Recruitment of apolipoprotein E facilitates Herpes simplex virus 1 attachment, entry, and release

  1. Lifeng Liu
  2. Konrad Thorsteinsson
  3. Fouzia Bano
  4. Sanduni Wasana Jayaweera
  5. Damien Avinens
  6. Dario Conca
  7. Hudson Pace
  8. Hugo Lövheim
  9. Anders Olofsson
  10. Marta Bally

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Over two decades, epidemiological studies have revealed that interactions between human polymorphic apolipoprotein 4 (ApoE, isoform 4) and herpes simplex virus type 1 (HSV1) associate with higher risk of Alzheimer’s disease, a serious and increasing issue among elder populations worldwide. Nevertheless, little is known about the mechanisms behind ApoE-HSV1 interactions at molecular levels. Here, we investigate the effects of ApoE on the HSV1 infectious life cycle in in vitro cell experiments. Analysis of HSV1 growth curves shows that HSV1 production is promoted in presence of any of the three ApoE isoforms, with ApoE 3 or 4 demonstrating more proviral effects than ApoE 2. Quantification by qPCR reveals that the presence of ApoE 2, 3, or 4 leads to an increase of HSV1 extracellular release but unchanged levels of viral genome copies within cells or on the cell surface, indicating that virus replication, assembly, or transport to cell membrane are not affected. Further test of virus release directly demonstrates that HSV1 detachment from the cell surface is promoted by ApoE. Subsequent results reveal that ApoE is both present in purified HSV1 particles produced in ApoE-expressing cells after ultra-centrifugation and able to incorporate into HSV1 particles after purification, suggesting that harbouring ApoE may play a key role in the pro-viral effect of ApoE. Along these lines, we tested the infectious behaviour of ApoE-coated viruses and observed faster attachment kinetics and higher entry efficiencies of ApoE decorated HSV1. Our hypothesis that the association with ApoE leads to modified interactions of the virus with the cell membrane during entry and egress, was further validated in biophysical experiments. In such experiments, HSV1-membrane interaction kinetics and apparent affinity between HSV1 and native supported lipid bilayers (a plasma membrane mimic) were quantified using total internal reflection microscopy. HSV1 particles decorated with ApoE demonstrate both higher association ( k on ) and dissociation rate constants ( k off ), as well as less irreversible binding to the membrane, which is in line with the biological experiments. Overall, our results provide new insights into the roles of ApoE during HSV1 infections, which is worth to be considered when studying their involvement during Alzheimer’s disease development.

Article activity feed

  9. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    The authors do not wish to provide a response at this time

  10. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    There is mounting evidence pointing towards an association of HSV with ApoE and Alzheimer's disease. Although it has been shown that ApoE impacts HSV-1 spread in animal models in an isoform specific fashion, the molecular relationship between the virus and ApoE is unclear. The present study probes the role of ApoE on the viral life cycle, clearly an important aspect if one is to better understand how the virus may influence the disease. Using assays monitoring various steps of viral replication, the authors report that ApoE perturbs the interaction of the virus with the cell surface, both during the initial binding and following viral release. Furthermore, they show that ApoE 3 and 4 exert a proviral effect, with a smaller impact with ApoE 2.

    General Comments

    The comprehensive study addresses an important point that may help clarify the interaction between HSV-1 and Alzheimer's disease. It major strength is that it is systematic and elegantly performed. The paper convincingly shows that ApoE impact cell adhesion at the cell surface. In general, the data have properly been analyzed statistically. Some aspects are however enigmatic. First of all, how can ApoE be proviral if it prevents the initial binding of the virus to the cells? Second, less viral binding should mean less entry (as shown in figure 2) and subsequently fewer genome copies, but that is not what is reported in figure 3. This is not clearly stated in the discussion (lines 457-459 and again in lines 490-491). Finally, if ApoE is unstable as indicated on lines 495-497, how can it be active later on at 24 hpi and prevent viral release? How do the authors reconcile these observations? Of interest, ApoE 3 has a slightly greater impact on viral growth than other isoforms (fig 1), which is not quite fitting the model that ApoE4 is the main culprit for Alzheimer's disease. Could the authors comment? Where are the ApoE proteins normally expressed in cells? At the cell surface or intracellularly? This may provide a hint as to where the virus picks it up when incorporating it. Immunofluorescence would be a great addition (e.g., Huh-7 cell line). Similarly, does the virus impact the expression level of ApoE? One could resolve the dilemma that ApoE blocks the initial binding of the virus but stimulates viral release at later time points if the virus induces ApoE at those late times. Furthermore, if the Vero or SH-SY5Y cells don't normally express ApoE, then it should not be important for the virus in that context. How about keratinocytes, the normal host of the virus? These considerations should be addressed in the manuscript by Western blotting at different time points.

    Significance

    This study adds an interesting twist and advances the infectious etiology model of Alzheimer's disease and should appeal to a broad authorship ranging from the neurobiology to virology.

