All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The authors improved the manuscript, including additional material, answering to the referees, and improving and clarifying the research presented. Therefore this work is suitable for publication in this form.
Further to your formal Appeal on my earlier decision, I am willing to reconsider my decision. Please provide a revised version of the manuscript including all the issues emphasized by the referees and clarifiyng the most controversial points.
Appeal text from authors (7/6/2015):
I am writing you regarding the Decision on my Manuscript "Pore size is a critical parameter for
obtaining sustained protein release from electrochemically synthesized mesoporous silicon
microparticles" (#2015:05:5106:0:0:REVIEW). I understand that PeerJ politics regarding
manuscript publication are based on methodological and scientific soundness of the data
presented. As you noted in your comments, the decision for the rejection was heavily based on the
comments of Reviewer 3. While I deeply respect academic decisions, I think the comments of
Reviewer 3 that question the methodological soundness of the work are scientifically flawed.
Because of that, I would kindly request you to analyze my responses to Reviewer 3 below, and
consider the re-evaluation of this work for publication in PeerJ.
Thank you very much in advance. Looking forward to hearing from you.
Yours sincerely,
Marcos Garcia-
Responses to REVIEWER 3 (Reviewer in RED, MGF in black):
Reviewer #3 (R3): Pastor and coworkers present the controlled preparation of mesoporous silicon
based microparticles and their use for controlled release of protein.
The synthesis and characterization of material has been previously described by the same authors
and it has been here improved in order to tail the pore-size of particles under controlled conditions.
Such a material has been then used as support for protein adsorption and controlled delivery,
claiming its possible use as medical device. The theoretical design and practical development of
this experimental part present a lot of flaws and scientific issues that finally make the entire
research work unable to be published in PeerJ.
Main issues:
• Why did the authors use aCT in its soluble form as model protein? Being a proteolytic enzyme
and working in absence of an inhibitor as benzamidine, aCT can undergo to autolysis generating a
complex mixture of oligo and polypeptide. This behavior generates a serious problem when the
quantification of protein release must be carried out by BCA assay. In fact, the BCA chemistry
expects the chelation of copper atoms with protein in an alkaline environment forming a blue
colored complex. In this reaction, known as the biuret reaction, peptides containing three or more
amino acid residues are involved. The intensity of the color produced is proportional to the number
of peptide bonds participating in the reaction. Hence, due to the autolysis of aCT and carrying out
experiment for long times (14 days at 37C), the number of peptides available for biuret reaction will
increase in a time dependent manner generating a false absorbance increase. Enzymatic activity
or other protein detection assays together with the systematic use of SDS-PAGE should have been
mandatory. Moreover, in these long-time experiments, a control is always required in order to
exclude the autolysis.
MGF: I understand that this is the major methodological concern of the referee concerning this
manuscript. There are theoretical considerations and experimental evidence that evidence there
such methodological problem does not exist as I will describe in the following paragraphs.
First, note that the word “chymotripsin” is not mentioned even one time in the whole manuscript,
but rather the word “chymotripsinogen (i.e. the acronym aCT)”. This is because we have not
worked with the enzyme but rather with the zymogen (pro-enzyme). This zymogen is activated
by endopeptidaases in the brush border, and thus it does not transform to the active form under the
in vitro release conditions tested here.
On a second note, as the reviewer suggest, BCA reaction is based on peptide bonds, and contrary
to his reasoning there is no strict proportionality between number of peptide molecules and
observance, but rather between total protein mass and absorbance. In fact, if you see the Micro
BCA instruction (Life Technologies, Inc.) you will see that this assay is optimized for BSA in 0.5-20
ug/mL range, and they specifically mentioned “… and exhibits very low levels of protein-to-protein
variability”. Therefore, in the unlikely scenario were some peptide degradation would occur, the
impact over BCA observance would be minimum.
Finally, we have experimental evidence that this process as a whole does not occur. When
we first used aCT as a model protein for BMP-7 drug delivery devices (Reguera-Nuñez et al.
