
    4. Pender SM
    5. Liu YB
    6. Slack BE
    7. Blusztajn JK

    (2017) Perinatal choline supplementation reduces amyloidosis and increases choline acetyltransferase expression in the Hippocampus of the APPswePS1dE9 alzheimer's Disease Model Mice

    PLOS ONE 12:e0170450.

    https://doi.org/10.1371/journal.pone.0170450

    • PubMed
    • Google Scholar
46. 1. Mendelsohn LG
    2. Smith MC
    3. Lucaites VL
    4. Kerchner GA
    5. Ghetti B

    (1988) Autoradiographic localization of insulin-like growth factor II receptors in cerebellar cortex of Weaver and purkinje cell degeneration mutant mice

    Brain Research 458:361–366.

    https://doi.org/10.1016/0006-8993(88)90479-9

    • PubMed
    • Google Scholar
47. 1. Minatohara K
    2. Akiyoshi M
    3. Okuno H

    (2016) Role of Immediate-Early genes in synaptic plasticity and neuronal ensembles underlying the memory trace

    Frontiers in Molecular Neuroscience 8:14599.

    https://doi.org/10.3389/fnmol.2015.00078

    • Google Scholar
48. 1. Morgan DO
    2. Edman JC
    3. Standring DN
    4. Fried VA
    5. Smith MC
    6. Roth RA
    7. Rutter WJ

    (1987) Insulin-like growth factor II receptor as a multifunctional binding protein

    Nature 329:301–307.

    https://doi.org/10.1038/329301a0

    • PubMed
    • Google Scholar
49. 1. Nishimoto I
    2. Murayama Y
    3. Katada T
    4. Ui M
    5. Ogata E

    (1989)

    Possible direct linkage of insulin-like growth factor-II receptor with guanine nucleotide-binding proteins

    The Journal of Biological Chemistry 264:14029–14038.

    • PubMed
    • Google Scholar
50. 1. Nixon RA

    (2005) Endosome function and dysfunction in Alzheimer's disease and other neurodegenerative diseases

    Neurobiology of Aging 26:373–382.

    https://doi.org/10.1016/j.neurobiolaging.2004.09.018

    • PubMed
    • Google Scholar
51. 1. Nolan CM
    2. Creek KE
    3. Grubb JH
    4. Sly WS

    (1987) Antibody to the phosphomannosyl receptor inhibits recycling of receptor in fibroblasts

    Journal of Cellular Biochemistry 35:137–151.

    https://doi.org/10.1002/jcb.240350207

    • PubMed
    • Google Scholar
52. 1. Oka Y
    2. Rozek LM
    3. Czech MP

    (1985)

    Direct demonstration of rapid insulin-like growth factor II receptor internalization and recycling in rat adipocytes insulin stimulates 125I-insulin-like growth factor II degradation by modulating the IGF-II receptor recycling process

    The Journal of Biological Chemistry 260:9435–9442.

    • PubMed
    • Google Scholar
53. 1. Okamoto T
    2. Katada T
    3. Murayama Y
    4. Ui M
    5. Ogata E
    6. Nishimoto I

    (1990) A simple structure encodes G protein-activating function of the IGF-II/mannose 6-phosphate receptor

    Cell 62:709–717.

    https://doi.org/10.1016/0092-8674(90)90116-V

    • PubMed
    • Google Scholar
54. 1. Ouyang H
    2. Shiwaku HO
    3. Hagiwara H
    4. Miura K
    5. Abe T
    6. Kato Y
    7. Ohtani H
    8. Shiiba K
    9. Souza RF
    10. Meltzer SJ
    11. Horii A

    (1997)

    The insulin-like growth factor II receptor gene is mutated in genetically unstable cancers of the endometrium, stomach, and colorectum

    Cancer Research 57:1851–1854.

    • PubMed
    • Google Scholar
55. 1. Pascual‐Lucas M
    2. Viana da Silva S
    3. Di Scala M
    4. Garcia‐Barroso C
    5. González‐Aseguinolaza G
    6. Mulle C
    7. Alberini CM
    8. Cuadrado‐Tejedor M
    9. Garcia‐Osta A

    (2014) Insulin‐like growth factor 2 reverses memory and synaptic deficits in APP transgenic mice

    EMBO Molecular Medicine 6:1246–1262.

    https://doi.org/10.15252/emmm.201404228

    • Google Scholar
56. 1. Phillips RG
    2. LeDoux JE

    (1992) Differential contribution of amygdala and Hippocampus to cued and contextual fear conditioning

    Behavioral Neuroscience 106:274–285.

    https://doi.org/10.1037/0735-7044.106.2.274

    • PubMed
    • Google Scholar
57. 1. Richter JD
    2. Klann E

    (2009) Making synaptic plasticity and memory last mechanisms of translational regulation

    Genes & Development 23:1–11.

    https://doi.org/10.1101/gad.1735809

    • PubMed
    • Google Scholar
58. 1. Roy DS
    2. Arons A
    3. Mitchell TI
    4. Pignatelli M
    5. Ryan TJ
    6. Tonegawa S