  11. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    In the manuscript "Recruitment of apolipoprotein E facilitates Herpes simplex virus 1 release", by Liu et al, the authors investigate the effect of ApoE protein on HSV-1 replication. Treatment of infected cultures with ApoE proteins appeared to increase the production of infectious titer. The authors perform several experiments to determine which steps of the virus replication cycle are affected. ApoE proteins reduce virus attachment, but subsequent reductions in entry are explained by the reduction in attachment, suggesting that the efficiency of entry is not affected. Viral DNA replication and cell-surface virus amounts appear to be unaffected, but the release of virus titer to the supernatant is increased. These results suggest that ApoE effects virus replication in two ways: reduces attachment if inoculum, but subsequently increases release of progeny. To determine whether this is the case, the authors then measure the release of attached virus particles from native membranes in the presence or absence of ApoE4, and derived from cells inducibly expressing ApoE4.

    The manuscript is generally well written and the experiments generally appear to be performed well. However, the importance and impact of this manuscript are limited by two major weaknesses:

    1. It seems that effects are only seen with high concentrations of ApoE. How does this concentration compare to what would be found in blood plasma/tissues/secreted by Huh-7 cells? Thus, these results may not be biologically relevant. It is difficult to determine what concentrations of ApoE are used in some cases, e.g. Fig 6. Please provide this information in the figure or figure caption.
    2. While there are some interesting results here, this manuscript does not get to the point of establishing mechanism. In the discussion, it is speculated that ApoE functions via GAGs/HSPGs, which are known to affect HSV-1 attachment/release. It would make the manuscript much stronger to include experiments adding soluble heparin or treating cells with heparinase, or producing gC-null virus particles, to see if this abolishes the attachment/release effects of ApoE.

    Minor points:

    Fig. 3B: It is difficult to compare virus genomes by qPCR to virus titer of supernatants. If ApoE is promoting release of cell surface virus, why does an increase in titer in the supernatants not show a corresponding decrease in cell surface virus?

    Fig. 6B: "Spdi I" term is not used in the results section or figure legend. I do not know what "Spdi I" means.

    Fig. 6E, Fig 7: Normalized values can be misleading. Please provide raw values. Please show dissociation curves, as in Fig. 6D, for Fig.7.

    It would be nice to perform the same analysis of supernatant vs. cell surface vs. intracellular virus as in Fig. 3B, and the release on ice measures as in Fig 4, using the inducible expression HEK cell line.

    Discussion: Degradation of ApoE (line 495-500). The degradation of ApoE in these infection experiments could be measured by e.g. western blot of cell supernatants. This suggestion is a bit troubling: If the ApoE is degraded during the first replication cycle, how is it able to have an ongoing effect? How can ApoE simultaneously be present to promote release of progeny, while being degraded so as not to prevent attachment to subsequent cells.

    I do not see that "GMK AH-1" cells are available from ATCC, as stated in the methods. Is this a synonym for Vero cells?

    Although I understand what is meant, "dissolvent" is not a common term. "diluent" or "vehicle" is more common.

    Significance

    General assessment: The manuscript is generally well written and the experiments generally appear to be performed well. However, the significance of this manuscript are limited by two major weaknesses:

    1. It seems that effects are only seen with high concentrations of ApoE. How does this concentration compare to what would be found in blood plasma/tissues/secreted by Huh-7 cells? Thus, these results may not be biologically relevant. It is difficult to determine what concentrations of ApoE are used in some cases, e.g. Fig 6. Please provide this information in the figure or figure caption.
    2. While there are some interesting results here, this manuscript does not get to the point of establishing mechanism. In the discussion, it is speculated that ApoE functions via GAGs/HSPGs, which are known to affect HSV-1 attachment/release. It would make the manuscript much stronger to include experiments adding soluble heparin or treating cells with heparinase, or producing gC-null virus particles, to see if this abolishes the attachment/release effects of ApoE.
  12. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    This manuscript attempts to identify the molecular basis for the reported interactions between apolipoprotein E (ApoE) and herpes simplex virus 1 (HSV1), known to be a significant marker for Alzheimer's disease. The authors employ a combination of cellular and in vitro assays designed to assess the effect of ApoE on different stages of the HSV1 life cycle. These experiments reveal an effect of ApoE on virus binding to and detachment from cell membranes, but not in other aspects of the viral life cycle. Further, the isoform ApoE4 was found to be the most effective in exerting these changes, possibly due to competitive binding with cell surface receptors that can associate with both ApoE and HSV1. Prior studies were referenced appropriately. For the most part, sufficient details about the data acquisition and analysis workflow were included in the study, although a couple of exceptions have been noted. More information about number of data sets for each panel and statistical analysis need to be included in the manuscript (figure legends and a separate section in Materials & Methods). Some of the key conclusions from the data require additional context and information to justify their interpretation in the present form. The additional experiments suggested are reasonable in terms of time and resources and are critical for strengthening the key conclusions in the manuscript. These have been noted below. Comments on evidence, reproducibility and clarity

    Major Comments:

