GRAFTING METHOD FOR ESTIMATING GENOTYPIC DIVERSITY IN ACROPORA CERVICORNIS

The overarching goal of this method is to establish who are the unique genets and who the clonal ones among a set of ramets of unknown genetic nature. That is to say, to unveil genotypic diversity. This in turn will ideally have the objective of ensuring or increasing the genetic diversity of restored reefs, thus promoting reef resilience. However, prior to designing the plan, it is important to establish what are the specific objectives of the study, or in other words, what will be the specific application of the genetic information obtained, since that will give a better idea of how appropriate it is to use this method compared to other more advanced methods of genotyping.

Some examples of its utilization are:

• To organize coral ramets in a nursery by genotype-specific structures or labels, for later evaluation of important traits such as bleaching tolerance, growth rates, disease resistance, etc.
• To organize coral ramets in a nursery by genotype-specific structures or labels, for later genetically organized outplant.
• To ensure genotypic uniqueness of new wild-collected ramets introduced in a nursery.

Available resources will determine the viability of performing this method. These include human, economic, and logistical resources for both in-water and on land operations. Primarily, you must have access to a diving center for the provision of tanks, boat, and dive gear; and have a team of restoration technicians or trained divers who can carry out the tests, as well as monitor and retrieve them. In addition, it is imperative to have the necessary government permits to handle coral and for the establishment of underwater structures if necessary. In already up and running restoration programs, this would incur no additional cost to ongoing restoration operations. However, if it is done occasionally without any established restoration program, all expenses associated with this activity must be considered (tanks, dive gear, boat, payment of personnel and training-associated costs, permit fees, etc.). Also, some tools and materials would be needed:

• Clippers

• Nylon

• Labels (for instance laminated waterproof paper, or plastic numbered labels)

• Zip ties

• Manpower and nursery structures materials in the case of setting up a new structure where to hang the tests from. For instance: o Rope structure: rope, anchors, and buoys.

  o Table, dome, or A-frame: rebar.

  o PVC tree: PVC pipes, driller, nylon, anchor, buoy.

For the analysis of the tests once the tissue growth period is concluded, a stereoscopic microscope is required (for instance: SMZ-143 N2LED, MOTIC).

Depending on the objective, it will be necessary to decide how many and which corals

are going to be studied. The more corals you want to evaluate, the greater the number of

combinations necessary between them, and therefore the necessary amount of tissue and

the time required will increase. Regarding tissue availability is important to take into

account that the fragments to be collected of each ramet will ideally correspond to the

apical parts of the ramet, so the more branched the coral, the easier it will be to collect

fragments and the more tests can be performed.

Combinations between tissue of different ramets will be done in pairs (i.e., individual 1 with individual 2) but will be arranged in a five-fragment bundle test to optimize number of tests needed for all contacts without compromising the ability to distinguish fragments once tissue grows. The five-fragment bundle is comprised of a larger fragment (10 ± 2 cm) and four smaller fragments (5 ± 1 cm) zip tied to both the apical and basal parts of the larger piece (Fig. 1). In this way, for each test there will be 6 contacts in pairs between the fragments (the two short ones from the apical part with the large one, the two short ones from the basal part with the long one, and the small ones from both parts between them).

Ramets in the study will be assigned a number from 1 to as many ramets there are, in order to facilitate combinations planning and arrangement. Letters can also be used instead but we will be using numbers in the present example. Each five-fragment test will also be identified by a number or a letter. Everyone will be combined with the rest of individuals (i.e., 1-2, 1-3, 1-4, and so on). Combinations between oneself could also be performed as acceptance controls. These combinations should be replicated across the tests in order to provide rigor to the outcomes. It is recommended to arrange a minimum of 4 or 5 replicates, depending on the tissue, time and logistics available. There is no specific order in which the combinations should be done, but they must be designed in such a way that the number of tests with the desired combinations is optimized, ideally without obtaining many redundant combinations. An example of combinations for 10 ramets is shown in the figure 2 below. Table representations will be important to follow a specific placement of the shorter fragments when arranging the test. More details in this regard are explained in section GRAFTING TESTS ASSEMBLY.

Depending on the number of corals for study and the planned combinations and replicates, the number of fragments needed should be calculated in advance to assess whether there is enough tissue available to collect for each individual without compromising the health of the colony. For the previous example of test for 10 individuals (Fig. 2), the required tissue is presented in table 1 below.

Table 1 . Number of fragments needed to arrange combinations between the ten ramets (Fig. 2).

A	B	C
Number of fragments needed
Ramet ID	Large	Short
1	5	14
2	5	13
3	4	17
4	1	23
5	5	14
6	5	16
7	5	16
8	5	14
9	4	18
10	4	17

Firstly, the structure where tests will be hanged from needs to be determined. It can be any kind of nursery structure (ropes, frames, coral trees...) although it is recommended that the tests hang with nylon off the structure rather than being fixed to it, as this way tissue will grow faster. An example of a rope structure with grafting tests is shown in figure 3. If there is no such structure already in the nursery, then the setup of this prior to test arrangement must be taken into account.