    (2016) Memory retrieval by activating Engram cells in mouse models of early Alzheimer's disease

    Nature 531:508–512.

    https://doi.org/10.1038/nature17172

    • PubMed
    • Google Scholar
59. 1. Sakano K
    2. Enjoh T
    3. Numata F
    4. Fujiwara H
    5. Marumoto Y
    6. Higashihashi N
    7. Sato Y
    8. Perdue JF
    9. Fujita-Yamaguchi Y

    (1991)

    The design, expression, and characterization of human insulin-like growth factor II (IGF-II) mutants specific for either the IGF-II/cation-independent mannose 6-phosphate receptor or IGF-I receptor

    The Journal of Biological Chemistry 266:20626–20635.

    • PubMed
    • Google Scholar
60. 1. Santini E
    2. Huynh TN
    3. MacAskill AF
    4. Carter AG
    5. Pierre P
    6. Ruggero D
    7. Kaphzan H
    8. Klann E

    (2013) Exaggerated translation causes synaptic and behavioural aberrations associated with autism

    Nature 493:411–415.

    https://doi.org/10.1038/nature11782

    • PubMed
    • Google Scholar
61. 1. Santini E
    2. Huynh TN
    3. Klann E

    (2014) Mechanisms of translation control underlying long-lasting synaptic plasticity and the consolidation of long-term memory

    Progress in Molecular Biology and Translational Science 122:131–167.

    https://doi.org/10.1016/B978-0-12-420170-5.00005-2

    • PubMed
    • Google Scholar
62. 1. Schmidt EK
    2. Clavarino G
    3. Ceppi M
    4. Pierre P

    (2009) SUnSET, a nonradioactive method to monitor protein synthesis

    Nature Methods 6:275–277.

    https://doi.org/10.1038/nmeth.1314

    • Google Scholar
63. 1. Schrick C
    2. Fischer A
    3. Srivastava DP
    4. Tronson NC
    5. Penzes P
    6. Radulovic J

    (2007) N-cadherin regulates cytoskeletally associated IQGAP1/ERK signaling and memory formation

    Neuron 55:786–798.

    https://doi.org/10.1016/j.neuron.2007.07.034

    • PubMed
    • Google Scholar
64. 1. Smith M
    2. Clemens J
    3. Kerchner GA
    4. Mendelsohn LG

    (1988) The insulin-like growth factor-II (IGF-II) receptor of rat brain: regional distribution visualized by autoradiography

    Brain Research 445:241–246.

    https://doi.org/10.1016/0006-8993(88)91185-7

    • PubMed
    • Google Scholar
65. 1. Steinmetz AB
    2. Johnson SA
    3. Iannitelli DE
    4. Pollonini G
    5. Alberini CM

    (2016) Insulin-like growth factor 2 rescues aging-related memory loss in rats

    Neurobiology of Aging 44:9–21.

    https://doi.org/10.1016/j.neurobiolaging.2016.04.006

    • PubMed
    • Google Scholar
66. 1. Steinmetz AB
    2. Stern SA
    3. Kohtz AS
    4. Descalzi G
    5. Alberini CM

    (2018) Insulin-Like growth factor II targets the mTOR pathway to reverse Autism-Like phenotypes in mice

    The Journal of Neuroscience 38:1015–1029.

    https://doi.org/10.1523/JNEUROSCI.2010-17.2017

    • PubMed
    • Google Scholar
67. 1. Stern SA
    2. Kohtz AS
    3. Pollonini G
    4. Alberini CM

    (2014) Enhancement of memories by systemic administration of insulin-like growth factor II

    Neuropsychopharmacology 39:2179–2190.

    https://doi.org/10.1038/npp.2014.69

    • PubMed
    • Google Scholar
68. 1. Stork O
    2. Welzl H

    (1999) Memory formation and the regulation of gene expression

    Cellular and Molecular Life Sciences 55:575–592.

    https://doi.org/10.1007/s000180050316

    • PubMed
    • Google Scholar
69. 1. Sutton MA
    2. Schuman EM

    (2006) Dendritic protein synthesis, synaptic plasticity, and memory

    Cell 127:49–58.

    https://doi.org/10.1016/j.cell.2006.09.014

    • PubMed
    • Google Scholar
70. 1. Taubenfeld SM
    2. Milekic MH
    3. Monti B
    4. Alberini CM

    (2001) The consolidation of new but not reactivated memory requires hippocampal C/EBPβ

    Nature Neuroscience 4:813–818.

    https://doi.org/10.1038/90520

    • Google Scholar
71. 1. Tischmeyer W
    2. Grimm R

    (1999) Activation of immediate early genes and memory formation

    Cellular and Molecular Life Sciences 55:564–574.

    https://doi.org/10.1007/s000180050315

    • PubMed
    • Google Scholar
72. 1. Vasileiou T
    2. Foresti D
    3. Bayram A
    4. Poulikakos D
    5. Ferrari A