(2014) Biomaterials 35, 2859–2867), we performed this very simple test: we compared the readout
in the MicroBCA test between two aCT aliquots of the same sample, one stored for three weeks at
4ºC (fridge, control) and another at 37ºC. Both sample were in PBS +5%BSA, and the test was
performed in duplicate. As we expected, there was no significant differences in the BCA readout
between the test and control samples. These results obviously support the consistency of the
results presented in the present manuscript where we observed sometimes close to 100% of
protein recovery, but never more than that as it would be expected if the processes suggested by
the reviewer took place.
In any case, we plan to introduce some clarifying sentences in the manuscript to explain that aCT
is an inactive pro-enzyme loaded here as a predictive protein model for BMP-7 (see last question
for further data on this protein selection).
• R3: Why did the author not use the BCA protein mass detection for BMP7? In this case, it would
have been more useful than ELISA assay.
MGF: The protein doses required for therapeutic purposes for morphogens and similar proteins are
very small, in the order of micrograms, sometimes even nanograms. This is why concentrations
of these samples fall absolutely outside the range for BCA. In fact, ELISA has not only been
used by us many times for detecting this kind of molecules (see for instance Reguera-Nuñez et al.
(2014) Biomaterials 35, 2859–2867; González-Gómez et al. (2015) Oncotarget 6, 10950–10963; d’
Angelo et al. (2010) Mol. Pharmaceutics 7, 1724–1733), but it is the standard method to
measure growth factor and morphogen release for drug delivery devices in the literature.
On a separate note, a further advantage of using ELISA as the method is that it ensures that the
released protein is at least antigenically active; we know this does not ensure ultimately its
bioactivity, but in previous studies we have found good correlation between antigenically active
BMP-7 and bioactive BMP-7 (Reguera-Nuñez et al. (2014) Biomaterials 35, 2859–2867). Since
BMP-7 is a therapeutic protein, and the device could be used in the future for the design of a
pharmaceutical system, we believe that it is beneficial to have preliminary data supporting protein
bioactivity after in vitro release.
• R3: Why did the authors loaded the proteins in force conditions? I strongly disagree with the
claim in L 175-176. By this strategy, it is impossible to check the real protein loading because
after the solvent evaporation, a solid mass composed by the protein mass plus the microparticle
mass will be always obtained. Hence, it will be impossible to estimate the protein amount really
adsorbed inside the porous and discriminate it by the protein mass precipitated outside or not
adsorbed at all (as in the reported case). By this way, the protein loading will always be
quantitative and a burst effect will always be promoted (as depicted in Figure 2A). The loading
method described by the same authors in Pastor et al 2011 is undoubtedly more reliable.
MGF: Both the method used in our previous publication and this one share the same limitation:
they do not allow direct measurement of the associated protein. In this regard, please note that in
our previous publication (Pastor et al. (2011) Colloids and Surfaces B: Biointerfaces 88, 601–609)
we were only able to quantify the protein that was NOT associated to the platform. While we
believe that studies such as those published in the previous publication are more suited to
measure physicochemical parameters in the adsorption process, a method that achieves close to
quantitative loadings of a biopharmaceutical in non-equilibrium conditions are preferable for
pharmaceutical development. Because we were cautions of the fact that our system does not allow
the determination of the total amount of protein loading, we made sure that we extended protein
release times in all studies to a point where reasonably high protein recovery (>70%) had been
obtained (see Fig. 2A and 2B). Note that recovery of prototype C in 2A, even if not shown in the
graph, was 77.2% ± 4.2 (n=3) after two weeks (stated in line 197 of the manuscript).
R3: It is impossible that such a different proteins (especially considering the pI and size of both in
relation to the 6nm pore size) show almost the same release kinetic profile.
In sum, considering the above detailed comments together with many others issues encountered
along the manuscript, this referee considers this research work unable to be published in PeerJ.
MGF: aCT selection as a model protein was based on a proteomic screening of databases
for protein similarity. In fact, the difference in Mw between both proteins represents just over
10% (10.7%) of the whole weight of BMP-7, and the difference in isoloelectric point (pI) was just
1.4 units. Even more important, since both pI values were above 7.5, this ensured that net charge
was similar for both proteins during protein loading and in the release medium. I think any minimal
analysis of the pharmaceutical or drug delivery literature will show that most model proteins used in
other papers have far less structural similarity than aCT and BMP-7. In our case, we have already
used aCT as an indicative model for loading and release studies with BMP-7 (Reguera-Nuñez et
al. (2014) Biomaterials 35, 2859–2867). However, note that although we assume that aCT might a
predictive model for BMP-7, we do not expect nor claim to have achieved the same release kinetic
profile for both proteins.