It is also important to think about who is going to set up the structure, assemble the tests, who is going to monitor their growth and clean them of macroalgae if necessary, who is going to collect and analyze them, etc; and when will all these activities take place. In addition, it is necessary to decide how long they are going to let them grow. In our experience and based in previous literature (García-Urueña et al. 1995), 10 weeks is enough to get a clear outcome between the fragments. Other logistical aspects associated with the previous step need to be considered (i.e., dive logistics for structures setup, transport of fragments to the lab or space where they will be analyzed once retrieved, etc.)

Combinations of each test must be planned before underwater arrangement and should be represented in tables that include test number and ramet numbers to be used (see section 4). It is recommended that this information is printed in a waterproof paper or written down on a diving slate for when tests are performed (Fig. 4).

To arrange the test bundles, these instructions should be followed:

The five fragments should all have their apex pointing towards the same direction (to the right in the example of this protocol).

This way, the apical fragments would be represented by the third (right) column of the table; and the basal ones would correspond to the first (left) column of the table).

Both apical and basal fragments will be zip tied to the larger fragment following a clockwise order. To do this, the person arranging the test should be looking towards the apex (for apical ones) and then towards the bases (for the basal ones) (Fig. 5)

The clockwise order will always start with the larger fragment (number 1 on the clock) and will be followed by the fragment represented on the first row of the table. The third fragment to be placed following the clockwise order will then be the one on the third row of the table.

The same principle stated in the previous step will be followed for both apical and basal fragments.

To optimize the time needed to carry out the tests, it is advisable to first fragment all necessary tissue (thus the importance of having the number of fragments needed as in table 1 and figure 4) and keep it separated by ramet ID. This can be done by using previously numbered plastic nests or bags (Fig. 6).

Fragments will then be joined together using thin zip ties and following the afore mentioned steps (section GRAFTING TESTS ASSEMBLY: Fragment placement and orientation). Additionally, a small label indicating the test number and desired additional information such as the date will be added to each bundle. Labels can be laminated pieces of waterproof paper, or any kind of manufactured label that can resist being underwater for a long period. This label can be tied to the test with nylon (preferably tied to one of the zip ties). Also, some nylon from the label could be left longer and used later on to hang it from the structure.

The previous step can be either done underwater or at the surface. This decision would be a trade-off between time availability and reducing coral stress.

Once arranged, the labeled tests can be tied with nylon to the structure, making sure they do not crash into each other to avoid stress.

The monitoring frequency can vary depending on the interest in evaluating the evolution of the acceptance or rejection response. If there is interest in this aspect, retrieving the tests to be examined under the microscope several times before the final retrieval could be a possibility. However, this will increase the stress in the corals and could affect their health and competition performance. It is recommended that tests are checked at least every two weeks, to ensure that they do not get covered in macroalgae, get entangled or other possible events that would have detrimental effects on the tests.

Before evaluating the contacts under the microscope, it is recommended to prepare a spreadsheet or table that contains a row for each combination and replicate, as well as the corresponding test number. Each test will be evaluated contact by contact (sometimes the flange can be removed to see if the contact between fragments keeps them together). It is also recommended to take photographs of each contact and write down whether acceptance or rejection is observed, as well as any additional information of interest.

Here we will present three types of responses that we have previously identified for Acropora cervicornis , one of them corresponding to an acceptance or tissue fusion response, and the other two corresponding to rejection or fission responses, one in a stronger way than the other.

Acceptance or fusion response

Acceptance responses involve the fusion of both skeletal components and soft tissues. When this happens, it is no longer possible to establish the start and end of both fragments, as they are smoothly sealed. Some examples under the microscope are shown in figure 7.

Rejection or fission response

Strong fission

Here we call strong fission what has previously been described as rejection in order to distinguish it from a weaker, less clear rejection described in the next section. The rejection response normally involves the formation of a suture line at the skeletal interface separating the tissues of the two fragments Neigel & Avise (1983). This area is devoid of connecting soft tissue. Rejection responses can also be characterized by the overgrowth of one of the fragments over the other, sometimes on a linear hierarchy pattern (Chadwick-Furman & Rinkevich, 1994; Frank & Rinkevich, 1994). Bleaching, anomalous growth, soft tissue death and incomplete development of the polyps have been also observed in the rejection response (Hildemann et al. 1977; Rinkevich, 2004). Some examples under the microscope are shown in figure 8.

Weak fission

In this type of response, what seemed like a certain fusion of the skeletal components was observed, not as clear and smooth as what is described as acceptance response. Also, in most of cases a white area was observed between both fragments, probably indicating lack of colonization by the symbionts (Hildemann et al. 1977). Some examples under the microscope are shown in figure 9.

Once outcomes data is collected, it is important to assess their consistency for each combination across replicates in case there is any inconsistency. That is to say, that the same combination (for example 4-5) sometimes fuses and sometimes fissions. If this happens, it is recommended to redo some more tests of the same combination. This may be due to human error during the test arrangement, using the wrong ramet. With this information, an estimation of the genetic relationships between the individuals in the study can now be established: those that are always fused will be taken as belonging to the same genotype, and those that are always rejected will belong to different genotypes. It is possible that those that reject each other weaker are more genetically related despite not being the same genotype.