    (2016) Toward Contactless Biology: Acoustophoretic DNA Transfection

    Scientific Reports 6:1–10.

    https://doi.org/10.1038/srep20023

    • Google Scholar
73. 1. Veyrac A
    2. Besnard A
    3. Caboche J
    4. Davis S
    5. Laroche S

    (2014) The transcription factor Zif268/Egr1, brain plasticity, and memory

    Progress in Molecular Biology and Translational Science 122:89–129.

    https://doi.org/10.1016/B978-0-12-420170-5.00004-0

    • PubMed
    • Google Scholar
74. 1. Wang ZQ
    2. Fung MR
    3. Barlow DP
    4. Wagner EF

    (1994) Regulation of embryonic growth and lysosomal targeting by the imprinted Igf2/Mpr gene

    Nature 372:464–467.

    https://doi.org/10.1038/372464a0

    • PubMed
    • Google Scholar
75. 1. Wang DO
    2. Kim SM
    3. Zhao Y
    4. Hwang H
    5. Miura SK
    6. Sossin WS
    7. Martin KC

    (2009) Synapse- and stimulus-specific local translation during long-term neuronal plasticity

    Science 324:1536–1540.

    https://doi.org/10.1126/science.1173205

    • PubMed
    • Google Scholar
76. 1. Wang Y
    2. Buggia-Prévot V
    3. Zavorka ME
    4. Bleackley RC
    5. MacDonald RG
    6. Thinakaran G
    7. Kar S

    (2015) Overexpression of the Insulin-Like growth factor II receptor increases β-Amyloid production and affects cell viability

    Molecular and Cellular Biology 35:2368–2384.

    https://doi.org/10.1128/MCB.01338-14

    • PubMed
    • Google Scholar
77. 1. Wang Y
    2. MacDonald RG
    3. Thinakaran G
    4. Kar S

    (2017) Insulin-Like growth Factor-II/Cation-Independent mannose 6-Phosphate receptor in neurodegenerative diseases

    Molecular Neurobiology 54:2636–2658.

    https://doi.org/10.1007/s12035-016-9849-7

    • PubMed
    • Google Scholar
78. 1. Wylie AA
    2. Pulford DJ
    3. McVie-Wylie AJ
    4. Waterland RA
    5. Evans HK
    6. Chen YT
    7. Nolan CM
    8. Orton TC
    9. Jirtle RL

    (2003) Tissue-specific inactivation of murine M6P/IGF2R

    The American Journal of Pathology 162:321–328.

    https://doi.org/10.1016/S0002-9440(10)63823-0

    • PubMed
    • Google Scholar
79. 1. Ye X
    2. Kapeller-Libermann D
    3. Travaglia A
    4. Inda MC
    5. Alberini CM

    (2017) Direct dorsal hippocampal-prelimbic cortex connections strengthen fear memories

    Nature Neuroscience 20:52–61.

    https://doi.org/10.1038/nn.4443

    • PubMed
    • Google Scholar
80. 1. Zhang WP
    2. Guzowski JF
    3. Thomas SA

    (2005) Mapping neuronal activation and the influence of adrenergic signaling during contextual memory retrieval

    Learning & Memory 12:239–247.

    https://doi.org/10.1101/lm.90005

    • PubMed
    • Google Scholar
81. 1. Zhou M
    2. Ma Z
    3. Sly WS

    (1995) Cloning and expression of the cDNA of chicken cation-independent mannose-6-phosphate receptor

    PNAS 92:9762–9766.

    https://doi.org/10.1073/pnas.92.21.9762

    • PubMed
    • Google Scholar

Acceptance summary:

This study uses a combination of biochemical and behavioral approaches to demonstrate a critical role for the CIM6P/IGF2 receptor in the hippocampus in long-term memory. The authors report that neuronal CIM6P/IGF2R is required for learning-induced de novo protein translation and works as a target mechanism for memory enhancement. The authors provided a thoughtful and detailed response in regards to previous reviewer comments. The previous concerns included a request for an additional experiment, which the authors addressed based on the current literature and the confounds associated with the inability to do additional experiments during COVID-19. Given eLife policy regarding the current shutdown, the authors' detailed response is sufficient.

Decision letter after peer review:

Thank you for submitting your article "A role for CIM6P/IGF2 receptor in memory consolidation and enhancement" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Lisa M Monteggia as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Gary Westbrook as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Kobi Rosenblum (Reviewer #2).

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:

This study uses IHC, behavior, biochemistry, genetics and molecular research to show the role of IGF2 receptor specifically in memory consolidation. They show that CIM6P/IGF2R receptor has an important role in hippocampus-dependent memory consolidation in rats and mice. Blocking CIM6P/IGF2R receptor 15 min before fear conditioning training impaired memory consolidation. In addition, using pharmacological and genetic approaches, the authors show that CIM6P/IGF2R receptor controls de novo protein synthesis and up regulation of Arc, Egr1 and c-Fos proteins following fear condition training. Moreover, administration of mannose-6-phosphate/IGF2, CIM6P/IGF2R ligands, enhance memory retention which was occluded by CIM6P/IGF2R receptor cre induced expression in the dorsal hippocampus in mice. This is a well carried out study that shows IGF2R is required for hippocampal memory consolidation in mice and rats.