Considering the comments and the critical points exposed by the referees (specially the third one) , the manuscript is in a very preliminary form, lacking of a set of important experiments, and therefore unfortunately this work cannot be accepted for publication in PeerJ.
The paper shows the control of the protein release from a support having different pore size. Results are clear, smaller pores permit a very slow release. However, they did not show if the enzyme is still active, or if the support may have any kind of interaction with these enzymes.
Detailed list of suggestions may be found below:
There some text that papers in all pages of the paper, very likely a mistake in the conversion of the files (the text is “ically active form”), making in some cases impossible to read the real sentence.
Abstract should contain main hypothesis, conclusions and results, introduction parts should be in the introduction section of the paper. Please, improve this section.
In introduction they need to mention the forces that keep the proteins on the support (what kind of adsorption is achieved, or the enzyme is just “trapped inside the support”). Moreover, using medical proteins also effect of the immobilization on stability and activity of the enzyme must be considered, a proper immobilization may improve enzyme stability, but a support able to produce many enzyme-support uncontrolled interactions may have a bad result. A protein is not like standard drug:
Scopus
EXPORT DATE:09 Jun 2015
Guzik, U., Hupert-Kocurek, K., Wojcieszynska, D.
Immobilization as a strategy for improving enzyme properties- Application to oxidoreductases
(2014) Molecules, 19 (7), pp. 8995-9018. Cited 13 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84904793107&partnerID=40&md5=6f3db285ab5deb45f9e666b2643077d7
DOI: 10.3390/molecules19078995
DOCUMENT TYPE: Review
SOURCE: Scopus
Min, K., Yoo, Y.J.
Recent progress in nanobiocatalysis for enzyme immobilization and its application
(2014) Biotechnology and Bioprocess Engineering, 19 (4), pp. 553-567. Cited 3 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84907236859&partnerID=40&md5=67c2833bb1c54b5396f8d2994555fc0f
DOI: 10.1007/s12257-014-0173-7
DOCUMENT TYPE: Review
SOURCE: Scopus
Stepankova, V., Bidmanova, S., Koudelakova, T., Prokop, Z., Chaloupkova, R., Damborsky, J.
Strategies for stabilization of enzymes in organic solvents
(2013) ACS Catalysis, 3 (12), pp. 2823-2836. Cited 34 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84890319993&partnerID=40&md5=80eed6151fa892b3b4782b68c9a175bd
DOI: 10.1021/cs400684x
DOCUMENT TYPE: Review
SOURCE: Scopus
Barbosa, O., Torres, R., Ortiz, C., Berenguer-Murcia, A., Rodrigues, R.C., Fernandez-Lafuente, R.
Heterofunctional supports in enzyme immobilization: From traditional immobilization protocols to opportunities in tuning enzyme properties
(2013) Biomacromolecules, 14 (8), pp. 2433-2462. Cited 53 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84881561590&partnerID=40&md5=4264b273db718c8ce2b6003e15234dd2
DOI: 10.1021/bm400762h
DOCUMENT TYPE: Review
SOURCE: Scopus
Rodrigues, R.C., Ortiz, C., Berenguer-Murcia, A., Torres, R., Fernández-Lafuente, R.
Modifying enzyme activity and selectivity by immobilization
(2013) Chemical Society Reviews, 42 (15), pp. 6290-6307. Cited 201 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84873693763&partnerID=40&md5=9a1e95235d50440b1eedf170b1d78087
DOI: 10.1039/c2cs35231a
DOCUMENT TYPE: Review
SOURCE: Scopus
Hwang, E.T., Gu, M.B.
Enzyme stabilization by nano/microsized hybrid materials
(2013) Engineering in Life Sciences, 13 (1), pp. 49-61. Cited 59 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84872415717&partnerID=40&md5=3672a520dfb5847a3ddf7f74dcbfe0ca
DOI: 10.1002/elsc.201100225
DOCUMENT TYPE: Review
SOURCE: Scopus
Garcia-Galan, C., Berenguer-Murcia, A., Fernandez-Lafuente, R., Rodrigues, R.C.