    1. The authors mention in the Discussion section that they have ruled out interaction of ApoE with HSV1 glycoproteins B, C, D, and E based on immunoprecipitation data that is not included in the manuscript. In view of this, how do they justify using HSV1 gC as a marker for checking for association of HSV1 with ApoE (Fig 5)? Further, the authors should consider including said immunoprecipitation data in the manuscript, since those would be of immense value in further studies looking for other interaction partners of ApoE (as the authors have stated in the Discussion section).
    2. How strong are the interactions between ApoE and HSV-1? In other words, what fraction of the available ApoE could be expected to associate stably with HSV1? Can ApoE or ApoE-associated complexes act as a trap for the virus and therefore represent latent virus pools in cells? How were possible contributions to HSV1 detachment from other cellular factors associated with ApoE ruled out?
    3. The Discussion section of the manuscript clearly places the detected interaction of ApoE with HSV1 in the context of previous literature on related facets of the crosstalk between ApoE and viral infections. The authors may also consider including a paragraph on the more physicochemical attributes of ApoE interaction with HSV1, which would make the Discussion section more well-rounded and provide some background for understanding the biophysical experiments reported here (Fig. 6, 7). For example, how do the dissociation rates they report in Fig. 6 compare to those reported earlier for viruses on SLBs or cellular membranes? Could these dissociation rates be readily converted to (at least semi-quantitative) estimates of the thermodynamics of ApoE binding to HSV1 or would other factors need to be explicitly considered for such analytical exercises? Would it be difficult to measure binding affinities using HSV1 and purified ApoE by complementary approaches such as calorimetry or surface plasmon resonance? On a related note, what kind of ApoE concentration could HSV1 encounter in a cellular milieu and would it be in the range (5 μM) at which they report significant effects of ApoE on HSV1? What could be expected to happen at even higher concentrations of ApoE (reported for other cellular pathologies)?
      Why is the isoform dependence of ApoE effects observed predominantly in case of HSV1? Could this be related to the more complex fusion machinery available to this virus?
    4. It is strongly recommended that details about ApoE purification and characterization are included in the manuscript, along with appropriate references. It is also not clear how a 4h period was deemed to be sufficient for incubation with ApoE (Fig. 2).
    5. The data in Fig 4 is not entirely sufficient to support claims of ApoE enrichment in virus particles released into the supernatant. The authors may consider including additional experiments to check for (and quantify, if possible) ApoE levels in these virus fractions (since these conditions are drastically different than that used for reporting co-sedimentation of ApoE and HSV-gC in Fig. 6B).
    6. Fig 5: It is not clear why the authors tested only for gC (especially when they note that co-immunoprecipitation experiments have ruled out gC as a possible interaction partner for ApoE; also see comment 1). Is the shift in the ApoE band to higher kD values from fraction 1 to 6 significant? What do the error bars in Fig 5B represent, if data was generated from two independent replicates? How do you reconcile the very high viral titer of fraction 3 (Fig 5C) with the moderate level of gC_HSV-1 in the same fraction (Fig 5B)? Does this indicate heterogeneity in gC content across seemingly equivalent viral titers (fractions 1-3, based on Fig 5C) ?
    7. Several inconsistencies were noted in figures and figure legends that could affect a clear understanding of the data by readers. For all figures, please indicate clearly if no notation for statistical tests denote an absence of significance (ns) or that significance was not tested (such as for Fig. 2B, C). Please include sufficient information regarding number of independent replicates for each panel (i.e., what do error bars represent). For t-tests, please indicate clearly the reference data sets used for testing statistical significance and define symbols used for different p-values only for that specific figure. For example, a p-value corresponding to * is defined in the legend to Fig. 2, although that p-value is not indicated in the figure.

    Minor Comments:

    1. P. 4: abbreviation HSPG is not defined
    2. Inconsistent figure formatting noted in terms of non-uniform axis labels, color coding, inclusion of error bars, clear label of all lanes in blots. These should be reviewed and modified as appropriate.
    3. Fig 1A: The authors may consider reverting the order of ApoE concentration (ascending, instead of descending) in X-axis label to make it more intuitive.
    4. Line 167: please specify that data in Fig S1b refers only to SH-SY5Y cells.
    5. In Fig 2A, value for dissolvent should be set to 100%, since rest of the data are normalized with respect to that.
    6. Line 279 refers to Fig. 3C (not present in the manuscript).
    7. Ladders not clearly visible in some blots, such as that for 50 kD in Fig S1 and Fig 5A (HSV1 infected panel).
    8. Please indicate clearly in the Materials & Methods if ApoE induction (Fig. 7) is performed in HEK cells or HEK-293T cells.

    Significance

    This study sheds light on the molecular basis of the interaction between HSV1 and ApoE and represents a conceptual advance in the field. Since such interactions have been reported to be a marker for patients at high risk of developing Alzheimer's disease, these findings would be important in designing future clinical studies on prognostic and diagnostic advances in neurodegenerative diseases. As such, this manuscript would be of interest to a broad spectrum of scientists and clinicians including virologists, biochemists, and biophysicists.

  13. Version published to 10.1101/2023.02.10.526562v2 on bioRxiv
  14. Version published to 10.1101/2023.02.10.526562v1 on bioRxiv