This paper is an extension of the authors previous work on IGF2 and suggests that IGF2R in the hippocampus may be more important than IGF2 in memory. Although this is a strong study that convincingly shows IGF2R is required for hippocampal memory consolidation there were concerns that should be addressed.

Major comment:

All three reviewers raised the question regarded how the authors link IGF2R and the downstream IEG translation. Is there any mechanistic insight into how IGF2R links to IEG translation?

Comments:

1) The authors show the manipulation does not affect retrieval. Does it affect reconsolidation?

2) Measuring intensity in the IHC can be problematic. Do the authors have data for n of cells? Can they use an internal control (not relevant brain structure)?

3) Can the authors speculate why the window of IGF2R engagement is so long? It seems unlikely that circulating ligands are present for 8hrs after learning. Indeed, blocking IGF2R leads to rapid memory impairments (5 min-1hr after learning). To clarify this point, the authors could inject the antibody immediately and 5 min or 1hr after training.

4) This rapid engagement of IGFR2 is in contrast to the authors previous work (Chen et al., 2011) showing that IGF2 expression is only increased 20hrs after avoidance training in rats. No increase was observed prior to that time point. Is this because other ligands are important immediately after learning in the normal context?

5) Subsection “Neuronal CIM6P/IGF2R is required for the formation of hippocampus-dependent memories”. Figure 4A is not knock down data, this is Figure 4G-I.

6) Discussion, eighth paragraph, last sentence, please correct.

7) In figure legend please specify rat or mice according to the experiment.

8) It is not clear which subregions are targeted in the stereotaxic injections of rats and mice as the coordinates are missing in the "Materials and methods".

9) Figure 1 – western blot data in Figure 1C, D. Figure 1C shows one upper band and a lower band ~ 250kDa, while Figure 1D shows two upper bands without the lower band ~ 250kDa. It is not clear which band is the real signal, raising concern about the specificity of the antibody in rat tissue although it seems to be specific in mouse tissue.

10) Figure 2: Is the inference that the single injection is ineffective because the receptor is already engaged sufficiently? Would it be useful to have this be measured somehow?

11) Figure 3: Are we talking about the same cohort of animals across all panels?

12) Figure 4F the authors see impaired contextual memory at 5 min but not 1 min after the training. It is not clear how to define these two processes temporally. Could the freezing level of mice 1 min after training be due to the response to the shock but not the index of learning? The authors state n=8 for the IHC data. Given that they further state in the Materials and methods 2 slices/animals, over a range of 0.4mm in rats and 0.5mm in mice, this should be defined more specifically.

13) Figure 5: The IHC data here needs to account for the range over which it was collected over, in order to be comparable to mRNA studies, conducted with extracted/dissected hippocampi.

Are there any differences in the number of positive cells for the protein products of these IEGs?

14) Figure 6: it was shown previously in Figure 3C in rats that there is significant effect of anti CIM6P/IGF2R on memory tested 1h after training, so why in in Figure 6 protein synthesis using puromycin was measured 2 h following inhibitory place avoidance training (subsection “CIM6P/IGF2R is required for the training-induced increase in de novo synthesized protein”).

15) Figure 7: Considering the dose response obtained with M6P in rats, and the lack of an effect at the high dose, this might be particularly revealing of the potential and limitations associated with the ligand. Does the high dose cause defects in physiology or any other memory-related paradigms?

[…]

Major comment:

All three reviewers raised the question regarded how the authors link IGF2R and the downstream IEG translation. Is there any mechanistic insight into how IGF2R links to IEG translation?

We agree with the reviewers that this is one of the most important next questions that should be addressed. We are working extensively towards the identification of the mechanisms directly controlled by CIM6P/IGF2R.

The current knowledge on the mechanisms of action of CIM6P/IGF2R has derived mostly from studies carried out on tumor cell lines and embryonic development, which then have been corroborated by knockout mice investigations. These studies identified a role for CIMPR/IGF2R in trafficking certain ligands such as IGF2 and lysosomal enzymes (mannose-6-phosphate tagged proteins) to and from endosomal compartments, lysosomes, Golgi apparatus, and/or plasma membrane.

To our knowledge, ours is the first study that has examined the role of CIM6P/IGF2R in the healthy brain and has identified its crucial function in memory formation. Because CIM6P/IGF2R has a small intracellular domain and does not have known direct signal transduction mechanisms (El-Shewy and Luttrell, 2009), hypothesizing molecular mechanisms directly activated downstream is not straightforward.

In our study, we therefore approached the question of CIM6P/IGF2R downstream mechanisms by testing whether the receptor controls fundamental and rapidly engaged cellular processes (close to plasma membrane) that are known to be promptly recruited by learning and required for long-term memory. Hence, we assessed whether the receptor regulates activity-induced de novo translation. Using in vivo SUnSET, we found that de novo translation evoked by learning is, in fact, downstream of the receptor; blocking the receptor completely blocked learning-induced translation.