Potential of different enzyme immobilization strategies to improve enzyme performance
(2011) Advanced Synthesis and Catalysis, 353 (16), pp. 2885-2904. Cited 257 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-81555222964&partnerID=40&md5=3306501554b28e08f1941b4945f950ef
DOI: 10.1002/adsc.201100534
DOCUMENT TYPE: Review
SOURCE: Scopus
Rodrigues, R.C., Berenguer-Murcia, Á., Fernandez-Lafuente, R.
Coupling chemical modification and immobilization to improve the catalytic performance of enzymes
(2011) Advanced Synthesis and Catalysis, 353 (13), pp. 2216-2238. Cited 86 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-80051787511&partnerID=40&md5=c718fe5c44b6548f1fd4568930a6422e
DOI: 10.1002/adsc.201100163
DOCUMENT TYPE: Review
SOURCE: Scopus
Brady, D., Jordaan, J.
Advances in enzyme immobilisation
(2009) Biotechnology Letters, 31 (11), pp. 1639-1650. Cited 260 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-70350257632&partnerID=40&md5=072b0fd0b1c701bc9de521f385cb2f37
DOI: 10.1007/s10529-009-0076-4
DOCUMENT TYPE: Review
SOURCE: Scopus
Iyer, P.V., Ananthanarayan, L.
Enzyme stability and stabilization-Aqueous and non-aqueous environment
(2008) Process Biochemistry, 43 (10), pp. 1019-1032. Cited 322 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-50849139867&partnerID=40&md5=7bb955ce7b039d5fb6be3949f0ecd009
DOI: 10.1016/j.procbio.2008.06.004
DOCUMENT TYPE: Review
SOURCE: Scopus
Mateo, C., Palomo, J.M., Fernandez-Lorente, G., Guisan, J.M., Fernandez-Lafuente, R.
Improvement of enzyme activity, stability and selectivity via immobilization techniques
(2007) Enzyme and Microbial Technology, 40 (6), pp. 1451-1463. Cited 1005 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-33947602594&partnerID=40&md5=fe827afa9c1f8e7a9259e0a770543389
DOI: 10.1016/j.enzmictec.2007.01.018
DOCUMENT TYPE: Review
SOURCE: Scopus
The fact that the enzyme suffers two modifications if the enzyme modification strategy is use is true, but the first one (immobilization) may greatly improve the resistance to the coating with the hematin, not only the general enzyme behavior as stated above.
Rodrigues, R.C., Berenguer-Murcia, Á., Fernandez-Lafuente, R.
Coupling chemical modification and immobilization to improve the catalytic performance of enzymes
(2011) Advanced Synthesis and Catalysis, 353 (13), pp. 2216-2238. Cited 86 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-80051787511&partnerID=40&md5=c718fe5c44b6548f1fd4568930a6422e
DOI: 10.1002/adsc.201100163
DOCUMENT TYPE: Review
SOURCE: Scopus
They should also explain the likely effect of the pore size. A concavous pore should have a higher interaction with the protein when the pore diameter is smaller. Moreoiver, if protein aggregates are immobilized (e.g., using evaporation), these large aggregates may have problems to go outside. Authors should give some introduction to this:
Scopus
EXPORT DATE:09 Jun 2015
Bayne, L., Ulijn, R.V., Halling, P.J.
Effect of pore size on the performance of immobilised enzymes
(2013) Chemical Society Reviews, 42 (23), pp. 9000-9010. Cited 7 times.
http://www.scopus.com/inward/record.url?eid=2-s2.0-84887192289&partnerID=40&md5=cc9654ab65e132e02edf125fde499932
DOI: 10.1039/c3cs60270b
DOCUMENT TYPE: Review
SOURCE: Scopus
Li, Y., Gao, F., Wei, W., Qu, J.-B., Ma, G.-H., Zhou, W.-Q.
Pore size of macroporous polystyrene microspheres affects lipase immobilization
(2010) Journal of Molecular Catalysis B: Enzymatic, 66 (1-2), pp. 182-189. Cited 50 times.
DOI: 10.1016/j.molcatb.2010.05.007
Concentration of buffer??