The effect on translation was then extended to the translation of specific proteins, i.e., IEGs. Blockade of the receptor did not affect the induction of IEG transcription upon learning, but completely blocked their translation. These results provided key information on mechanisms downstream of the CIM6P/IGF2R and suggested that our future studies should focus on mechanisms that link endosomal trafficking, lysosomal degradation and Golgi functions to learning-induced de novo translation.

This is a novel and complex area of investigation, and while we are devoting many resources to answering this important question, finding the answer will require multiple approaches (in vitro and in vivo, targeted and unbiased) and will not be a matter of months. Given the significant amount of work involved, we feel that this aim is outside the scope of the current manuscript and hope that the reviewers agree.

Comments:

1) The authors show the manipulation does not affect retrieval. Does it affect reconsolidation?

In Figure 3E, we show that injecting the blocking antibody 15 min before memory test (retrieval) given at 24 hours after training had no effect on that test, hence on memory retrieval. We then tested again the memory of those animals 7 days later, and again found no effect on memory retention. These data indicated that reconsolidation was not affected. Like what we found for consolidation, where a pre-training injection completely blocked memory, if CIM6P/IGF2R was required for the rapidly induced fragility and restabilization of memory upon retrieval, i.e. reconsolidation, pre-retrieval injection should have impaired memory at the 7 day test. In our previous submission we did not emphasize these results because the paper focuses on the consolidation process.

Now we included the explanation of the no-effects at the 7 days test after the pre-retrieval injection of the receptor blocking antibody, hence the lack of an effect on reconsolidation, in the Abstract, Results, and Discussion sections.

2) Measuring intensity in the IHC can be problematic. Do the authors have data for n of cells? Can they use an internal control (not relevant brain structure)?

To address this question, we have now re-analyzed our intensity data in order to normalize them against the total number of cells present in the quantified samples.

For each image, we quantified the total number of cells using DAPI (4′,6-diamidino-2-phenylindole) staining, a marker commonly used to visualize nuclei. As we have now detailed in the Materials and methods section, each image was subjected to background subtraction, thresholding, fill holes, watershed, and the cells were quantified using analyze particles (Hartig, 2013; Roy et al., 2016; Jedlicka et al., 2015; Vasileiou et al., 2016). The number of DAPI-positive cells for all images per sub-region were summed to calculate the total number of cells/sub-region, per rat. For each rat, 1100 – 1600 cells were quantified in CA1, and 1900 – 3000 cells in DG. The total number of cells per rat in each experimental group (effects of training and treatment) underwent statistical analyses to determine if changes among the groups were found. No difference in the averaged total number of DAPI-positive cells was found across training and/or treatment groups, in either sub-region (these data are depicted in a new figure: Figure 5—figure supplement 1).

Then, we re-analyzed the immunofluorescence total intensity for each IEG, in both CA1 and DG, by normalizing the raw integrated density values against the total number of DAPI-positive cells for that sub-region (graphs in Figure 5B-D). This new analysis showed that training significantly increases the intensity of all three IEGs, and that the receptor-blocking antibody blocked this increase. The graphs reporting these data have been included in the updated Figure 5 and the new statistical analyses is shown in the Supplementary file 1.

In addition, on the same images, we quantified the number of IEG-positive cells, normalized this number against the total number DAPI-positive cells for each rat, and expressed the results in % change for each sub-region/rat. Statistical analyses across the grouped values determined changes related to training and/or treatment. As shown in the updated Figure 5 (statistical analyses in Supplementary file 1), training significantly increased the percent of Arc-positive cells in both CA1 and DG. The same effect was observed in DG for Egr1 and c-Fos in DG but not in CA1 where the % of positive cells did not change (but the intensity did as described above). All these changes were blocked by the receptor-blocking antibody.

To re-summarize, training significantly increased both the number of Arc-positive cells, as well as Arc total intensity. However, in the case of Egr1 and c-Fos, training differentially affected the two hippocampal sub-regions: while it significantly increased Egr1 and c-Fos cell number and intensity in DG, it only increased Egr1 and c-Fos intensity but not the number of positive cells in CA1, indicating that cells already expressing Egr1 and c-Fos upregulated their levels. All these training-induced changes required CIM6P/IGF2R, as they were completely blocked by the receptor-blocking antibody.

These findings are now described in the Results section; their statistical analyses are detailed in Supplementary file 1, and the related methods have been added to the Materials and methods section.

We think that using a putative “non relevant” brain region, presumably not involved in inhibitory avoidance memory consolidation, is most likely not the best control, because changes in IEGs are not selective for memory but occur in response to a variety of brain activations. These changes can occur for instance in response to sensory and motor stimuli. In other words, such an experiment would be informative only if no changes are found in the putative “non relevant” area.

3) Can the authors speculate why the window of IGF2R engagement is so long? It seems unlikely that circulating ligands are present for 8hrs after learning. Indeed, blocking IGF2R leads to rapid memory impairments (5 min-1hr after learning). To clarify this point, the authors could inject the antibody immediately and 5 min or 1hr after training.

We know from previous studies that, in the rat hippocampus, the phase of inhibitory avoidance consolidation that requires transcription starts very rapidly with training and then persists for more than 24 hours; finally returning to control conditions by 2 days after training (e.g. Taubenfeld et al., 2001; Bambah-Mukku et al., 2014; Garcia-Osta et al., 2006; Chen et al., 2011; Arguello et al., 2013).

Therefore, it is not surprising that the requirement for CIMPR/IGF2R during memory consolidation can extend for several hours. Circulating ligands could be engaged for the entire consolidation temporal window, but, even if they were involved for a more limited time, they could functionally influence subsequent processes for an extended time.

Toward the delineation of the temporal window of receptor requirement, we have carried out a relatively extensive temporal and concentration testing (shown in Figure 1). These experiments revealed that inhibition of the receptor acutely prior to training completely blocks memory formation; however, the receptor needs to be blocked for several hours after training, through multiple injections, in order to show its critical role during consolidation. In fact, even an acute injection of a 10 times higher concentration of the blocking antibody immediately after training has no effect on memory. Hence, these experiments indicated that training very rapidly engages the receptor, which then remains functionally recruited for an extended time.

In normal circumstances, we would have carried out one of the experiments suggested by the reviewer, and specifically testing the effect of 2 injections post-training separated by one hour. We believe that the other experiment of testing a double injection with the first one delivered immediately after training and the second 5 minutes later, is a condition already addressed by the single post-training injection at 10x higher concentration (no effect). However, given the uncertainty of when the experiment could be done due to the COVID-19 situation and the fact that our tests have already provided an extensive temporal investigation, we decided to clarify our results and add a discussion point (as detailed below).

We think that the additional experiment of 2 injections separated by one hour would only perhaps confirm that a shorter temporal window between 2 injections is also effective. We hope that the reviewer will be satisfied. However, if the reviewer feels that we should perform this experiment, we will do so after the COVID resolution, but we need to request an extension for our resubmission until when we will be allowed to return to the laboratory.

To summarize, a single injection given either before training or immediately after training, or at 8 hours after training, or even a single post-training injection at 10x concentration, along with double injections given 8 hours apart, showed that, once it is rapidly recruited by training, the CIM6P/IGF2R needs to remain functionally engaged for several hours. The mechanisms by which the receptor is functionally recruited for several hours following training remain to be identified. However, here we offer some speculations: 1) Our data based on in vivo SUnSET indicated that activation of CIMPR/IGF2R is upstream of learning-induced mRNA translation. Previous studies have shown that, in order for memories to be successfully consolidated, multiple waves of de novo mRNA translation are required (Barzilai et al., 1989; Bourtchouladze et al., 1998; Taubenfeld et al., 2001; Alberini, 2009; Stork and Welzl, 1999). Moreover, as mentioned, the requirement for de novo translation in the hippocampus following IA continues for more than 24 hours after training (Bambah-Mukku et al., 2014; Bekinschtein et al., 2007). Hence, if, as suggested by our data, CIM6P/IGF2R is temporally coupled to de novo translation, it is plausible that its functional engagement follows the same temporal window of requirement.

Indeed, a pre-training injection of the receptor blocking antibody, like that of translation inhibitors (e.g. anisomycin) completely block memory formation, whereas post-training injections of either blocking antibodies or protein synthesis inhibitors require more extended treatments (Bambah-Mukku et al., 2014; Milekic et al., 2006).

To address the reviewers’ question, we have now added these clarifications and speculations in the Discussion section.

4) This rapid engagement of IGFR2 is in contrast to the authors previous work (Chen et al., 2011) showing that IGF2 expression is only increased 20hrs after avoidance training in rats. No increase was observed prior to that time point. Is this because other ligands are important immediately after learning in the normal context?

In our previous work (Chen et al., 2011), we found that mRNA levels of IGF2 in the hippocampus did not change for several hours after IA training, but significantly increased at 20 hours post-training and then returned to control levels by 48 hours after training. IGF2 protein levels followed the same profile. Lack of protein induction in response to training, however, does not exclude that that protein, which is already present at a lower level at the time of training, is readily functionally engaged. Changes in expression may be additionally needed to either sustain the protein function during the temporal window of consolidation (effect on persistence) and/or to replenish the consumed protein.

A similar condition has been found with brain-derived neurotrophic factor (BDNF). BDNF is rapidly engaged through training and immediately required for the formation of long-term memory without rapidly changing its protein levels. Its level in fact significantly augments only at later timepoints, and this late increase, which is part of a positive autoregulatory loop, is required for memory consolidation (Bambah-Mukku et al., 2014).

Thus, IGF2 could be readily engaged via CIM6P/IGF2R at the time of training, and then increase its expression level at later time points, while continuing to be functionally required (Chen et al., 2011).