Methods sometime advance results, please correct this.
3.2 Fist lines should be moved to introduction
Why they did not check the activity of the desorbed enzymes??
It should be interesting to check if the matrix is able to make any kind of interaction with the enzyme molecules (e.g., it is able to adsorbe some protein under adsorption/deseoprtion conditions??) The authors refer to previous paper havoing similar chemistry and different pore and particuale size where these inetractions exited. Can they give any clue of the nature?? E.f., adsorption is more rapid or slow if ionic strength increase, or if some PEG is added?
It seems fine.
It seems fine.
.
This work is very interesting, it is well presented and the conclusions are well supported by data.
The work is original including relevant findings. It is well presented and the conclusions are well supported by data.
Authors manage to tailor the size of the inner pores of the particles and control the protein release process. They have emphasized the different preparation parameters that are critical for obtaining such small and uniform pore sizes.
This work is very interesting, it is well presented and the conclusions are well supported by data. In my opinion it deserves to be published in this journal. Authors manage to tailor the size of the inner pores of the particles and control the protein release process. They have emphasized the different preparation parameters that are critical for obtaining such small and uniform pore sizes.
Authors should address a couple of questions:
- Biodegradability and toxicity: No significant tissue accumulation, pathology, or toxicity has been reported from the ingestion of those insoluble or very slightly soluble silicon compounds by the toxicity of intravenously administrated silica nanoparticles should be commented [for example: Yu Y, Li Y, Wang W, Jin M, Du Z, Li Y, et al. (2013) Acute Toxicity of Amorphous Silica Nanoparticles in Intravenously Exposed ICR Mice. PLoS ONE 8(4): e61346. doi:10.1371/journal.pone.0061346]
- Up-load capacity of a drug delivery system is important as well as the release control. Please comment on the loading.
See General comments
See General comments
See General comments
Pastor and coworkers present the controlled preparation of mesoporous silicon based microparticles and their use for controlled release of protein.
The synthesis and characterization of material has been previously described by the same authors and it has been here improved in order to tail the pore-size of particles under controlled conditions. Such a material has been then used as support for protein adsorption and controlled delivery, claiming its possible use as medical device. The theoretical design and practical development of this experimental part present a lot of flaws and scientific issues that finally make the entire research work unable to be published in PeerJ.
Main issues:
• Why did the authors use aCT in its soluble form as model protein? Being a proteolytic enzyme and working in absence of an inhibitor as benzamidine, aCT can undergo to autolysis generating a complex mixture of oligo and polypeptide. This behavior generates a serious problem when the quantification of protein release must be carried out by BCA assay. In fact, the BCA chemistry expects the chelation of copper atoms with protein in an alkaline environment forming a blue colored complex. In this reaction, known as the biuret reaction, peptides containing three or more amino acid residues are involved. The intensity of the color produced is proportional to the number of peptide bonds participating in the reaction. Hence, due to the autolysis of aCT and carrying out experiment for long times (14 days at 37C), the number of peptides available for biuret reaction will increase in a time dependent manner generating a false absorbance increase. Enzymatic activity or other protein detection assays together with the systematic use of SDS-PAGE should have been mandatory. Moreover, in these long-time experiments, a control is always required in order to exclude the autolysis.
• Why did the author not use the BCA protein mass detection for BMP7? In this case, it would have been more useful than ELISA assay.
• Why did the authors loaded the proteins in force conditions? I strongly disagree with the claim in L 175-176. By this strategy, it is impossible to check the real protein loading because after the solvent evaporation, a solid mass composed by the protein mass plus the microparticle mass will be always obtained. Hence, it will be impossible to estimate the protein amount really adsorbed inside the porous and discriminate it by the protein mass precipitated outside or not adsorbed at all (as in the reported case). By this way, the protein loading will always be quantitative and a burst effect will always be promoted (as depicted in Figure 2A). The loading method described by the same authors in Pastor et al 2011 is undoubtedly more reliable.
• It is impossible that such a different proteins (especially considering the pI and size of both in relation to the 6nm pore size) show almost the same release kinetic profile.
In sum, considering the above detailed comments together with many others issues encountered along the manuscript, this referee considers this research work unable to be published in PeerJ.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.