Nevertheless, we cannot exclude, and agree with the reviewers’ suggestion, that at different timepoints after training, the receptor may engage other ligands in physiological conditions. In fact, it is known that CIM6P/IGF2R can bind a variety of ligands, including mannosylated proteins such as lysosomal enzymes, TGF-β1 precursor, proliferin, plasminogen, and retinoic acid. We have added a discussion point on this issue (Discussion).

5) Subsection “Neuronal CIM6P/IGF2R is required for the formation of hippocampus-dependent memories”. Figure 4A is not knock down data, this is Figure 4G-I.

This has been corrected.

6) Discussion, eighth paragraph, last sentence, please correct.

This sentence has been re-worded to make this point clearer.

7) In figure legend please specify rat or mice according to the experiment.

We have now clarified whether the experiment was conducted in rats, mice, or both in each figure legend.

8) It is not clear which subregions are targeted in the stereotaxic injections of rats and mice as the coordinates are missing in the "Materials and methods".

All injections in rats targeted dorsal hippocampus. We detailed the rat coordinates for cannula placement and injection in the “Materials and methods” sub-section “Cannula implants and hippocampal injections in rats”. The coordinates of stereotactic injections of viruses in mice are detailed in the “Materials and methods” sub-section “Viral injections in mice”.

9) Figure 1 – western blot data in Figure 1C, D. Figure 1C shows one upper band and a lower band ~ 250kDa, while Figure 1D shows two upper bands without the lower band ~ 250kDa. It is not clear which band is the real signal, raising concern about the specificity of the antibody in rat tissue although it seems to be specific in mouse tissue.

We apologize for any confusion in the western blot data. Both Figure 1C and D are data from rat lysates; however, the tissue processing was not identical, as part of the goal in Figure 1C was to compare the synaptoneurosomal and total fractions from untrained and trained rats. In the experiment shown in Figure 1D, the tissue was dissected and snap frozen, then later homogenized in RIPA buffer for protein extraction. In Figure 1C, the tissue was freshly dissected and homogenized in synaptoneurosomal buffer (see Materials and methods), in order to allow separation of synaptosomes. An aliquot of this suspension was extracted in RIPA buffer to generate the total input protein extract. It is possible that the difference in lysate preparation procedures may lead to the slightly different appearance in the bands.

Nevertheless, it is established that the CIM6P/IGF2R molecular weight is about 300 kDa. The 300 kDa band is the strongest band (can appear as a doublet relative to the separation) evident with either type of extraction protocols, and well above 250 kDa. This reactivity has been verified by Abcam to be blocked by CIMPR/IGF2R peptides. In our experiments we have quantified the strong 300 kDa signal. The lower band in Figure 1C could be a non-specific reactivity.

Notably, a similar 300 kDa band was also detected in total protein extracts obtained from mouse dorsal hippocampus (homogenized in RIPA buffer; Figure 4G). We have validated the specificity of this 300 kDa band with genetic knockdown: as shown in Figure 4G, the 300 kDa band was significantly decreased in homogenates obtained from mice that underwent Cre-recombinase-dependent knockdown of CIM6P/IGF2R.

10) Figure 2: Is the inference that the single injection is ineffective because the receptor is already engaged sufficiently? Would it be useful to have this be measured somehow?

Please see also response above to point 3. Our results indeed indicate that the timing of receptor engagement is key. When comparing a single injection given 15 min before training (Figure 2F) to one given immediately after training (Figure 2B), the outcome is radically different: the former completely blocks memory formation, while the latter has no effect. This suggests that the receptor is recruited very rapidly with training and very quickly activates downstream mechanism(s). The few minutes after training, that elapse before the antibody blocks the receptor, are sufficient for the receptor itself to be engaged sufficiently. Once engaged by the training event, the receptor seems to remain engaged for several hours, as a single injection no longer affects memory retention, but a double injection, hours apart, does. We do not know of methods available other than blocking the receptor (with the blocking antibody) to efficiently measure receptor engagement.

11) Figure 3: Are we talking about the same cohort of animals across all panels?

Independent cohorts of rats were used for each lettered panel in Figure 3, in order to avoid the confound of multiple testing. We have clarified this point in the text.

12) Figure 4F – the authors see impaired contextual memory at 5 min but not 1 min after the training. It is not clear how to define these two processes temporally. Could the freezing level of mice 1 min after training be due to the response to the shock but not the index of learning? The authors state n=8 for the IHC data. Given that they further state in the Materials and methods 2 slices/animals, over a range of 0.4mm in rats and 0.5mm in mice, this should be defined more specifically.

We do not interpret the freezing at 1 min as the result of shock only; rather, we think that our experiments show a very rapid effect on blocking the CIM6P/IGF2R on memory retention. In several studies, in fact, memory retention is tested immediately after training (e.g. Radulovic et al., 1998; Balogh et al., 2003). We are not aware of any work indicating that associative memory requires time to be expressed.

As the receptor is known to be involved in endosomal trafficking, the rapid mechanism could be, for example, vesicle trafficking and/or local mRNA translation. In line with the latter idea, our results showed that pre-training receptor blockade completely disrupts the de novo translation induced by learning, a process known to occur very rapidly. We added a discussion point (Discussion).

We apologize for any confusion regarding the coordinates for IHC sections. We have edited the text in the Materials and methods section, to clarify that the sections were taken at the two specified coordinates from each animal.

13) Figure 5: The IHC data here needs to account for the range over which it was collected over, in order to be comparable to mRNA studies, conducted with extracted/dissected hippocampi.

Are there any differences in the number of positive cells for the protein products of these IEGs?

The coordinates for both IHC and RNA extracted are the dorsal hippocampus, the focus of our studies. The RNA was extracted from the entire dHC. For the IHC analyses we quantified 1 section at approximately Bregma -3.6 mm and another at -4.0 mm for each animal on each side of the dorsal hippocampus. These sections provide an average of sample-changes along the dHC.

As detailed above in point 2, we quantified the number of IEG-positive cells, normalized this number against the total number DAPI-positive cells in each rat, and expressed the results as a percent of positive cells for each sub-region. As shown in the updated Figure 5, training significantly increased the percentage of Arc-positive cells in both CA1 and DG. This effect was blocked in the rats that received the receptor-blocking antibody. The same effect was observed in DG for Egr1 and c-Fos. However, training did not change the percentage of Egr1- or c-Fos-positive cells in CA1(but did change the intensity of expression). Graphs depicting this new analysis have been added to Figure 5, and their statistics added to Supplementary file 1.

All these training-induced changes required CIM6P/IGF2R, as they were completely blocked by the receptor-blocking antibody.

These findings are now described in the Results section; their statistical analyses are detailed in Supplementary file 1, and the related methods have been added to the Materials and methods section.

14) Figure 6: it was shown previously in Figure 3C in rats that there is significant effect of anti CIM6P/IGF2R on memory tested 1h after training, so why in in Figure 6 protein synthesis using puromycin was measured 2 h following inhibitory place avoidance training (subsection “CIM6P/IGF2R is required for the training-induced increase in de novo synthesized protein”).

With the in vivo SUnSET assay measured with IHC analysis we did not intend to correlate the result of puromycin incorporation with the 1 hour memory impairment, but aimed at determining the effect of receptor blockade on the level of de novo translation. We had established conditions in the lab working reliably for the in vivo SUnSET assessment at 2 hours after training; hence, we used that timepoint.

15) Figure 7: Considering the dose response obtained with M6P in rats, and the lack of an effect at the high dose, this might be particularly revealing of the potential and limitations associated with the ligand. Does the high dose cause defects in physiology or any other memory-related paradigms?

It is very common in pharmacology that dose-response curves of compounds show optimal concentration effects and then higher concentrations produce impairments. Usually these types of curves indicate that higher doses may have other effects. In fact, we found similar outcomes with systemic IGF2 injections: mice given a systemic injection of IGF2 only exhibited enhanced memory for contextual fear conditioning within a given concentration range, while lower and higher doses did not elicit memory enhancement (Stern et al., 2016a).

We did not detect any overt deficits in rats injected with the high dose of M6P, however, as stated in the manuscript, the dose was injected only once.

References:

Balogh, S.A., Radcliffe, R.A., Logue, S.F., Wehner, J.M., 2003. Contextual and cued fear conditioning in C57BL/6J and DBA/2J mice: Context discrimination and the effects of retention interval. Behav. Neurosci. 116, 947–57. doi:10.1037/0735-7044.116.6.947

Bekinschtein, P., Cammarota, M., Igaz, L.M., Bevilaqua, L.R.M., Izquierdo, I., Medina, J.H., 2007. Persistence of Long-Term Memory Storage Requires a Late Protein Synthesis- and BDNF- Dependent Phase in the Hippocampus. Neuron 53, 261–277. doi:10.1016/j.neuron.2006.11.025

Milekic MH, Brown SD, Castellini C, Alberini CM. (2006) Persistent disruption of an established morphine conditioned place preference. J Neurosci. 2006 Mar 15;26(11):3010-20.

Radulovic, J., Kammermeier, J., Spiess, J., 1998. Generalization of fear

responses in C57BL/6N mice subjected to one-trial foreground contextual fear conditioning. Behav Brain Res 95, 179–189. doi:10.1016/S0166-4328(98)00039-4.

Author details

1. Xiao-Wen Yu

   Center for Neural Science, New York University, New York, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology, Writing - original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8974-0085

2. Kiran Pandey

   Center for Neural Science, New York University, New York, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-8658-2942

3. Aaron C Katzman

   Center for Neural Science, New York University, New York, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

4. Cristina M Alberini

   Center for Neural Science, New York University, New York, United States

   Contribution

   Conceptualization, Data curation, Supervision, Funding acquisition, Investigation, Methodology, Writing - original draft, Writing - review and editing

   For correspondence

   ca60@nyu.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7386-0018

• 2,083

  Page views

• 282

  Downloads

• 14

  Citations

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